Neurosurgical Intensive Care Research Articles

Serum Procalcitonin in the diagnosis of pneumonia in the neurosurgical intensive care unit

Serum Procalcitonin in the diagnosis of pneumonia in the neurosurgical intensive care unit

Potential for reducing immobility times of a mobility monitor in-bed sensor system – a stepped-wedge cluster-randomised trial

BackgroundPressure ulcer prophylaxis is a central topic in clinical care. Pressure-relieving repositioning is strongly recommended for all pressure-sensitive patients. The Mobility Monitor (MoMo) is a technical device that records a patient’s movements and transmits the data to a monitor. This study investigated the extent to which the MoMo sensor system, which records and visualises patients’ movements in bed, supports nurses in performing pressure-relieving repositioning in neurological and neurosurgical intensive care units (ICU).MethodsThis stepped-wedge cluster-randomised trial involved two clusters: one neurological and one neurosurgical ICU. The study was carried out in two steps over three periods between November 2018 and May 2019, with a two-month interval between each step. At the beginning of the study, we equipped 33 beds across the two ICUs with a MoMo system. Our primary endpoint was the immobility rate, which is defined as the patient’s inactive time in bed exceeding two hours without pressure-relieving movements divided by the time the MoMo was in the bed. The immobility rate ranges from 0 to below 1, with higher values indicating lower mobility. Secondary endpoints were the rate of new pressure ulcers and the rate of relevant pressure-relieving repositionings. Relevant repositionings are defined as the number of repositionings identified by the MoMo as a pressure-relieving repositioning divided by the total number of repositionings,Results808 patients were included in the study, of whom 403 were in the control group and 405 were in the intervention group.The mean immobility rate was 0.171 during the control phase and 0.144 during the intervention phase. The estimated intervention effect was -0.0018 (95% confidence interval [-0.0471, 0.0436], p=0.94). The number of new pressure ulcers was 5/405 in the intervention phase and 15/403 in the control phase. We noted a small difference in the mean rate of relevant repositioningswith an estimated intervention effect of 0.046 (95% confidence interval [-0.018, 0.110], p=0.16).ConclusionOur results are insufficient to recommend the standardised use of mobility monitors in neurological or neurosurgical ICUs.Clinical trial registrationThe primary analysis was prespecified and the trial was registered in the German Clinical Trials Register (DRKS) under the reference number DRKS00015492 (31/10/2018).

Groote Schuur Hospital neurosurgical intensive care unit: A 2-year review of admission characteristics.

At Groote Schuur Hospital (GSH), the neurosurgical intensive care unit (NsICU) is a 6-bed unit headed by a specialist neurosurgeon with extensive experience in neurocritical care, working in close collaboration with intensivists from the Division of Critical Care. There is currently no detailed analysis of the demographics, diagnosis and management of patients admitted to the NsICU at GSH. To provide a detailed descriptive analysis of the demographics, diagnosis and management of patients admitted to the NsICU at GSH from 1 January 2020 to 31 December 2021. A retrospective descriptive analysis was done of patients who received treatment in the NsICU from 1 January 2020 to 31 December 2021. A total of 685 patients were admitted to the unit over a 2-year period, with a male preponderance (68.2%). The average age was 42.5 (standard deviation (SD) 17.2) years. The most common neurosurgical diagnoses were traumatic brain injuries (39.6%), brain tumours (22.6%) and aneurysmal subarachnoid haemorrhages (9.9%). Emergency admissions comprised 76.6% of the total and 86.7% of patients were admitted postoperatively. Three hundred and seventy-two patients (54.3%) required mechanical ventilation, 132 (19.3%) required both an intracranial pressure (ICP) monitor and brain tissue oxygenation monitor, 86 (12.5%) needed placement of an external ventricular drain, 50 (7.3%) needed placement of a tracheostomy tube and 16 (2.3%) needed placement of an ICP monitor only. The average duration of stay was 5.5 (1.3) days and NsICU mortality over 2 years was 11.1%. The NsICU at GSH manages predominantly male trauma patients and a significant number of admitted patients require specialised invasive intracranial monitoring. This is the first in-depth analysis of patients managed in a dedicated neurosurgical intensive care unit in South Africa. The work defines the patient population, neurosurgical pathologies and service level requirements that would likely be encountered by teams building a similar service.

Audit of Tracheostomy Care Practices in a Nigerian Tertiary Neurosurgical Intensive Care Unit According to Published Guidelines.

Chest infections are a frequently encountered problem in patients admitted to the intensive care unit (ICU), more so in tracheostomised patients. This study aimed to audit the tracheostomy care practices in patients with neurosurgical pathologies in the ICUof Wellington Clinics Abuja, a tertiary neurosurgical hospital in Nigeria. We conducted a closed-loop audit with mixed methods involving analysis of 24 patients who had tracheostomy within the first two weeks of neurosurgical pathology at a tertiary neurosurgical hospital and semi-structured interviews to determine tracheostomy care practices among the primary caregivers - nurses, intensivists, and doctors. Of the 161 ICU admissions in the first cycle, 22 patients received tracheostomy, 16 met the eligibility criteria. At re-audit (second cycle), eight of 40 patients met the criteria. All the patients received open suctioning through a dual cannula-cuffed tracheostomy tube and had independent portable suction units. In the baseline audit (first cycle), suction catheters were reused for 12-24 hours in each patient and were stored in varying combinations of normal saline and Savlon antiseptic (5 mg of cetrimide (0.5% w/w) and 1 mg of chlorhexidine digluconate (0.1% w/w)). The frequency, technique, and assessment of the need for airway suctioning were inconsistent among caregivers interviewed. All 16 patients had at least one episode of pneumonia, 10 patients had a second episode, and two patients had > two episodes. One mortality was recorded directly attributable to the complications of pneumonia. While in the re-audit, with adherence to recommendations, three patients suffered one episode of pneumonia and only one had a second episode. No mortality was recorded. A standard practice guideline wasnecessary for tracheostomy care in our ICU. In low-resource settings, stated recommendations such as single-use suction catheters and improved hygienic practices can reducerates of pneumonia in tracheostomised patients.

The impact of time to evacuation on outcomes in endoscopic surgery for supratentorial spontaneous intracerebral hemorrhage: a single-center retrospective study.

Supratentorial spontaneous intracerebral hemorrhage (SICH) can be treated with endoscopic surgery, but the optimal timing remains uncertain. We retrospectively analyzed data from 46 patients who underwent endoscopic surgery for supratentorial SICH. We examined the relationship between time to evacuation and functional outcome at 3months, adjusting for prognostic factors. Surgical outcomes and complications were compared between patients with early (≤ 12h) or late (> 12h) evacuation. Median time to evacuation was 12h, and the rate of unfavorable outcome (modified Rankin Scale > 3 at 3months) was 32.6%. Longer time to evacuation was independently associated with unfavorable outcome (odds ratio per hour delay: 1.26). Late evacuation carried a 7.25-fold higher risk of unfavorable outcome compared to early evacuation. This association held across subgroups based on hematoma volume, location, and intraventricular extension (P for interaction > 0.05). Patients with late evacuation had fewer spot signs (24% vs. 4.8%, P = 0.035) and markers of hemorrhagic expansion (36% vs. 9.5%, P = 0.018), longer neurosurgical intensive care unit (NSICU) stay (3.2 vs. 1.9days, P = 0.011) and hospital stay (15.7 vs. 11.9days, P = 0.014), and higher 30-day mortality (28.6 vs. 4%, P = 0.036) and complication rates (57.1% vs. 28.0%, P = 0.023). This study suggests a potential association between early endoscopic evacuation of supratentorial SICH and improved functional outcomes, lower 30-day mortality and reduced complications. The need for timely intervention in managing supratentorial SICH is highlighted, yet further validation through multi-center prospective studies is essential to substantiate these findings and provide a higher level of evidence.

Comparative assessment of nutritional characteristics of critically ill patients at admission and discharge from the neurosurgical intensive care unit in Korea: a comparison study

Comparative assessment of nutritional characteristics of critically ill patients at admission and discharge from the neurosurgical intensive care unit in Korea: a comparison study

A study on differences in energy expenditure according to neurosurgical intensive care unit admission period: indirect calorimetry versus predictive equations

A study on differences in energy expenditure according to neurosurgical intensive care unit admission period: indirect calorimetry versus predictive equations

Implementation of an Interprofessional Mobility Program in a Neurosurgical Intensive Care Unit.

Implementation of an Interprofessional Mobility Program in a Neurosurgical Intensive Care Unit.

Where to Draw Cerebrospinal Fluid from an External Ventricular Drain? Comparison of Cerebrospinal Fluid Parameters between Two Different Collection Sites

Background: High cerebrospinal fluid (CSF) sampling frequency is considered a risk factor for external ventricular drain (EVD)-associated infections. To reduce manipulation at the proximal port and potentially minimize the risk of an infection, we aimed to analyze whether CSF parameters sampled from the far distal collection bag could provide reliable results compared to the proximal port. Methods: We included patients who were treated with an EVD at our neurosurgical intensive care unit (ICU) between June 2021 and September 2022. CSF sampling, including microbiological analysis, was performed simultaneously from the proximal port and the collection bag. Spearman’s correlation coefficients were calculated to assess the correlation of CSF cell count, protein, lactate and glucose between the two sample sites. Results: We analyzed 290 pairs of CSF samples in 77 patients. Ventriculitis was identified in 4/77 (5%) patients. In 3/4 patients, microbiological analysis showed the same bacterial species at both sample sites at the same time. Spearman’s correlation coefficient showed that CSF cell count (r = 0.762), lactate (r = 0.836) and protein (r = 0.724) had a high positive correlation between the two collection sites, while CSF glucose (r = 0.663) showed a moderate positive correlation. Conclusion: This study shows that biochemical CSF parameters can be reliably assessed from the EVD collection bag.

Smartphone for heart sound measurement in hospital: which factors affect heart sound quality?

Abstract Background Smartphone has recently been utilized to measure heart sounds in general population, but not in hospitalized patients. Purpose To explore the factors affecting the quality of heart sounds collected from the patients using smartphone in hospital. Methods An App was developed for Android smartphone to use the built-in microphone for measuring heart sounds on cardiac apex for 20 – 30 seconds in patients. Heart sounds were visually assessed for the presence of the first and second heart sounds and categorized into good, uncertain or bad quality. The factors affecting the percentage of good-quality heart sound recordings were analyzed, including the patients’ baseline characteristics, medical history and blood test. Results Among the 296 patients enrolled, 267 (90%) patients had a good-quality heart sound recording. Good-quality heart sounds were less recorded in emergency room (76%), cardiac care unit (82%), intensive care unit (87%) and neurosurgical intensive care unit (88%). Sex, body mass index, body position (supine vs. sitting), heart rate, medical history of atrial fibrillation, coronary artery disease, heart failure or chronic obstructive pulmonary disease, did not significantly affect heart sound quality (all P > 0.05). An advanced age was related to a lower percentage of good-quality heart sounds (< 60 vs. ≥ 60 years: 94% vs. 87%, P = 0.048). Heart sound quality significantly decreased with red blood cell count, hemoglobin level and hematocrit (all P < 0.001). Conclusions Smartphone can be used in hospital settings for measuring heart sounds in good quality, but factors such as department of hospitalization, age and blood indicators need to be considered.Central illustration

Early Detection and Urgent Surgical Repair of Pharyngeal Injury after Anterior Cervical Surgery: A Case Report

This case report addresses pharyngeal perforation, an uncommon complication after anterior cervical discectomy and fusion (ACDF). A 78-year-old male, who initially presented with quadriparesis, underwent ACDF. Post-surgery, the porridge he consumed was observed in the drain tube, leading to complications such as respiratory distress and fever. An emergency exploratory surgery was performed, and a suspected pharyngeal perforation was identified and treated with primary repair and sternocleidomastoid flap. Through thorough post-operative observation, we identified critical signs requiring immediate intervention, allowing for a successful recovery without additional complications. This report emphasizes the importance of early detection and intervention for complications following ACDF, providing valuable insights for clinical practice in neurosurgical intensive care.

Evaluating the Prognostic Efficacy of Scoring Systems in Neurocritical and Neurosurgical Care: An Insight into APACHE II, SOFA, and GCS

Background: To investigate the utility of established prognostic scoring systems, such as the Acute Physiology and Chronic Health Evaluation (APACHE) II score, Sequential Organ Failure Assessment (SOFA) score, and Glasgow Coma Scale (GCS), for patients admitted to a neurosurgical intensive care unit (ICU).Methods: Among neurosurgical patients admitted to the neurosurgical ICU in a tertiary hospital from January 2015 and December 2022, only patients who had an ICU stay exceeding three days were included. The primary endpoint was in-hospital mortality. Results: In this study, a total of 3,417 patients were enrolled in the study. Of these, 3,052 (89.3%) survived until hospital discharge. Both the APACHE II and SOFA scores were significantly higher in non-survivors than in survivors (p<0.001 for both). Conversely, GCS and GCS motor score (GCS M) were substantially lower in non-survivors (p<0.001). Among the commonly used scoring systems, the APACHE II score emerged as the most effective predictor of in-hospital mortality (C-statistic of 0.887, 95% confidence interval: 0.869–0.887). Remarkably, the GCS M proved to be equally effective as the SOFA score in predicting in-hospital mortality (p=0.435) and offered the additional advantage of being simpler to use.Conclusions: Our findings indicate that these scoring systems offer valuable insights into the clinical prognosis of patients in the neurosurgical ICU. Moreover, the GCS M stands out as a feasible and reliable metric for predicting in-hospital mortality among neurosurgical ICU patients.

Ultrasound-guided placement of Midline catheters performed by Neurosurgery resident as a substitute for conventional central venous catheter placement in Neurosurgical Intensive Care Unit patients

Background: Midline catheters (MCs) and peripherally inserted central venous catheters (PICCs) are placed by a specialist or trained nurse to avoid the complications of conventional central venous catheters (CCVCs). The safety of MC placements by residents has not been evaluated. We investigated the safety of ultrasound-guided MC placement by neurosurgery residents as a substitute for CCVC placement in patients in the neurosurgical intensive care unit (ICU) requiring intravenous (IV) therapy ≤14 days.Methods: Between July 2022 and June 2023, 57 MCs for 44 patients, 36 CCVCs for 35 patients, and 32 PICCs for 32 patients were placed during their ICU stay. One resident performed all MC placements under ultrasound guidance. The baseline and procedure-related parameters of the three catheter groups (MC, CCVC, and PICC) were analyzed, with focus on the comparison of complications between MC and CCVC.Results: The first-attempt success rate was significantly higher (p=0.003) in the MC group (96.5%) than in the CCVC group (80%). Insertional injuries occurred more frequently in the CCVC group than in the MC group (13.3% vs 0%, p=0.012). The rate of completion of therapy was significantly higher (p=0.002) in the MC group (86.0%) than in the CCVC group (53.3%). The incidence of bloodstream infections did not differ significantly between the MC and CCVC groups (p=0.605). Conclusion: MCs placed under ultrasound guidance by neurosurgery residents for neurosurgical ICU patients is safe and may be a substitute for CCVC indicated for IV therapy ≤14 days.

Development and Validation of a Nomogram Prediction Model for Multidrug-Resistant Organisms Infection in a Neurosurgical Intensive Care Unit.

To develop a predictive model for assessing the risk of multidrug-resistant organisms (MDROs) infection and validate its effectiveness. Logistic regression analysis revealed that sex, hospitalization time, febrile time, invasive operations, postoperative prophylactic use of antibiotics, mechanical ventilator time, central venous catheter indwelling time, urethral catheter indwelling time, ALB, PLT, WBC, and L% were independent predictors of MDROs infection in the NICU. The area under the ROC curve for the training set and validation set were 0.880 (95% CI: 0.857-0.904) and 0.831 (95% CI: 0.786-0.876), respectively. The model's prediction curve closely matched the ideal curve, indicating excellent predictive performance. The prediction model developed in this study demonstrates good accuracy in assessing the risk of MDROs infection. It serves as a valuable tool for neurosurgical intensive care practitioners, providing an objective means to effectively evaluate and target the risk of MDROs infection.

Systematic use of standardized A-scan technique in neurosurgical intensive care unit.

Systematic use of standardized A-scan technique in neurosurgical intensive care unit.

Does Consciousness Level Determine the Timing of Successful Removal of Urinary Catheters at Neurosurgery Intensive Care Unit-Implication to Nosocomial Infection Prevention

Objective This study aimed to investigate whether the level of consciousness can be used as an indicator to determine the successful removal of urinary catheters in post-stroke patients admitted to the neurosurgical intensive care unit (NICU). Methods The study included 97 patients who had urinary catheters in NICU and were divided into three groups based on their level of consciousness. Our analyses was performed by various statistical methods, including the chi-square test, one-way ANOVA, univariate and multivariate regression. Results The overall success rate of catheter removal was found to be 62.9%. On average, the catheters were retained for 6.3 ± 3.6 days. The presence of diabetes mellitus was identified as the only factor that significantly influenced the successful removal of catheters (Adjusted Odds Ratio = 0.236, p = 0.035). No significant differences were observed among patients with different degrees of consciousness impairment. Implementing a nursing-driven circle strategy and bladder sonography scan protocol significantly could decrease the catheter-associated urinary infection (CAUTI) rate from 10.25/1,000 catheter days to 6.69/1,000 catheter days. Conclusions Based on our findings, we recommend removing urinary catheters five days after initial placement for patients with a Glasgow Coma Scale (GCS) score greater than eight and for non-intubated patients, as well as for intubated patients and aphasic patients in the NICU with a GCS score greater than 4. This protocol can potentially prevent CAUTI cases resulting from delayed UC removal

Evaluation of a clinical pharmacist-led antimicrobial stewardship program in a neurosurgical intensive care unit: a pre-and post-intervention cohort study.

Background: Antimicrobial resistance poses a significant challenge in neurosurgical intensive care units (ICU). The excessive use of broad-spectrum antibiotics is closely linked to the emergence and dissemination of drug-resistant bacteria within neurosurgical ICUs. This study assessed the effects of implementing a comprehensive Antimicrobial Stewardship (AMS) program in a neurosurgical ICU setting. Methods: From April 2022 to September 2022, an AMS program was implemented in the neurosurgical ICU. The program involved the regular presence of a pharmacist and an infectious disease physician who conducted prospective audits and provided feedback. To assess the impact of the AMS program, the outcome measures were compared between the AMS period and the 6months before AMS implementation (pre-AMS period). The primary outcome was the use of antibacterial agents, including anti-pseudomonal beta-lactams (APBLs), polymyxin, and tigecycline. Additionally, the study evaluated the appropriateness of antimicrobial de-escalation and the susceptibility of Gram-negative bacilli to antimicrobial agents. Results: A total of 526 were included during the AMS period, while 487 patients were included in the pre-AMS period. The two groups had no significant differences in disease severity and mortality rates. During the AMS period, there was a notable decrease in the use of APBLs as empiric treatment (43.92% vs. 60.99%, p < 0.001). Multi-drug resistant organism (MDRO) infections decrease significantly during AMS period (11.03% vs. 18.48%, p < 0.001). The number of prescription adjustment increased significantly in all patients (0 item vs. 0 item, p < 0.001) and MDRO-positive patients (3 items vs. 2 items, p < 0.001) during the AMS period. Additionally, appropriate antimicrobial de-escalation for patients with MDRO showed improvement during the AMS period (39.66% vs. 20%, p = 0.001). Polymyxin utilization also decreased during the AMS period (15.52% vs. 31.11%, p = 0.034). Furthermore, the susceptibility of Gram-negative Bacilli isolates to APBLs was significantly higher during the AMS period. Conclusion: Implementing a comprehensive pharmacist-led AMS program led to a decrease in the use of antibacterial agents. This reduction in usage is significant because it can potentially delay the emergence of bacterial resistance.

Implementation of an Interprofessional Mobility Program in a Neurosurgical Intensive Care Unit.

BACKGROUND: Emerging research suggests the need for implementation of mobility protocols and consistent evaluation of the impact on patient outcomes. Standardized mobility guidelines may be a solution for promoting nurse-driven mobility efforts and influencing a shift in the culture of mobility among nursing teams. In a 36-bed neurosurgical intensive care unit, 2 key areas of opportunity were identified related to patient mobility: strengthening of frontline nursing engagement in mobility and accuracy of documented early and safe mobilization. METHODS: Using a plan-do-study-act performance improvement framework, an interprofessional team developed a comprehensive early patient mobility program. Defined criteria and mobility rounds assisted in identifying the patient's functional ability and level of assistance. Nursing staff received education in case study format to enhance understanding and improve practical application of the phases of mobility. Chart reviews served as continuous assessment of documented mobilizations within the appropriate phases of mobility, and surveys evaluated staff perceptions of program success. RESULTS: On a rating scale of 1 to 5, clinicians reported a rating of 4.32 in overall observation that patients are safely mobilized earlier and more frequently and 4.48 on overall improvement in the neurosurgical intensive care unit culture of mobility. CONCLUSION: An interprofessional mobility program focusing on safe and early mobilization may improve overall culture, confidence, and empowerment of the frontline clinicians.

A prospective observational study on evaluating the efficacy of bedside optic nerve sheath diameter in assessing clinical progression of patients admitted in neurosurgical ICU with comparisons to CT scans and GCS score

Background: Ultrasound of the optic nerve sheath diameter (ONSD) is a non-invasive, repeatable tool that can be used to measure intracranial pressure in a dynamic way with high diagnostic accuracy. The goal of this study was to find out if a bedside ultrasonographic measurement of optic nerve sheath diameter (ONSD) can accurately predict the computed tomography (CT) findings of high intracranial pressure (ICP) and changes to the Glasgow Coma Scale (GCS) in adult head injury patients in the Neurosurgery ICU. Methods: For 54 patients in the neurosurgical intensive care unit, we conducted a retrospective analysis of the results of cranial ultrasounds. Those under the age of 18 and those with apparent visual injuries were ineligible. Both horizontal and vertical optic nerve sheath diameters were measured 3 mm beneath the globe in each eye using a 7.5-10MHz ultrasonographic probe. A binocular change in optic nerve sheath diameter of more than 2.00 mm was deemed abnormal in two consecutive readings in the same patient. Patients in the neurosurgical ICU were given a GCS score, which was used to classify their level of brain damage as mild, moderate, or severe. The accuracy of the optic nerve sheath diameter was evaluated using cranial CT findings of shift, oedema, or effacement that suggested an increased intracranial pressure. Results: The research has 54 participants. According to the results, 68.5% of those who took the study were men, while 31.5% of those who did so were women. Nearly 16.7 per cent of respondents were between the ages of 18 and 40, while 40.7% of respondents were between the ages of 40 and 60, and 42.6% of respondents were above 60. The significant change in ONSD- fall in GCS and CT-progression scan-findings correlation was very strong. When compared to CT scan progression, the ONSD bedside sonographic test had an 86.7% sensitivity and an 89.7% specificity for detecting elevated ICP. The Positive Predictive Value of the reduction in GCS with advancement in CT scan was 80% and the Negative Predictive Value was 89.7%, respectively. Conclusions: The sensitivity, specificity, and positive predictive value of bedside ONSD ultrasonography in predicting high intracranial pressure are significant to those of progression in CT scan and drop in GCS. An ONSD bedside measurement may be used to determine elevated ICP since it is non-invasive and repeatable.

Treatment and prognostic risk factors for intracranial infection after craniocerebral surgery.

The objective of this study was to determine risk factors of pejorative evolution course in patients suffering from postoperative cranial infection. The data of patients who developed an intracranial infection after craniocerebral surgery in the neurosurgical intensive care unit of the First Affiliated Hospital of Nanjing Medical University in Nanjing, Jiangsu, China, from February 2018 to August 2019 were retrospectively analyzed. Logistic regression was used to analyze the factors influencing the prognosis of intracranial infection treatment. Sixty-four patients developed an infection after craniocerebral surgery, and 48 of them with negative CSF cultures received experimental anti-infectives. In 16 patients, cerebrospinal fluid culture showed pandrug-resistant pathogens, including 11 Acinetobacter baumannii (11), Klebsiella pneumoniae (3), Escherichia coli (1), and Candida glabrata (1). Nine patients received intraventricular or intrathecal injections of polymyxin B. The mean duration of infection treatment was 22.2 ± 9.9 days, and the clinical cure rate was 85.9% (55/64). Logistic multivariate regression analysis showed that inadequate CSF drainage (OR, 6.839; 95% CI, 1.130-41.383; P = 0.036) and infection with drug-resistant bacteria (OR, 24.241; 95% CI, 2.032-289.150; P = 0.012) were independent risk factors for postoperative intracranial infection. Intracranial infection with positive CSF culture and inadequate CSF drainage are factors contributing to the poor prognosis of intracranial infection. Moreover, early anti-infection treatment and adequate CSF drainage may improve patient outcomes. In particular, intraventricular or intrathecal injection of polymyxin B may be a safe and effective treatment strategy for MDR/XDR gram-negative bacilli infection.