Risk factors for neurosurgical site infections after craniotomy: a critical reappraisal of antibiotic prophylaxis on 4578 patients

The objective of this study was to evaluate incidence and risk factors of postoperative infections, with emphasis on antibiotic prophylaxis, in a series of 4578 craniotomies. A prospective database was implemented for surveillance of postcraniotomy infections. During period A, no antibiotic prophylaxis was prescribed for scheduled, clean craniotomies, lasting less than 4 h, whereas emergency, clean-contaminated or long-lasting craniotomies received cloxacillin or amoxicillin-clavulanate. During period B, prophylaxis was given to every craniotomy. The effect of prophylaxis on craniotomy infections, independently of other risk factors, was studied by multivariate analysis. The overall infection rate was 6.6%. CSF leak, male gender, surgical diagnosis, surgeon, early re-operation, surgical duration and absence of prophylaxis were independent risk factors. CSF leak had the highest odds ratio. Antibiotic prophylaxis decreased infection rate from 9.7% down to 5.8% in the entire population (p<0.0001) mainly by decreasing rates in low risk patients from 10.0% down to 4.6% (p<0.0001). Antibiotic prophylaxis in craniotomy is effective in preventing surgical site infections even in low-risk patients.

Prevention of external ventricular drain--related ventriculitis

OBJECTIVE

The purpose of this study was to test if a reduction of external ventricular drains (EVD) related ventriculitis could be achieved by a strict protocol of care and if protocol violation was associated with a higher incidence of EVD-related ventriculitis.

METHODS

A written protocol for EVD insertion, nursing and surveillance was implemented. A retrospective comparison of EVD-related ventriculitis incidence was performed between control (161 EVD in 131 patients) and study periods (216 EVD in 175 patients). Risk factor analysis was performed in patients in whom an EVD was inserted during the study period including the relationship between protocol compliance and ventriculitis. A score for the number of protocol violations (absence of hair clipping, absence of a tunnelled EVD, absence of shampooing, incorrect dressing change, inappropriate CSF bag or tap samplings and EVD manipulation) was established for each patient.

RESULTS

Incidence of patient-related ventriculitis decreased from 12.2% (1999) down to 5.7% (p<0.05) as well as incidence of EVD-related ventriculitis (9.9% vs 4.6%, p<0.05). During the study period, the only statistically significant risk factors for infection were CSF leak and protocol violations. The mean protocol violation score was 4 times higher in the infected versus the non-infected patients (p<0.0001). Patients with a violation score of 0 or 1 had no infection (EVD duration 2 to 42 days).

CONCLUSION

EVD can be left safely, as long as needed, provided that meticulous care is taken for EVD insertion and nursing. EVD duration seems to have no effect on infection incidence.

Empirical treatment of adult postsurgical nosocomial meningitis

BACKGROUND

The combination of cefotaxime and fosfomycin (CTX-FOS) has been proposed in France for the empirical treatment of postoperative nosocomial meningitis since the late 1980s. The purpose of this work was to evaluate this strategy today, as well as other possible treatments.

METHODS

Each patient undergoing a neurosurgical procedure was prospectively included in a database designed for the surveillance of surgical site infection (SSI). For each meningitis detected, we analysed the in vitro susceptibility of the causative micro-organisms to cefotaxime alone (CTX), cefotaxime-fosfomycin (CTX-FOS), vancomycin (VAN) and cefotaxime-vancomycin (CTX-VAN) combinations. The patient population was divided into two groups according to the presence or absence of CSF shunting material.

FINDINGS

116 patients had had a postoperative meningitis/ventriculitis during the last 36 months, among 6447 patients undergoing neurosurgery in our department (1.8%). Ten patients had aseptic meningitis (8.6%). Overall sensitivity to CTX was 69.8%, as compared to 77.3% with CTX-FOS combination (NS). This result was due to a large proportion of fosfomycin resistant cocci in our population. The CTX-VAN combination increased the overall in vitro susceptibility up to 91.5%, but the benefit of this combination was only significant in CSF shunting material patients. In these latter patients, VAN was as effective as CTX-FOS combination.

INTERPRETATION

CTX-FOS combination is no longer the best choice for empirical treatment of post neurosurgical meningitis. CTX alone can be safely used in patients without a CSF shunt; in those with either a ventriculostomy or a CSF shunt associated ventriculitis, a CTX-VAN combination could improve treatment efficacy, provided that high doses of vancomycin are used to ensure correct CSF diffusion.

Anaphylaxis to rocuronium

Reports about anaphylactic and anaphylactoid reactions to rocuronium have increased recently. We report two new cases of documented grade III anaphylaxis, leading to death in one patient. The first case occurred in an 81-year-old ASA II woman scheduled for emergency abdominal surgery. Severe hypotension and tachycardia were observed after rocuronium, without bronchospasm. Neosynephrine allowed rapid resuscitation, and the patient recovered fully. The second patient was a 64-year-old ASA II man scheduled for abdominal surgery. Severe haemodynamic instability and bronchospasm occurred after rocuronium. Despite immediate life support, the postoperative period was complicated by persistent low systolic pressure, acute respiratory distress syndrome, acute renal failure, disseminated intravascular coagulation and pancreatitis, leading to the death of the patient.

A multi-state model for evolution of intensive care unit patients: prediction of nosocomial infections and deaths

Nosocomial (hospital-acquired) infections are very frequent in intensive care units (ICU). The risk of death after severe infection is high, but the precise rate of death in ICU attributable to nosocomial infection is not known. The goal of this project was to build a statistical model to predict the occurrence of nosocomial infections in ICU and the outcome of the patients. We collected data on 676 consecutive patients admitted to an ICU for more than 24 hours between 1993 and 1996. The following data were collected for each patient: history; clinical examination at entry; subsequent infections; outcome. A multi-state heterogeneous semi-Markov model was determined and then validated; the initial data set was randomly split into two groups: two-thirds (450 patients) to build the model and one-third (226 patients) to validate it. The model defined five states: ICU admission; first simple infection; first complicated infection; death, and discharge from the ICU. Transitions between these states determined nine different events. The global model of patient histories can be divided into nine survival models, each corresponding to one of these events. The possible events from a given state were considered to be competing. Since many risk factors induced non-proportional hazard functions, piecewise exponential models were used to model event occurrence. The effect of continuous covariates on hazard functions has been described with a non-parametric method that enables non-linear relations to be shown. Among other things, the model allows patients' post-admission histories to be predicted from data available at ICU admission. The bootstrap estimator of the attributable risk of death due to simple or complicated nosocomial infections is 44.2 percent (95 percent CI 26.0-61.6 percent). We were also able to characterize the most highly exposed patients, those who comprise the high-risk group on whom prevention efforts must be focused.

[Ecological consequences of preventive antibiotic prescriptions]

Assessing the ecological impact of preventive antibiotherapy in hospital practice is an important piece in the strategies aiming at circumventing the development of bacterial resistance. In the present review of the literature, two situations will be taken into account: surgical antibioprophylaxis and selective digestive decontamination. Only the consequences of these on bacterial flora will be considered. Despite some discrepancies, only partially attributable to methodological differences, data as a whole are consistent. For antibioprophylaxis, they confirm the importance of a strict observance of the right therapeutic regimen, especially the duration of treatment. Selective digestive decontamination unquestionably encounters hazards of selecting a resistant flora. Monitoring the intestinal flora under treatment is mandatory. The indications must remain strictly limited.

[Infectious risk from ventriculostomy]

Ventriculostomy is a useful technique for the management of acute hydrocephalus or increased intracranial pressure. The mean rate of ventricular infections is 10%. This risk can be decreased by selecting indications, adherence to aseptic insertion techniques, avoiding CSF leakage, tunneling the catheter, using closed systems and limiting line manipulations. Duration of ventriculostomy drainage remains controversial, as well as systematic change of drain every five days of drainage. The value of local or general prophylactic antibiotic treatment remains to be substantiated.

Risk factors for neurosurgical site infections after craniotomy: a prospective multicenter study of 2944 patients. The French Study Group of Neurosurgical Infections, the SEHP, and the C-CLIN Paris-Nord. Service Epidémiologie Hygiène et Prévention

OBJECTIVE

To determine the incidence and risk factors of surgical site infections (SSIs) after craniotomy and to test the risk index score proposed by the National Nosocomial Infections Surveillance (NNIS) system, which, to our knowledge, has not been validated in neurosurgery to date.

METHODS

During a 15-month period, every adult patient undergoing craniotomy in 10 neurosurgical units was prospectively evaluated for development and risk factors of SSI. The follow-up period was at least 30 days. SSIs were defined according to the Center for Disease Control definitions. Incidence was calculated per patient. Multivariate analyses were conducted at first to include all significant risk factors of univariate analysis and then only those known preoperatively. Finally, the NNIS risk index was tested in this population.

RESULTS

Of a total of 2944 patients, 117 patients (4%) with SSIs were observed, including 30 with wound infections, 14 with bone flap osteitis, 56 with meningitis, and 17 with brain abscesses. Independent risk factors for SSIs were postoperative cerebrospinal fluid leakage (odds ratio, 145; 95% confidence interval, 72-293) and subsequent operation (odds ratio, 7; 95% confidence interval, 4-12). Independent predictive risk factors were emergency surgery, clean-contaminated and dirty surgery, an operative time longer than 4 hours, and recent neurosurgery. Absence of antibiotic prophylaxis was not a risk factor. The NNIS risk index was effective in identifying at-risk patients.

CONCLUSION

Independent risk factors for SSIs after craniotomy involve postoperative events. However, the NNIS risk index is effective in identifying at-risk patients.

[Cerebral abscess and empyema]

Brain abscess and subdural empyema are serious infections which can be metastasis of chronic suppurative diseases (bronchectasia, lung or abdominal abscesses) or of congenital cardiopathy, but they are more frequently seen in healthy adults suffering from chronic sinusitis or otitis. Brain CT scan with contrast media injection is the best tool for diagnosis and follow-up. It has transformed the prognosis of brain abscesses. Anaerobic oropharyngeal microflora is the main source of bacteria responsible for suppurative brain diseases. Surgical treatment consists of aspiration or, rarely now, of excision of the lesion. Medical treatment alone can be successful in selected cases, provided patients are closely monitored and antibiotics with good penetration into the brain parenchyma are used.

Inhaled nitric oxide in acute respiratory failure: dose-response curves

OBJECTIVE

To determine the dose-response curve of inhaled nitric oxide (NO) in terms of pulmonary vasodilation and improvement in PaO2 in adults with severe acute respiratory failure.

DESIGN

Prospective randomized study.

SETTING

A 14-bed ICU in a teaching hospital.

PATIENTS

6 critically ill patients with severe acute respiratory failure (lung injury severity score > or = 2.5) and pulmonary hypertension.

INTERVENTIONS

8 concentrations of inhaled NO were administered at random: 100, 400, 700, 1000, 1300, 1600, 1900 and 5000 parts per billion (ppb). Control measurements were performed before NO inhalation and after the last concentration administered. After an NO exposure of 15-20 min, hemodynamic parameters obtained from a fiberoptic Swan-Ganz catheter, blood gases, methemoglobin blood concentrations and intratracheal NO and nitrogen dioxide (NO2) concentrations, continuously monitored using a bedside chemiluminescence apparatus, were recorded on a Gould ES 1000 recorder. In 2 patients end-tidal CO2 was also recorded.

RESULTS

The administration of 100-2000 ppb of inhaled NO induced: i) a dose-dependent decrease in pulmonary artery pressure and in pulmonary vascular resistance (maximum decrease--25%); ii) a dose-dependent increase in PaO2 via a dose-dependent reduction in pulmonary shunt; iii) a slight but significant decrease in PaCO2 via a reduction in alveolar dead space; iv) a dose-dependent increase in mixed venous oxygen saturation (SVO2). Systemic hemodynamic variables and methemoglobin blood concentrations did not change. Maximum NO2 concentrations never exceeded 165 ppb. In 2 patients, 91% and 74% of the pulmonary vasodilation was obtained for inhaled NO concentrations of 100 ppb.

CONCLUSION

In hypoxemic patients with pulmonary hypertension and severe acute respiratory failure, therapeutic inhaled NO concentrations are in the range 100-2000 ppb. The risk of toxicity related to NO inhalation is therefore markedly reduced. Continuous SVO2 monitoring appears useful at the bedside for determining optimum therapeutic inhaled NO concentrations in a given patient.

Prevention of gram negative nosocomial bronchopneumonia by intratracheal colistin in critically ill patients. Histologic and bacteriologic study

OBJECTIVE

To evaluate the efficiency of intratracheal colistin in preventing nosocomial bronchopneumonia (BPN) in the critically ill.

DESIGN

Study evaluating the clinical incidence of nosocomial BPN in 2 groups of critically ill patients who receive or did not receive intratracheal colistin. BPN was assessed clinically in survivors and histologically in non-survivors.

SETTING

A 14-bed surgical intensive care unit.

PATIENTS

598 consecutive critically ill patients were studied during a prospective non-randomized study over a 40-month period.

INTERVENTIONS

251 patients--31 non-survivors and 220 survivors--did not receive intratracheal colistin and 347-42 non-survivors and 305 survivors--received intratracheal colistin for a 2-week period (1,600,000 units per 24 h).

MEASUREMENTS AND RESULTS

The incidence of nosocomial BPN was evaluated clinically in survivors, using repeated protected minibronchoalveolar lavages, and histologically in non-survivors via an immediate postmortem pneumonectomy (histologic and semi-quantitative bacteriologic analysis of one lung). The clinical incidence of nosocomial BPN was of 37% in coli (-) survivors and of 27% in coli (+) survivors (p < 0.01). This result was histologically confirmed in non-survivors, where the incidence of histologic BPN was of 61% in coli (-) patients and of 36% in coli (+) patients (p < 0.001). Emergence of BPN due to colistin-resistant micro-organisms was not observed. Because colistin was successful in preventing Gram-negative BPN and did not change the absolute number of Gram-positive BPN, the proportion of BPN caused by staphylococcus species was higher in group coli (+) patients (33% vs 16%). Mortality was not significantly influenced by the administration of colistin.

CONCLUSION

This study suggests that the administration of intratracheal colistin during a 2-week period significantly reduces the incidence of Gram-negative BPN without creating an increasing number of BPN due to colistin-resistant micro-organisms.

[Antibiotic prophylaxis in multiple trauma patients]

Infection prophylaxis in multiple trauma patients includes prophylaxis of infections due to surgery, which is the true one as well as the prophylaxis of secondary acquired infections which are more frequent, especially in case of co-existing shock. The association an aminopenicillin with a beta-lactamase inhibitor is recommended for prophylaxis of surgical infections. These antibiotics need to be administered early and in high doses, as the pharmacokinetic parameters are modified in trauma patients, with an increased volume of distribution and a shortened half-life of elimination. Prevention of secondary infection relies on a medico-surgical treatment of haemorrhagic shock. Other preventive measures, such as early enteral nutrition, selective decontamination of the digestive tract and immunotherapy, still need to prove their efficacy.

Selective decontamination of the digestive tract in neurosurgical intensive care unit patients: a double-blind, randomized, placebo-controlled study

OBJECTIVES

The aim of this study was to assess, in a selected population, the effects of selective decontamination of the digestive tract on colonization of the oropharynx, trachea, stomach and rectum, and on the infection rate. An economical assessment was also performed.

DESIGN

A prospective, double-blind, randomized, placebo-controlled, dual-center trial.

SETTING

Two neurosurgical intensive care units.

PATIENTS

A total of 191 comatose patients admitted emergently and intubated within < 24 hrs were enrolled. Of these patients, 68 were excluded because they either died, got an early infection, or were extubated within the first 5 days. A total of 123 patients were analyzed: 63 treated and 60 placebo patients.

INTERVENTIONS

Topical antibiotics (tobramycin, polymyxin E, amphotericin B) were applied in the oropharynx and in the stomach. Vancomycin was added in the oropharyngeal paste. Placebo patients received the same regimen (i.e., a suspension of fluid and a paste) but without antibiotics. No parenteral antibiotics were given during the study period.

MEASUREMENTS AND MAIN RESULTS

Bronchopneumonia episodes were diagnosed with protected specimen brush or plugged telescoping catheter and other infections were diagnosed according to the Center for Disease Control of Atlanta criteria. Antibiotic costs and cost per survivor were calculated. Selective decontamination of the digestive tract significantly reduced Gram-negative bacilli colonization as well as the number of episodes of bronchopneumonia, urinary tract infections, and sinusitis. Despite the addition of vancomycin, Staphylococcus aureus remained the main potential pathogen causing tracheal colonization and subsequent bronchopneumonia. The reduction in bronchopneumonia rate was observed in head-trauma patients only. We were able to show that: a) the trachea was the main reservoir of microorganisms responsible for pneumonia; b) pneumonia developed after tracheal colonization. Total charges for antibiotics were 2.8 times higher in the treated group than in the placebo group; in calculating the cost per survivor, selective decontamination of the digestive tract might be beneficial due to the reduced length of stay.

CONCLUSIONS

Selective decontamination of the digestive tract is an effective technique in reducing infectious morbidity in comatose neurosurgical patients. Because of its cost, this technique should be used only in selected populations.

[Humoral immunity, 5 years after anti-tetanus vaccination, in a group of malaria-infected and malnourished African children]

In 1978 a campaign of vaccination against tetanus was conducted in a savannah biotope of Burkina Faso (Garango). The effects of 1 or 2 tetanus toxoid injections and of concomitant malnutrition and malaria infection were assessed by measurements of specific antibody and cell-mediated responses. None of these 2 variables did interfere with the development of anti-tetanus immunity. In 1983, 5 years later, similar results were obtained, giving evidence that in spite of malnutrition and malaria, factors known for their immunosuppressive action, a good degree of specific protection was acquired. This local survey revealed also that multiple schemes of vaccination, 1 to 5 injections of vaccine over 5 years, had been performed by unidentified operators. The issues raised by such incongrous, costly and possibly detrimental practices are discussed within the frame of national vaccination policies.

[Treatment with a cefotaxime-fosfomycin combination of staphylococcal or enterobacterial meningitis in adults]

Thirty-two patients were included in this trial: 22 with staphylococcal meningitis (including 5 methicillin-resistant) and 10 with enterobacterial meningitis. Mean duration of treatment was 14.5 and 15.9 days respectively. The combination was synergistic in vitro against 10 of the 12 strains of Staphylococcus and 5 of the 6 strains of Enterobacteriaceae studied. Bacteriological sterilization occurred in all cases which could be evaluated, and clinical recovery was obtained in 95.2% of patients with staphylococcal meningitis (4 unrelated deaths) and 100% of patients with enterobacterial meningitis (2 deaths). Bactericidal power of the cerebro-spinal fluid, often less than 1/8, was not correlated with effectiveness against Staphylococci. Mean CSF concentrations of cefotaxime, desacetylcefotaxime and fosfomycin on the 2nd and 15th days of treatment were 4, 3.5 and 39.8 mg/l and 2.2, 2.1 and 28.0 mg/l, respectively. Clinical and biological acceptability was satisfactory. There were three cases of superinfection or colonization, by Pseudomonas and Enterobacter.

Effect of postoperative extradural morphine on ADH secretion

The effect of extradural morphine on antidiuretic hormone (ADH) secretion was assessed for the first 6 h after surgery in three groups of patients. Surgery was conducted under extradural bupivacaine: thereafter patients in group I (n = 6) received further injections of bupivacaine, patients in group II (n = 6) received an extradural injection of morphine and in patients in group III (n = 5) both bupivacaine and morphine, were administered extradurally. In group I, plasma ADH values remained unchanged throughout the study. In contrast, in the two groups of patients receiving extradural morphine a delayed and stepwise increase in plasma ADH concentration was documented. These results indicate that extradural morphine induces ADH secretion and suggest that this effect is the consequence of the migration of morphine to the brainstem.

[Cell-mediated immunity study by skin testing in 129 critically ill patients (author's transl)]

Delayed hypersensitivity skin testing was performed in 129 critically ill patients. Six intradermal antigens were used: tuberculin, candidin, varidase, epidermophytin, trichophytin and CCB (a polyvalent microbial vaccine from the Pasteur Institute). The response was judged as positive when one test or more were positive. Patients were devided in four groups: group A (40 cases): non-infected patients, a priori without immunodeficiency; group B (14 cases); suspected of immunodeficiency (cancers, hemopathies, collagen diseases receiving corticosteroids); group C (24 cases): decompensated chronic respiratory insufficienceis; group D (50 cases): overwhelming sepsis (septicaemias, septic acute respiratory distress syndromes, thoracic empyemas, purulent meningitis, peritonitis, mediastinitis). A significant diminution of delayed hypersensitivity was observed in groups B, C and D. No relation was found between delayed hypertensitivity and prognosis in groups C and D.