Attempt to establish a classification of patients suffering from coma and admitted to a hospital structure for short or medium term treatment: medical aspects and costs of hospital care.

We dispose of a database, constituted between 1987 and 1993, containing medical and cost information concerning 515 patients suffering from coma and admitted, after a period of resuscitation, to a French hospital establishment--Etablissement Hélio-Marin of Berck-sur-Mer (EHMB)--for short and medium term treatment, between 1974 and 1986. From this base, which contains demographic and clinical data (age, sex, condition upon admission, duration of consciousness disorders, Glasgow Outcome Scale (GOS) upon discharge) we devised a hierarchical classification analysis following a factorial analysis of multiple correspondences, on 2 sets: a sample of 515 patients (all causes of coma being merged) and a sample of 266 patients suffering from brain injuries. Four groups were determined for each typology. These groups were first described on the basis of the variables used for their construction, and later by considering other available variables: origin of coma, duration of stay at EHMB, future evolution of patients and cost of treatment (cost of specific care, average daily cost, total cost of hospitalization). Thus, typical clinical situations were identified in each classification, depending on age of patient, origin of coma and condition upon admission. These situations led to extremely different treatment costs (ratio from 1 to 5 in the general typology and 1 to 2.85 in the classification of brain injuries.

Prophylaxis, cost and effectiveness of therapy of infections caused by gram-positive organisms in neutropenic children.

Gram-positive infections are being reported with increasing frequency in children with haematological malignancies. Staphylococcus epidermidis, once considered a non-pathogenic skin contaminant, is emerging as a major cause of severe infection. However, in infants Gram-negative septicaemias are more frequent than in older children. A teicoplanin and ceftriaxone combination was assessed for use as empirical therapy of febrile episodes in neutropenic children with acute leukaemias. Of 47 patients, fever was of unknown origin in 21, and documented in 26 with 28 strains isolated; 19 Gram-positive (all sensitive to teicoplanin) and nine Gram-negative. Within 48 h, 41 patients became afebrile and the pathogen was cleared if initially present. Mean duration of treatment was 16 days. Febrile relapse occurred in 24 patients with eight documented superinfections. The need for prophylactic cover against Gram-positive organisms at the time of intravenous catheter insertion is questionable. We studied 71 patients who were randomly allocated to receive teicoplanin when the central line was inserted and if febrile, with added ceftriaxone and amikacin (arm A) or the tri-antibiotic regimen when fever occurred (arm B). In arm A a febrile episode occurred after ten days in 34/35 patients with only one Gram-positive organism isolated. In arm B a febrile episode occurred in all 36 patients after five days and ten Gram-positive strains were isolated. Those patients in arm A also received fluconazole. Amphotericin B was administered in cases of failure or relapses in both groups.(ABSTRACT TRUNCATED AT 250 WORDS)