Characteristics and determinants of outcome of hospital-acquired bloodstream infections in intensive care units: the EUROBACT International Cohort Study

Purpose

The recent increase in drug-resistant micro-organisms complicates the management of hospital-acquired bloodstream infections (HA-BSIs). We investigated the epidemiology of HA-BSI and evaluated the impact of drug resistance on outcomes of critically ill patients, controlling for patient characteristics and infection management.

Methods

A prospective, multicentre non-representative cohort study was conducted in 162 intensive care units (ICUs) in 24 countries.

Results

We included 1,156 patients [mean?±?standard deviation (SD) age, 59.5?±?17.7?years; 65?% males; mean?±?SD Simplified Acute Physiology Score (SAPS)?II score, 50?±?17] with HA-BSIs, of which 76?% were ICU-acquired. Median time to diagnosis was 14 [interquartile range (IQR), 7–26] days after hospital admission. Polymicrobial infections accounted for 12?% of cases. Among monomicrobial infections, 58.3?% were gram-negative, 32.8?% gram-positive, 7.8?% fungal and 1.2?% due to strict anaerobes. Overall, 629 (47.8?%) isolates were multidrug-resistant (MDR), including 270 (20.5?%) extensively resistant (XDR), and 5 (0.4?%) pan-drug-resistant (PDR). Micro-organism distribution and MDR occurrence varied significantly (p?<?0.001) by country. The 28-day all-cause fatality rate was 36?%. In the multivariable model including micro-organism, patient and centre variables, independent predictors of 28-day mortality included MDR isolate [odds ratio (OR), 1.49; 95?% confidence interval (95?%CI), 1.07–2.06], uncontrolled infection source (OR, 5.86; 95?%CI, 2.5–13.9) and timing to adequate treatment (before day?6 since blood culture collection versus never, OR, 0.38; 95?%CI, 0.23–0.63; since day?6 versus never, OR, 0.20; 95?%CI, 0.08–0.47).

Conclusions

MDR and XDR bacteria (especially gram-negative) are common in HA-BSIs in critically ill patients and are associated with increased 28-day mortality. Intensified efforts to prevent HA-BSIs and to optimize their management through adequate source control and antibiotic therapy are needed to improve outcomes.     

Detection of colorectal tumors with water enema-multidetector row computed tomography

Purpose

To retrospectively determine the diagnostic capabilities of water enema-multidetector row computed tomography (WE-MDCT) in the detection of colorectal tumors.

Materials and methods

One hundred and one patients (55 male, 46 female) who had WE-MDCT and videocolonoscopy because of suspected colorectal tumors were included. Results of complete videocolonoscopy, surgery, and histopathologic analysis were used as standard of reference. Sensitivity, specificity, and accuracy, and positive and negative predictive values of WE-MDCT for the diagnosis of colorectal tumors were estimated with 95% confidence intervals (CIs).

Results

Ninety-two colorectal tumors (64 malignant, 28 benign) were confirmed in 71 patients (prevalence, 71/101; 70%). Overall sensitivity for colorectal tumor detection was 87% (80/92; 95%CI: 78%?C93%) on a per lesion basis. For malignant and benign tumor detection, sensitivity was 100% (64/64; 95%CI: 94%?C100%) and 57% (16/28; 95%CI: 37%?C76%), respectively. For colorectal tumors ??10?mm, sensitivity was 99% (76/77; 95%CI: 93%?C100%). Seventy-nine of the 83 colorectal tumors ??6?mm were detected, yielding a sensitivity of 95% (79/83; 95%CI: 88%?C99%) for this specific size category. On a per patient basis, WE-MDCT had a sensitivity of 100% (71/71; 95%CI: 94%?C100%), a specificity of 100% (30/30; 95%CI: 88%?C100%), an accuracy of 100% (101/101; 95%CI: 96%?C100%), a positive predictive value of 100% (71/71; 95%CI: 94%?C100%), and a negative predictive value of 100% (30/30; 95%CI: 86%?C100%) for the diagnosis of colorectal tumor.

Conclusion

Our results suggest that WE-MDCT is a promising imaging technique for the detection of malignant colorectal tumors. However, our results should be validated by larger and prospective studies.     

Treatment-related side effects and quality of life in cancer patients

Background

Cancer leads to a complicated pattern of change in quality of life (QoL).

Objective

The aims of this study were to assess the impact of treatment-related side effects on QoL in cancer patients and to explore which other factors, and to what extent, contribute to explain low QoL scores.

Methods

One hundred twenty-three cancer patients receiving chemotherapy completed the self-administered questionnaires (Medical Outcomes Short-Form-36 (SF-36) and 12-item General Health Questionnaire). Multiple regression analyses were conducted with the SF-36 physical component summary (PCS) and SF-36 mental component summary (MCS) scores as the dependent variables and demographic and clinical factors as independent variables.

Results

Seventy-two percent of patients experienced treatment-related side effects, and 32% resulted positive for psychiatric diseases. Two multivariate analyses showed that worse PCS scores, like worse MCS scores, were significantly and independently predicted by treatment-related side effects (odds ratio (OR)?=?5.00, 95%CI 1.29–19.45; OR?=?8.08, 95%CI 2.03–32.22, respectively) and changes in health over the last 12?months (OR =2.34, 95%CI 1.47–3.76; OR?=?3.21, 95%CI 1.90–5.41, respectively), after adjustment for age, gender, years of school, time from cancer diagnosis, and psychiatric disease.

Conclusions

Given the new emphasis on QoL, we suggest that physicians have a responsibility to openly discuss therapy efficacy, prognosis as well as the potential for adverse events with their patients. Changes in health, as perceived by patient, should also be monitored at follow-up.     

A prospective trial of elective extubation in brain injured patients meeting extubation criteria for ventilatory support: a feasibility study

Introduction

To assess the safety and feasibility of recruiting mechanically ventilated patients with brain injury who are solely intubated for airway protection and randomising them into early or delayed extubation, and to obtain estimates to refine sample-size calculations for a larger study. The design is a single-blinded block randomised controlled trial. A single large academic medical centre is the setting.

Methods

Sixteen neurologically stable but severely brain injured patients with a Glasgow Coma Score (GCS) of 8 or less were randomised to early or delayed extubation until their neurological examination improved. Eligible patients met standard respiratory criteria for extubation and passed a modified Airway Care Score (ACS) to ensure adequate control of respiratory secretions. The primary outcome measured between groups was the functional status of the patient at hospital discharge as measured by a Modified Rankin Score (MRS) and Functional Independence Measure (FIM). Secondary measurements included the number of nosocomial pneumonias and re-intubations, and intensive care unit (ICU) and hospital length of stay. Standard statistical assessments were employed for analysis.

Results

Five female and eleven male patients ranging in age from 30 to 93 years were enrolled. Aetiologies responsible for the neurological injury included six head traumas, three brain tumours, two intracerebral haemorrhages, two subarachnoid haemorrhages and three ischaemic strokes. There were no demographic differences between the groups. There were no unexpected deaths and no significant differences in secondary measures. The difference in means between the MRS and FIM were small (0.25 and 5.62, respectively). These results suggest that between 64 and 110 patients are needed in each treatment arm to detect a treatment effect with 80% power.

Conclusions

Recruitment and randomisation of severely brain injured patients appears to be safe and feasible. A large multicentre trial will be needed to determine if stable, severely brain injured patients who meet respiratory and airway control criteria for extubation need to remain intubated.     

Metastatic epidural spinal cord compression among elderly patients with advanced prostate cancer

Background

A recent randomized trial demonstrated that for metastatic epidural spinal cord compression (MESCC), a complication of advanced prostate cancer, surgical decompression may be more effective than external beam radiation therapy (RT). We investigated predictors of MESCC, its treatment, and its impact on hospital length of stay for patients with advanced prostate cancer.

Methods

We used the SEER-Medicare database to identify patients >65 years with stage IV (n?=?14,800) prostate cancer. We used polytomous logistic regression to compare those with and without MESCC and those hospitalized for treatment with surgical decompression and/or RT.

Results

MESCC developed in 711 (5 %) of patients, among whom 359 (50 %) received RT and 107 (15 %) underwent surgery?±?RT. Median survival was 10 months. MESCC was more likely among patients who were black (OR 1.75, 95 %CI 1.39–2.19 vs. white) and had high-grade tumors (OR 3.01, 95 %CI 1.14–7.94), and less likely in those younger; with prior hormonal therapy (OR 0.73, 95 %CI 0.62–0.86); or with osteoporosis (OR 0.63, 95 %CI 0.47–0.83). Older patients were less likely to undergo either RT or surgery, as were those with ≥1 comorbidity. Patients with high-grade tumors were more likely to undergo RT (OR 1.92, 95 %CI 1.25–2.96). Those who underwent RT or surgery spent an additional 11 and 29 days, respectively, hospitalized.

Conclusions

We found that black men with metastatic prostate cancer are more likely to develop MESCC than whites. RT was more commonly utilized for treatment than surgery, but the elderly and those with comorbidities were unlikely to receive either treatment.     

A descriptive study of persistent oxaliplatin-induced peripheral neuropathy in patients with colorectal cancer

Background

Prolonged neurotoxicity after systemic chemotherapy has the potential to impact on quality of life. We explored the frequency of persistent peripheral neuropathy in patients who received oxaliplatin for colorectal cancer at two local centres.

Patients and methods

Questionnaires were sent to patients who completed treatment with oxaliplatin for colorectal cancer at least 20 months prior to entering the study. Neuropathy questions were adapted from the FACT/GOG-Ntx (V.4) questionnaire.

Results

Of the 56 eligible patients, 27 returned the questionnaire. Twenty-five patients (93 %) experienced neuropathic symptoms during their treatment; 11 had grade-2, and two had grade-3 symptoms. At the time of completing the questionnaire, 17 patients (63.0 %; 95%CI 43.9–79.4 %) were still symptomatic with 12 patients (44.4 %; 95%CI 26.8–63.3) having grade-2 or grade-3 symptoms and three patients (11.1 %; 95%CI 2.9–27.3) having grade-3 neuropathic symptoms. Participants who received more than 900 mg/m² oxaliplatin had a significantly higher risk of persistent grade-2 or grade-3 neuropathy (p?=?0.031, RR?=?8.3 95%CI?=?1.2–57.4). There was a trend toward increased risk of persistent neuropathy of any grade among participants with a history of regular alcohol use (p?=?0.051; RR?=?1.7 95%CI 1.0–2.8).

Conclusion

Persistent oxaliplatin-induced neuropathy is not as uncommon as previously suggested, and the rate of grade-2 and grade-3 symptoms could be considerably higher than previous reports.     

Screening for obstructive sleep apnea in veterans with ischemic heart disease using a computer-based clinical decision-support system

Objectives

To assess the validity of a handheld clinical decision-support system (CDSS) in detecting obstructive sleep apnea (OSA) in veterans with ischemic heart disease against polysomnography (PSG) and to compare the diagnostic accuracy of the CDSS versus the Berlin questionnaire.

Methods

We enrolled prospectively 143 patients with underlying ischemic heart disease. Veterans with history of neurologic disease, systolic congestive heart failure, or receiving opiates were excluded from participation. Participants were asked to complete the Berlin Questionnaire and to answer all eight questions of CDSS-software. At the end of the interview, veterans were scheduled for an in-laboratory polysomnogram.

Results

Ninety one patients completed the study. The prevalence of OSA (AHI ≥5/h) was 74.7?% with a median AHI of 11.5/h (range 0–90). When compared to PSG, the CDSS and the Berlin questionnaire achieved a sensitivity of 98.5?% [95?% confidence interval (CI) 92.1–100] and 80.9?% (95?% CI 69.5–89.4) and a specificity of 86.9?% (95?% CI 66.4–97.2) and 39.1?% (95?% CI 19.7–61.5) at a threshold value of AHI ≥5 with a corresponding area under the curve of 0.93 (95?% CI 0.85–0.97) and 0.60 (95?% CI 0.49–0.70); respectively.

Conclusions

CDSS is a superior screening tool for identifying cardiac veterans with undiagnosed OSA than the BQ.     

In-hospital outcomes after elective and non-elective percutaneous coronary interventions in hospitals with and without on-site cardiac surgery backup

Background

Guidelines recommend on-site surgery backup (SB) when elective percutaneous coronary intervention (PCI) is performed. The evidence for this recommendation is however weak.

Objectives

The objective of the present study was to compare clinical outcomes in patients undergoing PCI in hospitals with SB or without surgery backup (non-SB).

Methods

Prospective German PCI registry in 36 hospitals throughout Germany. Consecutive procedures were collected and analyzed centrally.

Results

In 2006, a total of 23,148 patients were included; 12,465 patients (53.8%) in 11 hospitals with SB and 10,683 patients (46.2%) in 25 hospitals without on-site cardiac SB. Both patient groups were well-balanced with regard to age and gender. SB hospitals had more patients with ACS (OR 1.29; 95%CI 1.23–1.36) and less patients with stable angina (OR 0.78; 95%CI 0.74–0.82) than non-SB hospitals. There was no indication of a clinically relevant differential outcome for in-hospital death, MACE, non-fatal MI, non-fatal stroke/TIA, or emergency CABG between SB and non-SB hospitals for neither patients with ACS nor stable angina except for emergency CABG in ACS patients (more frequent in SB hospitals, OR 2.29; 95%CI 1.02–5.13).

Conclusions

There was no evidence of an excess risk associated with PCI-procedures performed in non-SB hospitals.     

Postal survey methodology to assess patient satisfaction in a suburban emergency medical services system: an observational study

Background

The epidemiology of acute pancreatitis in the United States is largely unknown, particularly episodes that lead to an emergency department (ED) visit. We sought to address this gap and describe ED practice patterns.

Methods

Data were collected from the National Hospital Ambulatory Medical Care Survey between 1993 and 2003. We examined demographic factors and ED management including medication administration, diagnostic imaging, and disposition.

Results

ED visits for acute pancreatitis increased over the study period from the 1994 low of 128,000 visits to a 2003 peak of 318,000 visits (p = 0.01). The corresponding ED visit rate per 10,000 U.S. population also increased from 4.9 visits (95%CI, 3.1–6.7) to 10.9 (95%CI, 7.6–14.3) (p = 0.01). The average age for patients making ED visits for acute pancreatitis during the study period was 49.7 years, 54% were male, and 27% were black. The ED visit rate was higher among blacks (14.7; 95%CI, 11.9–17.5) than whites (5.8; 95%CI, 5.0–6.6). At 42% of ED visits, patients did not receive analgesics. At 10% of ED visits patients underwent CT or MRI imaging, and at 13% of visits they underwent ultrasound testing. Two-thirds of ED visits resulted in hospitalization. Risk factors for hospitalization were older age (multivariate odds ratio for each increasing decade 1.5; 95%CI, 1.3–1.8) and white race (multivariate odds ratio 2.3; 95%CI, 1.2–4.6).

Conclusion

ED visits for acute pancreatitis are rising in the U.S., and ED visit rates are higher among blacks than whites. At many visits analgesics are not administered, and diagnostic imaging is rare. There was greater likelihood of admission among whites than blacks. The observed race disparities in ED visit and admission rates merit further study.     

Propofol and remifentanil versus midazolam and fentanyl for sedation during therapeutic hypothermia after cardiac arrest: a randomised trial

Purpose

To compare two protocols for sedation and analgesia during therapeutic hypothermia: midazolam and fentanyl versus propofol and remifentanil. The primary outcome was the time from discontinuation of infusions to extubation or decision not to extubate (offset time). Secondary outcomes were blood pressure, heart rate, use of vasopressors and inotropic drugs, pneumonia and neurological outcome.

Methods

This was an open, randomised, controlled trial on 59 patients treated with therapeutic hypothermia (33–34?°C for 24?h) after cardiac arrest in two Norwegian university hospitals between April 2008 and May 2009. The intervention was random allocation to sedation and analgesia with propofol/remifentanil or midazolam/fentanyl.

Results

Twenty-nine patients received propofol and remifentanil, and 30 midazolam and fentanyl. Baseline characteristics were similar. Sedation and analgesia were stopped in 35 patients, and extubation was performed in 17?of these. Sedation had to be continued for 24 patients. Time to offset was significantly lower in patients given propofol and remifentanil [mean (95?% confidence intervals) 13.2 (2.3–24) vs. 36.8 (28.5–45.1)?h, respectively, p?p?=?0.003). Incidence of pneumonia and 3-month neurological outcome were similar in the two groups.

Conclusions

Time to offset was significantly shorter in patients treated with propofol and remifentanil. However, the clinical course in 40?% of patients prevented discontinuation of sedation and potential benefits from a faster recovery. The propofol and remifentanil group required norepinephrine twice as often, but both protocols were tolerated in most patients.     

Comparison between ATS/ERS age- and gender-adjusted criteria and GOLD criteria for the detection of irreversible airway obstruction in chronic heart failure

Background

Global Initiative for Chronic Obstructive Lung Disease (GOLD) criteria (FEV1/FVC <70%) are extensively used for diagnosis of chronic obstructive lung disease in heart failure (HF). The American Thoracic Society (ATS)/European Respiratory Society (ERS) recommends the use of age- and gender-specific lower limit of normal (LLN) for FEV1/FVC. We compared the impact of these definitions on apparent prevalence of airway obstruction in chronic HF.

Methods

Standardized pre- and post-bronchodilator spirometry was performed in HF patients. Airway obstruction was defined by ATS/ERS criteria as diagnostic standard. Additionally, airway obstruction was calculated using the GOLD criteria.

Results

Of the 89 participants who fulfilled the ATS criteria for acceptability and reproducibility, 24.7% met ATS/ERS and 43.8% GOLD criteria for airway obstruction (Chi-square p?=?0.007, McNemar <0.001). Sensitivity of GOLD criteria was 100%, specificity 74.6%, positive predictive value 56.4% and negative predictive value 100%. Among all individuals with an FEV1/FVC?>?LLN, 25.4% were falsely identified when using the GOLD criteria. A majority of false positives qualified for airway obstruction GOLD stage I (FEV1% ≥80%), which was significantly less often observed among true positives (76.5 vs. 31.8%; p?Conclusions In all HF patients with persistent dyspnoea despite optimal HF treatment, spirometric testing should be performed. Application of the GOLD criteria leads to overdiagnosis of irreversible airway obstruction in patients with HF, which may result in inappropriate medical therapy and health-care decisions.     

Candida spp. airway colonization could promote antibiotic-resistant bacteria selection in patients with suspected ventilator-associated pneumonia

Objective

Candida spp. airway colonization could promote development of ventilator-associated pneumonia (VAP) caused by Pseudomonas aeruginosa, a potentially multidrug-resistant (MDR) bacteria, and worsen the outcome of VAP regardless of pathogen. We therefore address the question of the risk of MDR bacteria isolation within the airway of patients with suspected VAP, whether Candida spp. is present or not.

Design and setting

Prospective observational study in a teaching hospital.

Patients and methods

Consecutive patients with suspected VAP were included. Respiratory tract secretions were seeded on specific medium for yeast isolation in addition to standard culture. Outcome as well as presence of MDR bacteria were assessed according to fungal colonization.

Results

323 suspected VAP were analysed. Among these, 181 (56?%) cases presented with Candida spp. airway colonization. Colonized and noncolonized patients were similar regarding baseline characteristics, prior exposure to antibiotics and VAP severity. However, mortality rate was greater in patients with fungal airway colonization than in those without (44.2 versus 31.0?%, respectively; p?=?0.02). In addition, MDR bacteria isolation was 31.5?% in patients with Candida spp. colonization versus 23.2?% in those without (p?=?0.13). Moreover, Candida spp. airway colonization was one independent risk factor for MDR bacteria isolation [odds ratio (OR)?=?1.79, 95?% confidence interval 1.05–3.05; p?=?0.03], in addition to the time elapsed between intensive care unit (ICU) admission and VAP suspicion.

Conclusions

In patients with suspected VAP, Candida spp. airway colonization is frequent and associated with increased risk for MDR bacteria isolation. This could worsen outcome and should therefore be considered when choosing an empiric antibiotic therapy.     

Safety of performing fiberoptic bronchoscopy in critically ill hypoxemic patients with acute respiratory failure

Background

The safety of fiberoptic bronchoscopy (FOB) in nonintubated critically ill patients with acute respiratory failure has not been extensively evaluated. We aimed to measure the incidence of intubation and the need to increase ventilatory support following FOB and to identify predictive factors for this event.

Methods

A prospective multicenter observational study was carried out in eight French adult intensive care units. The study included 169 FOB performed in patients with a PaO₂/FiO₂ ratio ??300. The main end-point was intubation rate. The secondary end-point was rate of increased ventilatory support defined as an increase in oxygen requirement >50?%, the need to start noninvasive positive pressure ventilation (NI-PPV) or increase NI-PPV support.

Results

Within 24?h, an increase in ventilatory support was required following 59 bronchoscopies (35?%), of which 25 (15?%) led to endotracheal intubation. The existence of chronic obstructive pulmonary disease (COPD; OR 5.2, 95?% CI 1.6?C17.8; p?=?0.007) or immunosuppression (OR 5.4, 95?% CI 1.7?C17.2; p?=?0.004] were significantly associated with the need for intubation in the multivariable analysis. None of the baseline physiological parameters including the PaO₂/FiO₂ ratio was associated with intubation.

Conclusions

Bronchoscopy is often followed by an increase in ventilatory support in hypoxemic critically ill patients, but less frequently by the need for intubation. COPD and immunosuppression are associated with the need for invasive ventilation in the 24?h following bronchoscopy.     

Surrogate decision makers’ attitudes towards research decision making for critically ill patients

Purpose

To examine the attitudes and preferences of surrogate decision makers (SDMs) regarding their involvement in the consent to research process for ICU patients.

Methods

We presented 136 SDMs of critically ill patients in five ICUs with four hypothetical research scenarios: baseline interventional study of a placebo controlled RCT; study with higher risk of treatment complication; study comparing two accepted treatments; study with shorter enrolment window. For each we asked SDMs if they would want to be involved in the consent to research decision, and to rate the acceptability of their comfort with, and their sense of burden with their involvement. Participants were screened for symptoms of anxiety and depression using the Hospital Anxiety and Depression Scale.

Results

For the baseline scenario, most SDMs wished to be involved in research decision making (90?%; 95?% CI 84–95?%); responses varied little across study permutations. The majority considered their involvement to be acceptable (85?%; 95?% CI 77–90?%), whereas, a small minority rated it as being unacceptable (2?%; 95?% CI 1–6?%). Many were comfortable with being involved (50?%; 95?% CI 41–59?%), but the number decreased when risk of harm was higher (34?%; 95?% CI 26–43?%) or enrolment window was shorter (41?%; 95?% CI 33–50?%). A majority (62?%) reported symptoms of anxiety and many (38?%) had symptoms of depression.

Conclusion

Most of the interviewed SDMs wished to be involved in research decision making for critically ill and incapable loved ones. Variability existed, however, in their desire to be involved when decisions were time-sensitive or perceived risk was greater.     

Meta-analysis of dropout rates in randomized controlled clinical trials

Background

The interpretation of opioid studies in patients with chronic pain due to osteoarthritis is limited by a high dropout rate. Therefore, the implication of dropouts on the recommendation of opioids in chronic osteoarthritis pain was analyzed.

Data sources

The databases of Medline, Embase, the Cochrane Library, and the Internet from 1990?C2009 were searched.

Study selection

Two independent authors included randomized controlled clinical trials investigating the effects of chronic opioid treatment for the management of osteoarthritis pain. In order to calculate the odds ratio, only placebo-controlled trials were included.

Data extraction

The primary outcome parameter was the dropout rate. Secondarily, the effect size was calculated. Data extraction was conducted by two independent authors.

Results

A total of 19 studies reporting results of 3,871 treatment and 2,080 placebo outcomes were retrieved. Compared to placebo, opioid treatment was associated with a significantly increased total dropout rate (OR=1.3, 95%CI 1.2?C1.4). Discontinuation of treatment was related to adverse events (OR=4.0, 95%CI 3.4?C4.6). Lack of analgesia was associated with a significantly reduced dropout rate in opioid groups (OR=0.4, 95%CI 0.3?C0.5). Analgesic effects were significantly better in opioid-treated patients (p=0.01).

Conclusion

In spite of analgesic effects, many osteoarthritis patients prefer to stop chronic opioid use, because of adverse events. Therefore, opioids are not generally recommended in osteoarthritis.     

Impact of fluid balance on outcome of adult patients treated with extracorporeal membrane oxygenation

Purpose

To assess the relationship between early daily fluid balance (FB) and 90-day outcome in adult patients treated with extracorporeal membrane oxygenation (ECMO).

Design

Retrospective observational study.

Setting

Tertiary referral centre for ECMO.

Patients

115 patients treated with ECMO for refractory heart failure and 57 patients treated with ECMO for refractory respiratory failure.

Methods

We analysed the association between early daily FB versus hospital and 90-day mortality using multivariable logistic regression model, Cox proportional-hazards model and propensity score.

Results

We obtained detailed demographic, clinical, and biochemical data, daily FB, and continuous renal replacement days. Fifty-seven per cent of patients had acute kidney injury (AKI) at ECMO initiation, and 60 % (n = 103) of patients received continuous renal replacement therapy (CRRT) during ECMO course, beginning at a median of 1 (0–3.5) days after ECMO initiation. Overall 90-day mortality was 24 %. Survivors exhibited lower daily FB from day 3 to day 5. After adjustments, Acute Physiology and Chronic Health Evaluation (APACHE) III, CRRT during the first 3 days, major bleeding event at day 1 and positive FB on day 3 were independent predictors of 90-day mortality. Positive FB at ECMO day 3 remained an independent predictor of hospital and 90-day mortality, regardless of the statistical model used or the inclusion of a propensity score to have positive FB.

Conclusions

Positive FB at ECMO day 3 is an independent predictor of 90-day mortality. Further interventional studies aimed at testing the value of strategy of tight control of FB during the early ECMO period are now warranted.     

Risk factors for hypoglycaemia in neurocritical care patients

Purpose

To identify risk factors for hypoglycaemia in neurocritical care patients receiving intensive insulin therapy (IIT).

Methods

We performed a nested case–control study. All first episodes of hypoglycaemia (glucose <80?mg/dL, <4.4?mmol/L) in neurocritical care patients between 1 March 2006 and 31 December 2007 were identified. Patients were treated according to the local IIT protocol, with target blood glucose levels between 4.5 and 6.0?mmol/L (81.0–108.0?mg/dL). The first hypoglycaemic event of every patient (index moment) was used to match to a control patient. Possible risk factors preceding the index moment were scored using hospital records and analysed with conditional logistic regression.

Results

Of 786 neurocritical care patients, 449 developed hypoglycaemia (57.1?%). Independent risk factors for hypoglycaemia were lowering nutrition 6?h before the index moment without insulin dose reduction (odds ratio (OR) 5.25, 95?% confidence interval (95?% CI) 1.32–20.88), mechanical ventilation (OR 2.59, 95?% CI 1.56–4.29), lowering the dosage of norepinephrine 3?h before the index moment (OR 2.44, 95?% CI 1.07–5.55), a hyperglycaemic event (>10?mmol/L,?>180.0?mg/dL) in the 24?h preceding the index moment (OR 2.40, 95?% CI 1.26–4.58), gastric residual in the 6?h preceding the index moment without insulin dose reduction (OR 1.76, 95?% CI 1.05–2.96) and dosage of insulin at the index moment (OR 0.83, 95?% CI 0.76–0.90).

Conclusion

Hypoglycaemia occurs in a considerable proportion of neurocritical care patients. We recommend the identification of these risk factors in these patients to avoid the occurrence of hypoglycaemia.     

Contrast-enhanced ultrasonography with Sonazoid? for the evaluation of bowel ischemia

Purpose

To determine the usefulness of contrast-enhanced ultrasonography with the contrast agent Sonazoid? for the detection of bowel ischemia.

Methods

From March 2007 to February 2009, 65 patients (35 men and 30 women, mean age 70.4?±?16.1?years) were enrolled. Fifty-three patients complained of acute abdominal pain with small bowel dilatation (n?=?40) or reduced bowel peristalsis (n?=?13). Twelve patients were clinically suspected of having bowel ischemia. After Sonazoid? injection, bowel segments were scanned using harmonic imaging, and the signal intensities were classified as normal or diminished. The definitive diagnosis was confirmed by surgery in 30 patients, autopsy in 6, endoscopy in 3, angiography in 1, and clinical follow-up in 25.

Results

All 50 patients with normal signal intensities were confirmed not to have bowel ischemia. In the 15 patients with diminished signal intensities, 14 patients were confirmed to have bowel ischemia, resulting in an overall sensitivity of 100% [95% confidence interval (CI) 80.7–100%], a specificity of 98% [95% CI 89.5–99.9%], a positive predictive value of 93% (95% CI 68.1–99.8%), and a negative predictive value of 100% (95% CI 94.1–100%).

Conclusion

Contrast-enhanced ultrasonography with Sonazoid? is a highly sensitive and specific method for the diagnosis of bowel ischemia.     

Prevalence and factors associated with polypharmacy in older people with cancer

Purpose

Polypharmacy has been associated with drug–drug interactions, adverse drug events, hospitalisation and increased mortality. The purpose of this study was to investigate the prevalence and factors associated with polypharmacy in older people with cancer.

Patients and methods

Patients aged ≥70 years (n?=?385) presenting to the medical oncology outpatient clinic at Royal Adelaide Hospital between January 2009 and July 2010 completed a structured data collection instrument. The instrument included domains related to medications, diagnoses, instrumental activities of daily living (IADLs), Karnofsky Performance Scale (KPS), physical function (SF-36), pain (ten-point visual analogue scale, VAS), weight loss (patient self-reported over previous 6 months), exhaustion (CES-D) and distress (ten-point VAS). Frailty was computed using Fried’s frailty phenotype. Logistic regression was used to compute unadjusted and adjusted odds ratios (ORs) and 95% confidence intervals (CIs) for the association between polypharmacy (defined as five or more self-reported daily medications) and clinical parameters.

Results

Polypharmacy was present in 57 % (n?=?221) of patients. When adjusting for age, gender and Charlson Comorbidity Index (CCI), polypharmacy was associated with being pre-frail (OR?=?2.35, 95%CI?=?1.43–3.86) and frail (OR?=?4.48, 95%CI?=?1.90–10.54) compared to being robust. When adjusting for age, gender, exhaustion, KPS, IADLs, pain and distress, polypharmacy was associated with higher CCI scores (OR?=?1.58, 95%CI?=?1.29–1.94) and poorer physical function (OR?=?1.13, 95%CI?=?1.06–1.20).

Conclusions

Polypharmacy is highly prevalent in older people with cancer and associated with impaired physical function and being pre-frail and frail compared to being robust. Research is needed to identify strategies to minimize patients’ medication regimens.