Reduction of the severe sepsis or septic shock associated mortality by reinforcement of the recommendations bundle: a multicenter study.

INTRODUCTION: We determined whether the implementation of a bundle of 10 recommendations leads to the reduction of mortality in ICU patients with severe sepsis or septic shock. METHODS: All patients with severe sepsis or septic shock during two consecutive phases: a 6-month quality control period (observational) and secondly a 6-month intervention period based on the implementation of a bundle of 10 recommendations adapted from the Surviving Sepsis Campaign guidelines (initial bacteriological samples and initiating antibiotics, measurement of arterial lactate, volume expansion > or =20 ml/kg, targeted mean arterial pressure > or =65 mmHg and the assessments of central venous pressure and Scv(O2); glucose control, low doses of corticosteroids, a tidal volume < or =8 ml/kg in mechanically ventilated patients with ALI; adequate use of recombinant human activated protein C) were evaluated in 15 ICUs. The primary endpoint was the 28-day mortality rate and the secondary endpoint was the compliance with the recommendations of the care bundle. MEASUREMENT AND RESULTS: Four hundred and forty-five patients (230 and 215 in the observational and intervention periods, respectively) were included. In the two periods, the patients had similar characteristics. The 28-day mortality rate significantly decreased from 40% in the observational period to 27% in the intervention period (P=0.02). According to each recommendation, compliance with the care bundle was achieved in 9 to 100% of patients. CONCLUSION: The implementation of a care bundle adapted from the Surviving Sepsis Campaign guidelines decreases the 28-day mortality rate in patients with severe sepsis and/or septic shock.

[Visceral form of human anthrax imported from Africa]. / Forme viscérale de charbon humain importée d'Afrique.

Widespread vaccination has largely eliminated anthrax in Europe (the last case was reported in France in 1972) but the disease remains endemic in many developing countries. The usual cutaneous presentation (malignant pustules) is much more familiar than the various visceral manifestations including digestive tract, pulmonary or meningeal signs. We report a case of a 33-year-old immigrant living in France who was hospitalized for asthenia, dyspnoea, mucopurulant expectoration and moderate diarrhoea 3 days after a 3-month stay in Senegal and Gambia. The temperature was 39 degrees C at admission and blood pressure 110/70 mmHg. Crepitants were heard at the base of the right lung and the rest of the physical examination was normal. Blood was drawn for culture. Laboratory tests and the chest X-ray led to the diagnosis of pneumopathy and a treatment of amoxicillin and clavulanic acid was given with oxygenotherapy. The patient's temperature returned to normal but over the next 48 hours the dyspnoea worsened together with the black diarrhoea. The abdomen was painful. There were no skin lesions. The chest X-ray revealed an extension of the bilateral pulmonary images and bilateral pleural effusion. Laboratory tests revealed thrombopenia (platelet count 38,000/mm3) hyperleukocytosis (WBC 48,000/mm3) and haemolysis (Hb 4 milligrams). The diagnosis was made on the basis of the initial blood cultures which were positive for Bacillus anthracis. All other samples were negative, including HIV serology. Despite adapted antibiotic therapy (penicillin G, 8MU/day, was initiated on day 2), multiple organ failure occurred with septic shock and pulmonary oedema. The patient died in the intensive care unit on day 7. Fatal outcome due to anthrax is described in 25% of the visceral forms but reaches 100% in cases of septicaemia. The haemolysis observed in this case is not mentioned in the classical descriptions of anthrax. When treating septic syndromes in patients who have returned from endemic zones, clinicians should entertain the diagnosis of anthrax since the risk of fatal outcome is increased greatly in case of delayed diagnosis.