Potential lung cancer screening outcomes using different age and smoking thresholds in the ANRS-CO4 French Hospital Database on HIV cohort.

OBJECTIVES: In most lung screening programmes, only subjects ≥ 55 years old and smoking ≥ 30 pack-years are eligible to undergo chest low-dose computed tomography. Whether the same criteria should apply to people living with HIV (PLHIV) is uncertain, given the increased lung cancer risks associated with immunodeficiency and high rates of smoking. We assessed different outcomes obtained from simulating one round of lung cancer screening in PLHIV using different age and smoking thresholds for eligibility. METHODS: Data from the French Agence Nationale de Recherche sur le SIDA et les Hépatites Virales (ANRS)-CO4 French Hospital Database on HIV (FHDH) cohort of PLHIV and a national representative survey of PLHIV in care in 2011 (the ANRS-VESPA2 [enquête sur les personnes atteintes] study) were used to estimate the maximum proportion of incident lung cancers occurring between 2012 and 2016 that would have potentially been detected by screening in 2011. Secondary outcomes were numbers of eligible subjects in the cohort and numbers of subjects needed to screen (NNS) to detect one lung cancer. RESULTS: Among 77819 PLHIV in 2011 (median age 46 years; 66% men), 285 subjects subsequently developed lung cancer. Adoption of the US Preventive Services Task Force (USPSTF) recommendations (55-80 years; ≥ 30 pack-years) would have detected 31% of lung cancers at most. Lowering the minimum age to 50 and 45 years would have detected 49% and 60% of cancers, respectively, but would have greatly increased the number of eligible subjects and the NNS to detect one case of lung cancer. CONCLUSIONS: Use of the USPSTF criteria would have detected only a minority of lung cancers in a large French cohort of PLHIV in 2011. Screening PLHIV at younger ages (45 or 50 years) and/or the use of lower smoking thresholds (20 pack-years) may be beneficial, despite the consequently higher numbers of eligible subjects and NNS to detect one case of lung cancer, and should be evaluated in future studies.

Soluble biomarkers of immune activation and inflammation in HIV infection: impact of 2 years of effective first-line combination antiretroviral therapy.

OBJECTIVES: The aim of the study was to assess the impact of rapid and sustained viral control produced by combination antiretroviral therapy (cART) on HIV-associated immune activation and inflammation. METHODS: In this longitudinal observational study, we examined changes in interleukin-6 (IL-6), interferon-γ-inducible protein-10 (IP-10), monokine induced by interferon-γ (MIG) and soluble CD14 (sCD14) levels during 2 years of effective first-line cART. Biomarker levels before and after cART were compared with those observed in healthy subjects, using the Wilcoxon signed rank test. Elevated biomarker levels were defined with respect to values for healthy subject (mean + 2 standard deviations). Factors associated with persistently elevated biomarker levels after 2 years of cART were identified by logistic regression. RESULTS: We included in the study 139 patients with a median HIV-1 RNA level of 4.8 log10 HIV-1 RNA copies/mL and a median CD4 cell count of 294 cells/µL at cART initiation [day 0 (D0)]. At D0, all biomarker levels were higher than in healthy subjects (P < 0.05). After 2 years of cART, IL-6, IP-10 and MIG levels fell significantly, by a median of 0.54, 420 and 1107 pg/mL, respectively (all P < 0.001), and were no longer elevated in > 75% of patients. In contrast, sCD14 levels did not change significantly (0.18 × 10(6) pg/mL; P = 0.102) and remained elevated. Older age was associated with elevated levels of IP-10 [odds ratio (OR) 1.60 per 10 years older; P = 0.047] and MIG (OR 1.92 per 10 years older; P = 0.007) after 2 years of cART. CONCLUSIONS: The rapid and sustained viral suppression produced by first-line cART reduced IL-6, IP-10 and MIG to normal levels, while sCD14, a marker of monocyte activation, remained elevated. High levels of IP-10 and MIG tended to persist in older patients.

Viral efficacy maintained and safety parameters improved with a reduced dose of stavudine: a pilot study.

OBJECTIVES: Stavudine (d4T) is a potent but potentially toxic nucleoside reverse transcriptase inhibitor that is still widely used in developing countries. This study's aim was to determine the efficacy and safety profile of lower-dose d4T. METHODS: Multi-centre, open-label, single-arm, pilot, 48-week study in French patients weighing >60 kg with viral load <400 HIV-1 RNA copies/mL who were receiving d4T 40 mg twice daily and then switched to 30 mg twice daily. The primary endpoint was the proportion with plasma viral load <400 copies/mL at week 24. Secondary endpoints included the proportion with <50 copies/mL at weeks 24 and 48, changes in mitochondrial DNA, CD4 cell count and pharmacokinetics, and clinical and laboratory safety. RESULTS: Fifty-seven patients enrolled. Baseline CD4 count was 584 cells/microL; viral loads were <400 copies/mL and <50 copies/mL in 100% and 89%, respectively. Prior antiretroviral drug exposure was 6.9 years, d4T exposure was 6.3 years. Fifty-six out of 57 (98%) patients had viral load <400 copies/mL and 51 (89%) had viral load <50 copies/mL at week 24. Median CD4 count increased by 63 cells/microL at week 48 (P=0.006). At 48 weeks, total cholesterol decreased by 0.24 mmol (P=0.02), high-density lipoprotein cholesterol by 0.15 mmol (P=0.0001) and alanine aminotransferase by 5.74 mg/dL (P=0.01). Paired baseline DNA and week 24 RNA mutations were unchanged. Mitochondrial DNA (copies/cell) content increased from 672+/-254 to 682+/-269. d4T area under the plasma concentration time curve (AUC) decreased by 31% (P=0.003) and C(max) by 44% (P=0.004). Clinical and laboratory parameters improved or were unchanged. CONCLUSIONS: Reduced-dose d4T is effective with improved safety parameters.

Influence of control selection in nested case-control studies: the example of exposure to antiretroviral treatment (ART) and the risk of myocardial infarction (MI) in the French Hospital Database on HIV (FHDH-ANRS CO4).

PURPOSE: Soon after availability of protease inhibitors (PIs), a duration-related effect relationship between PI and myocardial infarction (MI) was shown. New antiretroviral treatments (ARTs) have allowed more individualized regimens. To study their influence established risk factors of MI and additional therapeutic options such as lipid-lowering drugs will have to be taken into account. A nested case-control is an interesting alternative raising the choice of controls. With the previous full cohort analysis as reference, we investigated the influence of control selection in nested case-control studies sampled in this cohort by testing nine sampling scenarios. METHODS: During the period 1996-1999, 49 MI occurred among male patients exposed to PI and followed-up in the French Hospital Database on HIV (FHDH-ANRS CO4). For each case, controls were selected using incidence-density sampling. The influence of additional matching criteria was tested. Random sampling and analysis was repeated 100 times with varying control-case ratios. RESULTS: When controls were randomly selected among patients of the same age who were free of MI at the date MI was diagnosed in the case, we observed a duration-related effect relationship between PI and MI in agreement with the results of the full cohort analysis. The use of four controls per case was sufficient. Estimates obtained with simple sampling were more precise than those obtained when controls were also matched for year of enrollment, initial CD4 cell count and HIV transmission group. CONCLUSION: To study ARTs as MI risk factors, nested case-control using incidence-density sampling without additional matching is one appropriate option.

[Evaluation of compliance with antiretroviral treatment in a cohort of 200 patients in Djibouti, 2005]. / Evaluation de l'observance au traitement antirétroviral au sein d'une cohorte de 200 patients à Djibouti (2005).

We determined the rate of compliance with antiretroviral therapy and investigated the factors that influence it among 86 HIV patients. Compliance ratio (number of tablets taken/number prescribed) was assessed by tablet count. The mean ratio of compliance was 92%. By tablet count, 77% of the patients were compliant (compliance ratio > or =90%). Non-compliance was significantly associated with side-effects, degree of confidentiality of the care centre and travelling. Compliance correlated significantly with viral load. In multivariate analysis, community support and level of education protected against non-compliance. Patients having already missed a dose and those dissatisfied with confidentiality had a 4 times greater risk of non-compliance.

Men from Sub-Saharan Africa Living in Worker Hostels in France: A Hidden Population with Poor Access to HIV Testing.

Delayed presentation to care among HIV-infected individuals continued to be frequent in France. Migrants are at high risk for late presentation. This cross-sectional study investigated barriers to HIV testing in the specific population of men from sub-Saharan Africa living in four migrant worker hostels in Paris, France. Factors associated with never having been tested for HIV were examined using logistic regression. In all, 550 men participated, coming mainly from Mali and Senegal, with 31 % having lived in France for less than 5 years, and 25 % without any health insurance. Only 37 % have ever been tested for HIV. Not having health insurance was the main risk factor for never-testing [adjusted odds ratio (aOR) 2.4; 95 % confidence interval (CI) 1.4-4.0]. Despite free and anonymous HIV testing available at dedicated public screening centers, 63 % of men living in migrant worker hostels had never been tested for HIV.

Survival benefit of repeat liver resections for recurrent colorectal metastases: 143 cases. Association Francaise de Chirurgie.

PURPOSE: Resection is the only chance for cure in patients with colorectal liver metastases (LM). Five-year survival rates are close to 25%. Unfortunately, recurrences occur in most patients. Some recurrent LM are technically resectable. The aim of this study was to determine the risks and benefits of repeat resections for recurrent LM. PATIENTS AND METHODS: Data from 130 patients who received 143 repeat liver resections for recurrent LM were collected. In 116 patients, only the liver was involved, while 14 had both liver and extrahepatic recurrences. RESULTS: In the first group, the operative mortality and morbidity rates were 0.9% and 24.7%, respectively. Two- and 3-year survival rates were 57% and 33%, respectively. Recurrences were observed in 66% of patients. Twelve patients underwent a third hepatectomy for recurrence. The mortality rate was nil, and the mean survival time was 12.5 months. In the group with liver and extrahepatic metastases, the operative mortality and morbidity rates were 0% and 25%. The mean survival time was 16 months. Eleven patients died and 13 had recurrences during the follow-up period. CONCLUSION: Some hepatic recurrences after surgical excision of colorectal metastases can be resected with a low operative risk and with a long-term survival rate similar to that obtained after first resections. This emphasizes the need for a careful follow-up after hepatectomy for colorectal metastases to detect resectable recurrences.

[Immunofixation compared to isoelectric focusing in the detection of oligoclonal bands in cerebrospinal fluid of multiple sclerosis patients]. / Comparaison de l'immunofixation après électrophorèse sur gel d'agarose avec la focalisation isoélectrique dans la détection des bandes oligoclonales d'IgG du liquide céphalo-rachidien de patients atteints de sclérose en plaques.

INTRODUCTION: Intrathecal immunoglobulins (Ig) synthesis, reflected by oligoclonal bands (OCBs) in cerebrospinal fluid (CSF) is observed in up to 90 percent of patients with clinically definite Multiple Sclerosis (MS). The gold standard laboratory test to establish the presence of OCBs in CSF of MS patients is isoelectric focusing (IEF). However, a quicker and less expensive method has been developed: immunofixation (IF). METHODS: The aim of this study was to compare these two methods carried out 74 CSF/serum pairs of MS, 103 CSF/serum pairs of subject controls and to determine their sensitivity and specificity. RESULTS: The agreement between results from IEF and IF was excellent (Kappa = 0.84). IEF sensitivity (78 percent) was not significantly different from that of IF (74 percent) (p = 0.3). Similarly, the specificity of IEF (93 percent) was not significantly different from that of IF (95 percent) (p = 0.2). CONCLUSION: IF is a semi automated method which is easier to perform than IEF and which appears to be as efficient as IEF.

Statistical analysis of splenic colony histology: an attempt to confirm that external factors could affect the channelling of CFUS differentiation.

The aim of this work was to analyze quantitatively experimental data, which suggest that CFUS determination can be manipulated by external fibers. This hypothesis was based on the comparison of the ratio of erythroid to granulocytic spleen colonies generated by normal bone marrow exposed to factors released by either untreated or arabinoside cytosine treated mouse bone marrow. To investigate mechanisms able to produce this modification of spleen colony histology, statistical analysis was performed by testing three different schemes of evolution of the number of splenic colonies according to this histology from normal to treated groups. The total number of colonies per spleen is similar in both groups, but a significant increase of the number of erythroid colonies per spleen and a significant decrease of the number of granulocytic colonies are observed in the treated group as compared to control. Bias due to the used experimental technique are investigated but could not explain the observed differences. A unique mechanism acting on only one committed stem cell population does not fit the experimental data. Although other possible mechanisms are suggested, the experimental observations can be interpreted as the consequence of a shift of CFUS differentiation toward the erythroid pathway at the expense of at least the granulocytic lineage due to some humoral factors, secreted by treated mouse bone marrow.

Frequency and risk factors of infectious complications in neutropenic patients infected with HIV.

OBJECTIVE: To determine causes, incidence and factors associated with infections in neutropenic [polymorphonuclear neutrophil (PMN), 1000 x 10(6)/l] HIV-infected patients. DESIGN: Prospective study. SETTING: Infectious disease service of a 1000-bed university teaching hospital in Paris, France. PATIENTS: HIV-infected patients with a PMN count of < 1000 x 10(6)/l confirmed on two occasions were included in the study. Baseline characteristics, cause of neutropenia and occurrence of infectious episodes were analysed. RESULTS: The cause of neutropenia was lymphoma in four cases (6.5%), antineoplastic chemotherapy in seven (11.3%), zidovudine in 32 (51%), trimethoprim-sulphamethoxazole (TMP-SMX) in 28 (45%) and ganciclovir in 11 (18%). Fifteen patients (24%) developed infectious complications. Neutropenia induced by chemotherapy or lymphoma was more frequently complicate by infectious episodes (P = 0.02). Neutropenia in the previous 3 months (P = 0.05), presence of a central venous catheter (P = 0.05) and a trough PMN count (P = 0.02) were the three risk factors of infection retained in a logistic model. CONCLUSION: Neutropenia induced by zidovudine, gangiclovir or TMP-SMX, are less complicated by infectious episodes than neutropenia induced by antineoplastic chemotherapy. Overall, infectious episodes in neutropenic HIV-infected patients appear lower than in patients with haemobiologic malignancies.

HIV antibody testing in France: results of a national survey.

Data from a knowledge, attitudes, and behavior survey conducted at the end of 1988 suggest that 13% of the population in France, aged 16 to 50 years, had been tested for HIV antibodies. Geographical differences in testing were observed, with a greater number of respondents tested in the regions where AIDS is most prevalent. Important variations in the percentage of HIV antibody-tested persons were found in relation to age, education, and occupation, but not to gender. Multiple logistic regression analysis identified significant predictors of HIV antibody testing. Behavioral predictors were multiple partners (odds ratio = 2.4), anal intercourse (odds ratio = 1.8), and drug use. A positive relationship was demonstrated between HIV testing and condom use. These correlations could indicate that the opportunity for anonymous and free testing encourages high-risk individuals to be tested.

Prognostic factors for neutropenic patients in an intensive care unit: respective roles of underlying malignancies and acute organ failures.

The admission of neutropenic patients to an intensive care unit (ICU) is still controversial, especially if mechanical ventilation is required. To avoid useless stays in ICU, the evaluation of the respective role of the underlying malignancy and acute organ failures might be useful for better definition of the categories of patients who could benefit from aggressive ICU support. For this purpose, we carried out a retrospective study of the charts of 107 consecutive neutropenic patients admitted to an ICU in a comprehensive cancer centre over a four-year period. The following characteristics were recorded within 24 h of admission: patient data, characteristics of neutropenia and the underlying malignancy, the type and number of organ system failures (OSFs) and simplified acute physiological scores (SAPS and SAPS II). The impact of each variable on outcome in the ICU was studied by univariate and multivariate (logistic regression) analysis. 59 patients died in the ICU (mortality rate: 55%). Patients with a haematological malignancy (n = 57, 53%) were more likely to experience respiratory failure, an underlying malignancy deemed rapidly fatal, and to have longer lasting neutropenia than patients with a solid tumour (n = 50, 47%). However, the mortality rate did not differ in the two groups (haematological malignancy 61% versus solid tumour 48%, p = 0.16). Respiratory and cardiovascular organ failure (p < 0.001 for both) correlated with mortality in the ICU. In the multiple logistic regression model, only the number of organ system failures and respiratory failure remained predictive of ICU mortality. In conclusion, the characteristics of the underlying malignancy are not relevant when deciding whether or not neutropenic patients should be admitted to an ICU. The main risk factors for death in an ICU are the number of organ failures on admission, and among them the presence of respiratory failure.

Third component of rat complement. Purification from plasma and radioimmunoassay in culture media from cell lines.

A double antibody radioimmunoassay for rat C3 has been developed. The assay required the preparation of C3 from plasma. A new purification procedure using ion exchange high performance liquid chromatography is described. The final product was homogeneous on SDS-PAGE analysis. Rat C3 has an apparent molecular weight of 187,000 and is composed of two polypeptide chains with molecular weights of 125,000 and 73,000, respectively. The purified C3 antigen with high hemolytic reactivity, as assessed by its specific functional activity, was used in preparing anti-C3 sera to perform a specific radioimmunoassay for quantifying C3 in the presence of heterologous sera contained in the cell culture media. All the validating criteria, such as precision, recovery and dilution studies, were investigated. The high sensitivity of the method allowed replicate determination of C3 in small aliquots of the cell culture medium.

Fiberoptic bronchoscopy in ventilated patients. Evaluation of cardiopulmonary risk under midazolam sedation.

One hundred seven acutely ill ventilated patients were prospectively studied to ascertain the severity and frequency of alterations in gas exchange and hemodynamic parameters during brief bronchoscopy. Sedation was performed using midazolam (0.1 mg/kg IV) without topical anesthesia. An average decline in PaO2 of 26 percent was observed at the end of the procedure, compared to the baseline value, and this was associated with a mild increase in PaCO2 in spite of the use of a special adapter. Alterations in mean systolic blood pressure appeared to be modest, consisting of a 10 percent decrease from the control level, related to sedation, and a 10 percent rise from baseline during the procedure, associated with a concomitant mild tachycardia. At that time, central hemodynamic measurements performed in a subset of 31 patients showed a significant increase in cardiac output associated with higher pulmonary wedge pressure. Fourteen patients developed hypoxemia of less than 60 mm Hg on FIO2 adjusted to 0.8. Of the ten risk factors univariately associated with hypoxemia, only the presence of ARDS (p less than 0.001) and "fighting" the ventilator during the procedure (p less than 0.05) remained significant after stepwise logistic regression. Attempts to prevent hypoxemia in critically ill patients should focus on inducing complete sedation, with careful attention to hemodynamic status, or providing maximal levels of oxygen to the ventilator (or both).

A case-control study of aplastic anaemia: occupational exposures. The French Cooperative Group for Epidemiological Study of Aplastic Anaemia.

BACKGROUND: Based on the national register of new cases of aplastic anaemia (AA) begun in France in 1984, a case-control study was conducted to explore the aetiology of the disease. METHODS: Cases were all included in the French national register of AA. Two different groups of controls were derived with individual matching; one group from hospitalized patients in the same hospital; the other group from neighbours named by the case. A 15-year occupational history was collected through interview and then grouped into exposure categories by jobs done for one year or more. The study included 98 cases, 181 hospitalized controls, and 72 neighbours aged 18-70 years. RESULTS: No differences appeared between the cases and both groups of controls relative to any group of occupation investigated. However, a borderline non-significant small excess for exposure to pesticides was observed among the cases when compared to hospitalized controls. Whatever the control group, no association was found between AA and exposure to solvents, ionizing radiation, fuel, oils and grease. A positive relationship between exposure to glues and AA was observed, as well as a trend towards an increased risk after exposure to paints. CONCLUSIONS: This large-scale case-control study confirmed the vanishing role of previously known toxic substances in the aetiology of AA. However, a higher proportion of AA patients reported exposure to paints and to glues, a relationship which needs further investigation because of the diversity of compounds included in these products.

Epidemiological survey of a major outbreak of nosocomial legionellosis.

Forty-seven nosocomial cases of legionellosis due to Legionella pneumophila serogroup 1 were diagnosed in one major outbreak from November 1982 to March 1983 in a 960-bed teaching hospital. Contaminated water was considered to be a possible source of infection because, during that period, monthly samples were found to be positive with averages of 10(4) CFU/l. After chlorination of hot water associated with flushing of outlets, nearly all samples taken in the next two years were found to be negative. A case-control study was performed to examine potential risk factors. Three groups of controls were randomly selected among eligible patients. In a multivariate analysis, only three clinical factors were found to be associated with legionellosis patients: malignant illness (relative risk, RR = 3.5), presence of an ultimately fatal disease (RR = 2.6), and exposure to corticosteroids prior to admission (RR = 7.9). Investigations of in-hospital exposures suggest that during this nosocomial outbreak diagnostic or therapeutic respiratory procedures had not increased the risk of illness. Although the epidemiological association between water contamination and disease remains unclear, the eradication of L. pneumophila from the identified supply seems to have been effective in preventing disease in this hospital.

Improved efficiency of remission induction facilitates autologous BMT harvesting and improves overall survival in adults with AML: 108 patients treated at a single institution.

A hundred and eight patients less than 60 years old with de novo acute myeloid leukemia were treated between 1982 and 1994 by protocols including final intensification with a transplant using autologous bone marrow purged by mafosfamide in first remission in the absence of an HLA-matched sibling donor available for allograft. From 1989, we attempted to improve tumor control by using high-dose anthracyclines in induction, by increasing from one to two the number of consolidation courses pre-transplant and by introducing intermediate doses of cytarabine in the first consolidation course. The CR rate was 77% (33/43) before 1989 and 90% (59/65) after 1989 (P = 0.06). Forty-five out of the 59 patients (76%) who achieved CR after 1989 could undergo bone marrow grafting in CR1 vs 16/33 (48%) before 1989 (P = 0.01). In spite of the higher proportion of patients above 50 years after 1989 (32%) toxicity was mild and an adequate graft was obtained more frequently after one collection. The principal factor relating to improvement in graft feasibility was the post-1989 modification of induction and consolidation regimens. This improvement in graft feasibility was associated with a better disease-free survival (DFS) (48 +/- 7% vs 32 +/- 8%, P = 0.04) and overall survival (OS) (53 +/- 6% vs 30 +/- 7%, P = 0.007) at 5 years. By multivariate analysis four factors were associated with overall survival (OS): karyotype, white blood cell count at diagnosis, treatment regimen and bone marrow grafting in CR1. This global approach should be prospectively compared with intensive chemotherapy.

Postdischarge nosocomial infections in primary care.

OBJECTIVE: To study both surgical and nonsurgical nosocomial infections (NIs) seen by primary-care physicians (general practitioners [GPs]) in France. DESIGN: Ongoing surveillance of postdischarge NIs by an organized group of GPs, from August 1997 to July 1999. Both the GP who personally examined the case spontaneously presenting with NI and the responsible hospital physician or surgeon were interviewed by telephone. SETTING: 305 general practices from all French regions. RESULTS: 2,199 (29%) of 7,540 patients referred for hospitalization reconsulted the GP within 30 days of discharge. In 21 (1%) of the 2,199 cases, an NI was diagnosed by the GP and confirmed as plausible by the responsible hospital physician. We diagnosed an NI in 8 (1.3%) of the post-surgical patients and in 13 (0.8%) of the non-surgical cases within the cohort. We saw eight urinary tract infections, seven surgical-site infections, three soft-tissue infections, two respiratory tract infections, and one primary bloodstream infection. In 19 patients (90%), clinical signs of NI appeared within 7 days of discharge. Assuming that all 5,431 patients who were missed for follow-up did not experience any NI, an attack rate of 0.3 per 100 admissions may be estimated for the whole group. CONCLUSION: We diagnosed 1% of NIs following discharge from a hospital in a cohort of 2,199 patients, of which 1.3% were seen post-surgery and 0.8% following nonsurgical admissions. The percentage of postdischarge visits that were for an NI in nonsurgical patients warrants a major effort with feedback to the hospital physician to reduce infection rates.

Effectiveness of simple measures to control an outbreak of nosocomial methicillin-resistant Staphylococcus aureus infections in an intensive care unit.

Between June 1985 and March 1986, 14 cases of severe nosocomial methicillin-resistant Staphylococcus aureus (MRSA) infection, including septicemia, were observed in the intensive care unit (ICU) of a 400-bed cancer reference center. Simple control measures including contact isolation of colonized patients and reinforcement of handwashing practices among personnel were followed by a sharp decrease in the rate of infection and colonization. An epidemiological investigation showed that a single serophage variant MRSA strain was involved; peak incidence of infection was 17 per 100 ICU patient discharges; the index case was identified as a patient admitted from another hospital and the epidemic strain was then transmitted from patient-to-patient in the ICU; risk factors for acquiring infection were length of prior hospitalization, invasive procedures and number of antibiotic treatments; dissemination of the strain to other wards was only anecdotal. These results stress the effectiveness of simple measures to control outbreaks of MRSA nosocomial infections even in immunocompromised cancer patients.

Safety of tracheotomy in neutropenic patients: a retrospective study of 26 consecutive cases.

OBJECTIVE: To evaluate the safety of tracheotomy in neutropenic ventilated cancer patients, in terms of infectious and haemorrhagic complications. DESIGN: Retrospective study. SETTING: A medical-surgical intensive care unit in a Cancer-hospital. PATIENTS AND PARTICIPANTS: 26 consecutive patients undergoing a tracheotomy in neutropenic period, from 1987 to 1990. INTERVENTIONS: Tracheotomy, performed at the bedside or in operating room. MEASUREMENTS AND RESULTS: In all neutropenic patients undergoing a tracheotomy, the characteristics and duration of both neutropenia and mechanical ventilation have been recorded. Stomal bleeding and infection, and infectious pneumonias and alveolar haemorrhage have been carefully reviewed. Platelets were transfused in 23 of the 26 patients at the time of the procedure; no local haemorrhage was observed. Neither stomal nor pulmonary infections secondary to tracheotomy were noted. No respiratory worsening was attributable to the tracheotomy. Nineteen patients (73%) died in ICU, without direct link between tracheotomy and death. CONCLUSIONS: These findings suggest that a tracheotomy can be safely performed in neutropenic patients requiring mechanical ventilation.