[Lesional pulmonary oedema following laparoscopy for haemorrhage: aetiological assessment and possible gas embolism]. / Oedème pulmonaire lésionnel après laparoscopie pour hémopéritoine : bilan étiologique et possible embolie gazeuse.

A 44-year-old woman had to undergo repeat laparoscopy, four hours after laparoscopic sigmoidectomy, because of persistent bleeding from the drain. The bleeding caused perioperative shock and necessitated transfusional support. The following day she developed dyspnea, revealing pulmonary oedema. As her respiratory status continued to deteriorate and the bleeding persisted, she was transferred to the intensive care unit on day 2, after corrective laparotomy. The echocardiogram ruled out cardiogenic pulmonary oedema. No inhalation had occurred during the three anaesthetic procedures, and onset several hours after transfusion argued against lesional pulmonary oedema secondary to packed red cell transfusion. Given the context of repeat surgery by laparoscopy because of active bleeding, a diagnosis of lesional pulmonary oedema complicating CO(2) embolism was made. She received symptomatic treatment and her respiratory status gradually improved. Four months later her clinical status was normal. The different causes of postoperative pulmonary oedema and the pathophysiologic mechanisms of pulmonary lesions induced by gas embolism are discussed.

[Prospective assessment of information programme for blood transfusion in critical care patients]. / Evaluation prospective d'une procédure d'information des patients transfusés en réanimation.

OBJECTIVES: Assessment of oral and written information programme for blood transfusion in critical care patients and study of factors associated with the biological follow-up. STUDY DESIGN: Prospective study in one intensive care unit. PATIENTS AND METHODS: All blood recipients in 2000 were orally informed of transfusion process. In regard to the French haemovigilance regulation, written paper was given to the patient or passed on to the next hospital physician. Four months later, a questionnaire was proposed to the patients. RESULTS: In 2000, blood transfusion was performed for 102 patients and data were obtained for 59 patients (34 number of survivor patients, follow-up was impossible for eight and one refused the questionnaire). Quality assurance programme permit a global orally and written information of critical care patients. However, thirty-eight patients (64%) remembered oral information. Patients were younger (55 +/- 22 vs 68 +/- 13, OR = 1.04; p = 0.03). Biological follow-up was obtained for 22 patients (37%), particularly when written information was directly given to the patient (22/24 vs 2/35; OR = 335; p < 0.0001). CONCLUSION: Biological follow-up was obtained for only one third of blood recipients, because, in first, only 64% of critical care patients remembered oral information, particularly the younger patients. Perhaps, written information systematically given to the patient could permit a high biological follow-up. This study confirmed the lack of sensitivity of all hospital physician for haemovigilance.

Usefulness of quantitative cultures of BAL fluid for diagnosing nosocomial pneumonia in ventilated patients.

STUDY OBJECTIVE: To evaluate the role of quantitative cultures of BAL for diagnosing nosocomial pneumonia in mechanically ventilated patients. DESIGN: Cohort study. SETTING: Medical ICU, Hôpital Bichat, Paris, France, an academic tertiary care center. PATIENTS: A total of 141 episodes of suspected lung infection in 84 consecutive patients mechanically ventilated for 48 h or more. MEASUREMENTS AND RESULTS: Microbiologic findings obtained using BAL were compared with those obtained with protected specimen brush (PSB) samples and their operating characteristics were determined. The level of qualitative agreement between BAL and PSB specimen cultures was high, with 83% of the organisms isolated in PSB specimens being recovered simultaneously from BAL fluid. In addition, the results of quantitative BAL and PSB cultures were significantly correlated (rho = 0.46, p < 0.0001). Fifty-seven cases of pneumonia were diagnosed based on the following criteria: PSB sample yielding > or = 10(3) cfu/mL of at least one microorganism and/or > or = 5% of cells containing intracellular bacteria on direct examination of BAL. The operating characteristics of BAL fluid cultures were determined using different ways to report the results and over a range of values. The discriminative value of 10(4) cfu/mL was found to be an optimal threshold, with a sensitivity of 82% (95% confidence interval [CI], 76 to 88) and a specificity of 84.5% (95% CI, 79 to 90). CONCLUSIONS: These results indicate that BAL fluid cultures can offer a sensitive and specific means to diagnose pneumonia in ventilated patients and may provide relevant information about the causative pathogens.

[Pulmonary valve replacement for endocarditis. Apropos of 2 cases]. / Remplacement valvulaire pulmonaire pour endocardite. A propos de 2 cas.

The authors report two cases of pulmonary valve endocarditis which required emergency surgical treatment. A 74 year old patient with trivalvular endocarditis (pulmonary, aortic, mitral), due to Sptreptococcus D bovis, developed cardiogenic shock with acute pulmonary oedema and underwent double aortic and pulmonary valve replacement with Carpentier-Edwards prostheses and simple resection of a mitral valve vegetation. Another 36 year old drug addict developed isolated pulmonary valve endocarditis due to Staphylococcus aureus infection complicated by pulmonary regurgitation with right ventricular failure and by septic pulmonary embolism with persistent sepsis: he underwent pulmonary valve replacement with a Bravo 300 bioprosthesis. The postoperative course was uncomplicated in both cases, with interruption of the infection and normalisation of the haemodynamic status. The insidious and severe nature of pulmonary valve endocarditis is demonstrated by these two cases, confirming previous reports which have underlined the poor prognosis of this condition. Surgery has been shown to be effective and well tolerated and should be integrated early in the therapeutic strategy, the results being all the better when an aggressive attitude is taken.

Closed drainage using Redon catheters for local treatment of poststernotomy mediastinitis.

BACKGROUND: Continuous irrigation has been used worldwide for the treatment of acute poststernotomy mediastinitis. However, its high rate of failure led to the development of new methods, among them closed drainage with Redon catheters. METHODS: We evaluated the results obtained with Redon catheters in 70 patients, and compared them to those obtained in 38 patients treated with continuous irrigation. RESULTS: The two treatment groups were not different for age, type of cardiac operation, and initial severity of illness. Local failure of Redon catheter drainage occurred less frequently (20 of 38 versus 9 of 70 patients; p = 0.0001). This reduced failure rate was mainly attributable to a lower incidence of superinfections (10 of 38 versus 2 of 70 patients; p = 0.0002), but also to a lower incidence of primary failure (10 of 38 versus 7 of 70 patients; p = 0.026). Mortality was significantly decreased (15 of 38 versus 12 of 70 patients; p = 0.01). The other major advantage of this technique was the simplicity of its use. CONCLUSIONS: The technique using Redon catheters should be considered an effective and convenient treatment of acute poststernotomy mediastinitis.

Evaluation of bronchoscopic techniques for the diagnosis of nosocomial pneumonia.

To compare the usefulness of specimens obtained by bronchoalveolar lavage (BAL) and using a protected specimen brush (PSB) in the diagnosis of nosocomial pneumonia, both procedures were performed via fiberoptic bronchoscopy just after death in a series of 20 ventilated patients who had not developed pneumonia before the terminal phase of their disease and who had no recent changes in antimicrobial therapy. These results were compared with both histologic and microbiologic postmortem lung features in the same area. The total number of bacteria obtained by culture of lung segments and the latters' histologic grade were closely correlated (rho = 0.79, p < 0.0001). PSB and BAL quantitative culture results were strongly correlated with lung tissue values (rho = 0.67 and 0.75, respectively; p < 0.0001). Using discriminative values of > or = 10(3) and > or = 10(4) bacteria/ml to define positive PSB and BAL cultures, respectively, these techniques identified lung segments yielding > or = 10(4) bacteria/g tissue with sensitivities of 82 and 91% and specificities of 89 and 78%, respectively. Moreover, upon direct observation, the percentage of BAL cells containing intracellular bacteria was closely correlated with the total number of bacteria obtained from corresponding lung samples (p < 0.001). These findings indicate that bronchoscopic PSB and BAL samples very reliably identify both qualitatively and quantitatively microorganisms present in lung segments with bacterial pneumonia, even when the infection develops as a superinfection in a patient already receiving antimicrobial treatment for several days.

Role of quantitative cultures of endotracheal aspirates in the diagnosis of nosocomial pneumonia.

To assess the reliability of quantitative cultures of endotracheal aspirates (EA) to diagnose ventilator-associated pneumonia, fiberoptic bronchoscopy was used to study 57 episodes of suspected lung infection in 39 patients with no recent changes in antimicrobial chemotherapy. A total of 19 cases (33%) of pneumonia were diagnosed based on the following criteria: protected specimen brush (PBS) sampling yielding > or = 10(3) cfu/ml of at least one microorganism and/or > or = 5% of cells containing intracellular bacteria on direct examination of bronchoalveolar lavage (BAL). The operating characteristics of EA cultures were calculated over a range of cutoff values (from 10(3) to 10(7) cfu/ml), and the threshold of 10(6) cfu/ml appeared to be the most accurate, with a sensitivity of 68% and a specificity of 84%. Microorganisms cultured from EA samples correlated weakly with those obtained using PSB specimens (rho = 0.32), with only 49 microorganisms among 123 (40%) found in both samples. These latter results and the relatively low sensitivity of the technique indicate that EA quantitative cultures are of limited value for the diagnosis of pneumonia in ventilated patients when fiberoptic techniques are available.

Analysis of vancomycin entry into pulmonary lining fluid by bronchoalveolar lavage in critically ill patients.

Vancomycin penetration into the fluid lining the epithelial surface of the lower respiratory tract was studied by performing fiberoptic bronchoscopy with bronchoalveolar lavage on 14 critically ill, ventilated patients who had received the drug for at least 5 days. The apparent volume of epithelial lining fluid (ELF) recovered by bronchoalveolar lavage was determined by using urea as an endogenous marker. Vancomycin levels in ELF ranged from 0.4 to 8.1 micrograms/ml (mean, 4.5 micrograms/ml), while the mean simultaneous level of the drug in plasma was 24 micrograms/ml (range, 9 to 37.4 micrograms/ml). There was a significant relationship (r = 0.64, P < 0.02) between vancomycin levels in plasma and those in ELF, with a correlation whose slope (0.15) indicated that the blood-to-ELF ratio of drug penetration was 6:1. Using the albumin concentration in ELF as a marker of lung inflammation, we found that vancomycin penetration was higher in patients with ELF albumin values of > or = 3.4 mg/ml than in patients with normal values (< 3.4 mg/ml) (P < 0.02). These results suggest that the vancomycin distribution includes the ELF of the lower respiratory tract at a concentration that is dependent upon the levels in blood and the alveolar capillary membrane protein permeability. These concentrations were well above the MICs for most staphylococci and enterococci.