Phillip D. Levin

Use of ultrasound guidance in the insertion of radial artery catheters.

ObjectiveTo assess the role of a portable ultrasound device in the insertion of radial artery catheters. DesignProspective, randomized, comparative study. SettingTertiary university hospital. PatientsElective surgery patients requiring arterial catheter insertion for intraoperative monitoring. InterventionsA portable ultrasound device was used to visualize the radial artery at the wrist and to direct arterial catheter insertion. This new technique of arterial catheter insertion was compared with the classic palpation technique. Measurements and Main ResultsA total of 69 patients requiring an arterial catheter were randomized to either the ultrasound (34 patients) or palpation technique (35 patients). The time taken from skin puncture to successful arterial catheter insertion, the time taken per insertion attempt, the number of attempts required, and the number of cannulae used were recorded for each group. The arterial cannula was inserted on the first attempt in 21 (62%) cases using ultrasound vs. 12 (34%) cases by palpation (p = .03). Significantly fewer attempts were required for catheter insertion using ultrasound as compared with palpation (mean ± sd, 1.6 ± 1.0 vs. 3.1 ± 2.4;p = .003); however, the time taken for each successful attempt was longer (26.1 ± 2.0 vs. 17.3 ± 1.6 secs, p = .001). A trend toward shorter overall time required for catheter insertion was found for the ultrasound group (55.5 ± 63.8 vs. 111.5 ± 121.5 secs, p = .17). There were four failures in the ultrasound group and one in the palpation group (not significant). ConclusionsUltrasound is a useful adjunct to arterial catheter insertion and increases the rate of success at first attempt. The technique is easy to learn and may reduce the time taken to insert the catheter.

Troponin elevation in severe sepsis and septic shock: The role of left ventricular diastolic dysfunction and right ventricular dilatation

Objective:Serum troponin concentrations predict mortality in almost every clinical setting they have been examined, including sepsis. However, the causes for troponin elevations in sepsis are poorly understood. We hypothesized that detailed investigation of myocardial dysfunction by echocardiography can provide insight into the possible causes of troponin elevation and its association with mortality in sepsis. Design:Prospective, analytic cohort study. Setting:Tertiary academic institute. Patients:A cohort of ICU patients with severe sepsis or septic shock. Interventions:Advanced echocardiography using global strain, strain-rate imaging and 3D left and right ventricular volume analyses in addition to the standard echocardiography, and concomitant high-sensitivity troponin-T measurement in patients with severe sepsis or septic shock. Measurements and Main Results:Two hundred twenty-five echocardiograms and concomitant high-sensitivity troponin-T measurements were performed in a cohort of 106 patients within the first days of severe sepsis or septic shock (2.1 ± 1.4 measurements/patient). Combining echocardiographic and clinical variables, left ventricular diastolic dysfunction defined as increased mitral E-to-strain-rate e′-wave ratio, right ventricular dilatation (increased right ventricular end-systolic volume index), high Acute Physiology and Chronic Health Evaluation-II score, and low glomerular filtration rate best correlated with elevated log-transformed concomitant high-sensitivity troponin-T concentrations (mixed linear model: t = 3.8, 3.3, 2.8, and –2.1 and p = 0.001, 0.0002, 0.006, and 0.007, respectively). Left ventricular systolic dysfunction determined by reduced strain-rate s′-wave or low ejection fraction did not significantly correlate with log(concomitant high-sensitivity troponin-T). Forty-one patients (39%) died in-hospital. Right ventricular end-systolic volume index and left ventricular strain-rate e′-wave predicted in-hospital mortality, independent of Acute Physiology and Chronic Health Evaluation-II score (logistic regression: Wald = 8.4, 6.6, and 9.8 and p = 0.004, 0.010, and 0.001, respectively). Concomitant high-sensitivity troponin-T predicted mortality in univariate analysis (Wald = 8.4; p = 0.004), but not when combined with right ventricular end-systolic volume index and strain-rate e′-wave in the multivariate analysis (Wald = 2.3, 4.6, and 6.2 and p = 0.13, 0.032, and 0.012, respectively). Conclusions:Left ventricular diastolic dysfunction and right ventricular dilatation are the echocardiographic variables correlating best with concomitant high-sensitivity troponin-T concentrations. Left ventricular diastolic and right ventricular systolic dysfunction seem to explain the association of troponin with mortality in severe sepsis and septic shock.

Institutional Control Measures to Curtail the Epidemic Spread of Carbapenem-Resistant Klebsiella pneumoniae: A 4-Year Perspective

OBJECTIVE To describe the implementation of an institution-wide, multiple-step intervention to curtail the epidemic spread of carbapenem-resistant Klebsiella pneumoniae (CRKP). DESIGN Consecutive intervention analyses. PATIENTS AND SETTING All patients admitted to a 775-bed tertiary care medical center in Jerusalem, Israel, from 2006 through 2010. INTERVENTIONS The effects of 4 interventions were assessed: (1) a policy of isolation for patients colonized or infected with CRKP in single rooms, which was started in March 2006; (2) cohorting of CRKP patients with dedicated nursing staff and screening of patients neighboring a patient newly identified as a carrier of CRKP, which was started in March 2007; (3) weekly active surveillance of intensive care unit patients, which was started during August 2008; and (4) selective surveillance of patients admitted to the emergency department, which was started in March 2009. Interrupted regression analysis and change-point analysis were used to assess the effect of each intervention on the CRKP epidemic. RESULTS Patient isolation alone failed to control the spread of CRKP, with incidence increasing to a peak of 30 new cases per 1,000 hospital beds per month. Institution of patient cohorting led to a steep decline in the incidence of CRKP acquisition (P < .001). Introduction of active surveillance interventions was followed by a decrease in the incidence of CRKP-positive clinical cultures but an increase in the incidence of CRKP-positive screening cultures. The mean prevalence of CRKP positivity for the period after cohorting began showed a statistically significant change from the mean prevalence in the preceding period (P < .001). CONCLUSIONS The cohorting of patients with dedicated staff, combined with implementation of focused active surveillance, effectively terminated the epidemic spread of CRKP. Cohorting reduced cross-infection within the hospital, and active surveillance allowed for earlier detection of carrier status. Both interventions should be considered in attempts to contain a hospital epidemic.

Cultural differences at the end of life.

‘‘We have no right as physicians to perform any act under any circumstance that will shorten the life of patients under our care.” Among intensive care physicians reading this article, some will consider this statement to be heresy, some stupidity, some would conclude that the author is uninformed or has no understanding of economics or rationing, and some, maybe, would agree. Many would argue their point with considerable vehemence. Yet, all these physicians are working side by side in the global village and treating similar patients. And why are there such varied responses? Because each one of us is influenced by our cultural background and the environment in which we work. “Culture” is defined as “the distinctive customs, achievements, products, outlook, etc. of a society or group; the way of life of a society or group” (1). In this article, an attempt will be made to describe the effects of association with different cultural groups on decisions made at the end of life in the intensive care unit (ICU). Investigations into cultural variations in end-of-life attitudes focus on two main areas: the practice of end-of-life care per se (how frequently care is withheld or withdrawn, what type of patients would have such care withheld or withdrawn, etc.) and the process of making the decision (who should be involved in making the decisions, physician, nurse, family, or patient, for example). Data will be presented from the three study types used to examine these cultural differences: comparisons and/or descriptions of practice from different societies, responses from culturally identified individuals to theoretical scenarios and questionnaires, and broadly descriptive studies. Many studies from different countries have described the practice of withholding and withdrawing care from intensive care patients at the end of their lives. In the United States, the vast majority of patients dying within an ICU (2) do so after a decision has been made to limit care, the vast majority of physicians have withheld or withdrawn care (3, 4), and the incidence of decisions to withhold and withdraw care is increasing (2). Family or patient involvement in these decisions is almost universal (2). The Canadian experience is similar (5), with limitation of care occurring before 66– 80% of deaths in 1993. No data are provided concerning the decision-making process before withholding or withdrawal of care. In Spain (6), of 644 ICU deaths, only 221 (34%) were preceded by a decision to withhold or withdraw therapy, with treatment being withheld from 66% and withdrawn from 34%. In contrast to the United States, patients’ families were not involved in the decision-making process at all in 28% of cases, whereas in 41% there was direct agreement between the family and physicians, in 28% indirect agreement, and in 3% the family refused to accept the physicians recommendation and yet therapy was withheld or withdrawn. The decision to curtail therapy was initiated by physicians in 93% of cases and by the family in the remainder. Only 9% of patients having therapy withheld or withdrawn had expressed a prior opinion on the matter. In France (7, 8), 50 –53% of ICU deaths were preceded by a decision to limit care. Families were informed of the decision to limit care in 59%, but participated in making the decision on 17–44% of occasions. The patients’ wishes were rarely considered in either study. In Israel (9), care was withheld from 91% of patients who died in or after ICU admission, whereas treatment was withdrawn only from brain-dead patients. More descriptive studies have been published regarding the end-of-life experience in Hong Kong and Japan. Limitation of therapy has been reported to occur for 23–61% of Chinese patients dying in Hong Kong ICUs (10), with a high degree of involvement of the patient’s family (11), and the use of do-notresuscitate orders is rare (12). In Japan, decisions to withhold or withdraw care are described as rare (13, 14); family involvement in the end-of-life decisionmaking process may be to the exclusion of the patients themselves (15). Questionnaires inquiring about physicians’ practices and responses to scenarios present even more of a varied picture. A European ethics questionnaire (16) revealed that the use of written do-notresuscitate orders, discussions regarding end-of-life issues with families, and decisions to withhold, withdraw, or even speed the dying process varied greatly across 16 countries. In general, Southern European countries such as Greece, Portugal, and Italy demonstrated more conservative practices than Northern European countries. Variations were considerable; for example, the number of physicians writing do-not-resuscitate orders varied from 8% in Italy to 91% in The Netherlands, and 8% of Portuguese physicians would administer large doses of drugs to speed death in a hopeless case scenario as opposed to 71% of French physicians. More religious physicians tended to be more conservative than their less religious counterparts—38% of religious physicians felt that withdrawal of therapy or the administration of large drug doses were unacceptable as opposed to 21% of nonreligious physicians. A similar questionnaire addressed the issue of communication between physicians and their patients and also showed large European variations (17). Clearly, differences exist in the way that different cultures deal with end-ofFrom the Department of Anesthesiology and Critical Care Medicine, Hadassah Hebrew University Medical Center, Jerusalem, Israel. Copyright

Continuous surveillance to reduce extended-spectrum β-lactamase Klebsiella pneumoniae colonization in the neonatal intensive care unit.

Background: Clinical illness caused by resistant bacteria usually represents a wider problem of asymptomatic colonization. Active surveillance with appropriate institution of isolation precautions represents a potential mechanism to control colonization and reduce infection. The neonatal intensive care unit (NICU) is an environment particularly appropriate for such interventions. Neonates are rarely colonized by resistant bacteria on admission and staff enthusiasm for infection control is high. Objective: To reduce extended-spectrum β-lactamase-producing Klebsiella pneumoniae (ESBL-KP) acquisition amongst neonates through a continuous active surveillance intervention. Methods: Fecal ESBL-KP cultures were performed weekly on all neonates over 4 years. Neonates with positive cultures were managed with contact precautions by dedicated nurses separately from other neonates. ESBL-KP acquisition amongst neonates staying >7 days was compared for the consecutive years. A subset of ESBL-KP isolates was typed with pulsed-field gel electrophoresis (PFGE). Results: Surveillance cultures were obtained from 1,482/1,763 (84%) neonates over 4 years. ESBL-KP acquisition decreased continuously from 94/397 (24%) neonates in 2006 to 33/304 (11%) in 2009 (p < 0.001, hazard ratio 0.75, 95% confidence interval 0.66–0.85, p < 0.001 for comparison of years). Hospital-wide ESBL-KP acquisition did not decrease outside the NICU. PFGE identified identical ESBL-KP strains from multiple neonates on six occasions and different strains from single neonates on seven occasions. Conclusions: ESBL-KP is probably both imported into and spread within the NICU. Continuous long-term surveillance with cohorting was associated with a decrease in ESBL-KP acquisition within the NICU. This low-risk intervention should be considered as a means to decrease neonatal acquisition of resistant bacteria.

Routine surveillance blood cultures: Their place in the management of critically III patients

The use of surveillance blood cultures has been advocated as a means to allow earlier detection of septic episodes amongst intensive care patients, and therefore earlier institution of appropriate antibiotic therapy. We compared the results of surveillance cultures and clinically indicated blood cultures for bacterial isolates grown and the influence of culture results on patient management. Blood cultures were obtained from all intensive care unit (ICU) patients over the course of 3 months at a set surveillance time (surveillance group) or according to clinical indications (clinical group). Bacteriological results were compared and real-time chart review performed to assess the influence of the surveillance cultures on patient management, with particular reference to antibiotic therapy. Two hundred and forty-nine blood culture sets were collected over 3 months, 99 in the surveillance group and 150 in the clinical group. A total of 256 bacterial isolates were grown, 95 in the surveillance group and 161 in the clinical group. For the surveillance group 36%, 20%, and 44% of the isolates represented bacteraemia, line colonization and culture contamination, respectively. For the clinical group the distibution was 69%, 7%, and 24% respectively (P<0.001, P<0.01, and P<0.0027 for comparisons of percentages within each classification). On only one occasion was antibiotic therapy started based on the result of a surveillance culture, and on only one occasion was a septic episode detected earlier by a surveillance culture; however, this culture result did not lead to a change in patient management. Surveillance blood cultures are expensive and add very little to the management of patients in the intensive care environment.

The use of the arterial line as a source for blood cultures

Abstract Objective: To determine the reliability of blood cultures obtained through indwelling arterial lines as compared to that of blood cultures obtained by venipuncture. Design: A prospective observational study. Setting: Six-bed mixed medical surgical intensive care unit (ICU) of a 550-bed university-affiliated medical center. Measurements: During a 3-month period blood culture sets, when clinically indicated, were drawn in parallel from indwelling arterial catheters and one-time venipuncture and the results compared. Each blood sample consisted of 15 ml and was distributed equally between three blood culture bottles: aerobic, anaerobic and one aerobic resin-containing bottle. Blood culture results from the two sources were compared according to preset definitions. Main results: During the study period 90 parallel blood culture sets (540 bottles) were obtained from 36 patients. Forty-three (16%) venipuncture bottles were positive versus 88 (32%) arterial line culture bottles (p<0.001). Of the parallel sets, 83% yielded equivalent results – either both sterile or both growing the same organism. Amongst the discordant sets, the arterial line cultures grew 37 gram-positive and 18 gram-negative isolates not found in venipuncture sets (i.e. 50% of 109 arterial line isolates), while only two gram-positive isolates were solely grown in venipuncture cultures (4% of all 55 venipuncture isolates, p<0.001). On clinical correlation, all the gram-positive organisms in the discordant cultures were found not to reflect bacteremia, while five of the 18 gram-negative isolates (28%) grown only in arterial line cultures probably did reflect ongoing bacteremia. Conclusion: The results of blood cultures taken from the arterial line are frequently equivalent to those taken from venipuncture. When discordant, the growth of gram-positive bacteria almost certainly reflects contamination or arterial line colonization, whereas the growth of gram-negative bacteria may have to be considered as reflecting bacteremia.

Impact of the resin blood culture medium on the treatment of critically ill patients.

OBJECTIVE To assess the relevance, both clinical and bacteriologic, of the use of resin-containing blood culture media in blood cultures taken from critically ill patients receiving antibiotics. DESIGN A prospective, open clinical trial. SETTING The mixed medical surgical intensive care unit (ICU) of a 550-bed urban hospital. PATIENTS All ICU patients admitted during a 3-month period (n = 49) with suspected sepsis requiring blood cultures as part of their laboratory investigations. INTERVENTIONS The use of an aerobic resin-containing blood culture medium, in addition to the regular aerobic and anaerobic media for all blood cultures taken. MEASUREMENTS AND MAIN RESULTS Each blood culture result was classified as to its clinical significance. Changes in patient management were recorded. Culture sets in which the resin-containing bottle provided the information central to the change in patient management were identified. Bacteriologically, the results from the resin-containing medium were compared with the results from the aerobic and anaerobic media. Of 266 blood culture sets, 103 (39%) were positive, growing 278 bacterial and fungal isolates. Clinically, the resin-containing medium alone provided relevant data leading to changes in patient management on three occasions. On two of these occasions, cultures from the regular media provided the same data within 72 hrs. Bacteriologically, 77 (29%) aerobic bottles, 55 (21%) anaerobic bottles, and 89 (33%) resin-containing bottles were positive (statistical comparison of percentages: aerobic vs. resin-containing bottles, nonsignificant; aerobic vs anaerobic bottles, p < .046; anaerobic vs. resin-containing bottles, p < .0027). A similar proportion of pathogens was isolated from the resin-containing bottles only (9%) and aerobic bottles only (6%). A higher proportion of contaminants was isolated from the resin-containing bottles only than the aerobic bottles only in the various sets (17% vs. 7%, p < .046). The resin-containing bottle showed a trend toward increased detection of Staphylococcus aureus and Pseudomonas aeruginosa bacteremia. CONCLUSIONS The resin-containing medium offers little clinical benefit to the majority of ICU patients. Bacteriologically, it seems to have a similar overall sensitivity as the regular aerobic medium (with the possible exception of a higher sensitivity for the isolation of S. aureus and P. aeruginosa), but a lower specificity. The wide-spread use of the resin-containing bottle cannot be recommended.

Use of the Nonwire Central Line Hub to Reduce Blood Culture Contamination

BACKGROUND The sterile conditions used when inserting a central venous catheter (CVC) might be thought to decrease the contamination rate of blood cultures taken at CVC insertion; however, a previous retrospective study showed the opposite, that such blood cultures are contaminated more frequently than peripheral venipuncture blood cultures. The current study explored whether use of the CVC nonwire hub as a source of blood cultures decreased contamination while maintaining detection of true pathogens. METHODS A prospective, observational study was performed from June 2010 to May 2011 in the general ICU of an academic, tertiary referral center. The proportions of blood cultures taken from wire and nonwire CVC hubs growing contaminants and true pathogens were compared. Risk factors for blood culture contamination were identified, and multivariate analysis was used to identify independent predictors of blood culture contamination. RESULTS Among 313 blood cultures taken from 227 CVCs in 139 patients, 27 of 141 wire hub (19%) vs nine of 172 nonwire hub (5%) cultures were contaminated (P < .001). Only hub of blood culture origin was associated with contamination on multivariate analysis (OR, 4.3; 95% CI, 1.9-9.5; P < .001). True pathogens grew in 19 of 141 wire hub (13%) vs 27 of 172 nonwire hub (16%) cultures (P = .581). CONCLUSIONS A higher proportion of blood cultures taken from the CVC lumen exposed to the guidewire were contaminated when compared with nonwire hub cultures; detection of true pathogens was equivalent. To limit detrimental sequelae of blood culture contamination, blood cultures obtained at CVC insertion should be taken from the nonwire hub.

Multicentre validation of manufacturers' weight-based recommendations for laryngeal mask airway size choice in anaesthetic practice: A retrospective analysis of 20,893 cases.

BACKGROUND Manufacturers’ weight-based recommendations for laryngeal mask airway (LMA) sizes are not scientifically founded and have never been validated clinically. OBJECTIVE The validation of manufacturers’ weight-based recommendations in clinical practice and whether other patient variables such as sex and age may influence LMA size choice. DESIGN A retrospective study. SETTING Four Israeli hospitals (two tertiary and two secondary hospitals). METHODS Data on LMA use were retrieved from the Anaesthesia Information Management System (AIMS). Multinomial logistic regression analysis including weight, sex and age amongst adults, adolescents and children was used to identify predictors of LMA size used (development data set). Concordance between the size of LMA used and that predicted according to the model and the manufacturers’ recommendations was assessed (validation data set). RESULTS LMA insertions were analysed in 13 743 (65.8%) adults, 1807 (8.6%) adolescents and 5343 (25.6%) children. Suitability of manufacturers’ recommendations was higher in children [4075/5343 (76.3%)] than adults [5200/13743 (37.8%)] or adolescents [885/1807 (49.0%); P < 0.001 for children vs. adults or adolescents]. Sex influenced LMA size in adults and adolescents, but not in children. Age had no influence. Amongst adults and adolescents, LMA size was better predicted by the regression model than by manufacturers’ recommendations [adults: anaesthetists/model 61.7%, 95% confidence interval (95% CI) 60.1 to 63.2, anaesthetists/manufacturer 38.4%, 95% CI 36.0 to 39.0, P < 0.001; adolescents: anaesthetists/model 68.1%, 95% CI 64.1 to 72.0, anaesthetists/manufacturer 46.9%, 95% CI 42.7 to 51.1, P < 0.001]. In children, the regression model and manufacturers’ weight-based recommendations performed equally (anaesthetists/model 78.0%, 95% CI 75.9 to 80.0, anaesthetists/manufacturer 75.7%, 95% CI 73.6 to 7.8, P = 0.126). CONCLUSION Manufacturers’ weight-based recommendations for LMA size reflect clinical practice in children, but not in adults and adolescents. In these age groups, choice of LMA size is influenced by sex and weight, although with wider ranges than those recommended by manufacturers.