An audit of perioperative cardiac arrests in a Southeast Asian university teaching hospital over 15 years.
An audit of the incidence, causes and outcome of perioperative cardiac arrest was conducted in a university hospital in Pakistan. All perioperative cardiac arrests from induction of anaesthesia to post anaesthesia care unit discharge or intensive care unit admission during noncardiac surgery, from January 1992 to December 2006 were included. Patients' demographic information, physical status and type of surgery and anaesthesia were noted. Outcome variables were noted as immediate survival and survival to discharge. Anaesthesia-related cardiac arrests were identified and their causes analysed. Forty-two cardiac arrests occurred among 140,384 patients. Overall frequency was 2.99 per 10,000 (95% confidence interval: 2.90 to 3.08). Twenty-four (3.77/10,000) were females. Thirty-four (13.59/10,000) patients were ASA physical status III to V, 10 (4.95/10,000) were children and 14 (4.28/10,000) above 60 years. Sixteen patients (6.48/10,000) were undergoing emergency surgery. Anaesthesia was deemed primarily responsible in nine cases (0.64/10,000). The causes of anaesthesia-related arrests were medication related (4), airway related (3), massive air embolism (1) and under-replacement of fluids (1). The event was considered to be avoidable in 26 cases. Seventeen patients died during the arrest, 15 survived more than one hour and 10 were discharged home. The number of perioperative cardiac arrests and their mortality was higher in patients with poor physical status and in emergency surgery. The number was also higher in infants, patients above 60 and females. The majority of the cases were considered avoidable, indicating the importance of prevention strategies.
Key Words: perioperative: morbidity, cardiac arrest
Analysis of anaesthesia-related complications is important to identify the preventable causes and to make prevention strategies. Cardiac arrest during the perioperative period has been used as an indicator of quality of anaesthetic care (1). Studies performed in various parts of the world have examined perioperative cardiac arrest (2-7), with the main aim of improving quality. Most of these studies are not directly comparable due to differences in methodology, including the surgical population, type of anaesthesia, definition of perioperative period etc. Despite these differences in methodology, sharing such data from different centres and from different parts of the world may be informative.
To the best of the authors' knowledge, no published data on incidence and causes of perioperative cardiac arrest is available from Pakistan and other parts of Southeast Asia. There has been only one report on anaesthetic contributions to perioperative deaths within 24 hours of anaesthesia (8). In this article we report an audit of all cardiac arrests that occurred during non-cardiac surgery from the time of induction of anaesthesia to post anaesthesia care unit (PACU) discharge or intensive care unit (ICU) admission, in a tertiary care university hospital of a developing Southeast Asian country over a 15-year period.
MATERIALS AND METHODS
After obtaining approval from our institutional ethical review committee, we analysed all reported perioperative cardiac arrests occurring in patients undergoing noncardiac surgery at the Aga Khan University Hospital from January 1, 1992 to December 31, 2006. All anaesthesia-related morbidity and mortality is routinely discussed in our departmental meetings held once a month. Cases were identified from the records of these meetings. A comprehensive form was developed and was filled in after going through the departmental files and patient's medical records.
The Aga Khan University Hospital is a 500-bed teaching hospital which caters for patients in the southern part of the country. The anaesthesia department provides an average of 10,000 anaesthetics per year for paediatric, vascular, orthopaedic, general, neurosurgery, ear-nose-throat surgery, urology, gynaecology and plastic surgery. Anaesthesia services are also provided in the day care and obstetrics units. All operating rooms (OR) are equipped with the American Society of Anesthesiologists (ASA) recommended standards of monitoring. There is a 13-bed PACU adjacent to the main OR, while day care and obstetrics units have their own PACUs. Anaesthesia services are also provided for electroconvulsive therapy and radiology procedures and data from these patients were included.
All anaesthesia-related morbidity is reported to two assigned anaesthesia consultants who also obtain feedback from other sources, e.g. recovery room staff, anaesthesia residents and technicians, to ensure that all events are reported. A form is filled out including patient's demographic information, preoperative condition, ASA physical status, seniority level of anaesthetist, type of surgery and sequence of events. This system has been in place since 1992. All these cases are discussed in depth at a monthly morbidity and mortality meeting. Two independent consultants not involved in the management of the patient give their written opinion regarding the possible cause of arrest, role of anaesthesia in the morbidity and whether it was avoidable. For the purpose of the audit, cardiac arrest was defined as an event that required cardiopulmonary resuscitation which included closed chest cardiac compressions. The perioperative period was defined as the time starting from the induction of anaesthesia to PACU discharge or ICU admission. This time period was specified because we planned to specifically analyse the arrests during the period when the patient is under the direct care of the anaesthesia team. Cardiac arrests that occurred during cardiac surgery were not included in this audit as cardiac surgical services started at our hospital in 1998 with one surgeon and have expanded gradually over the years.
Patient-related factors recorded were age, gender and preoperative ASA physical status. Variables related to surgery included surgical specialty and the type of surgery (open, laparoscopic, closed). Anaesthesia-related factors included the type of anaesthesia, monitoring used and senior-most anaesthetist present at the time of arrest. The main outcome variable looked at was survival to discharge from the hospital, while survival for more than one hour after initial resuscitation was also noted. We also compared the data between two periods, 1992 to 1999 and 2000 to 2006 to look for any changes in the trends of perioperative cardiac arrest, as expansion of anaesthetic services had taken place in the year 2000 with an additional operating room opening up both in the main OR suite and the day care unit and routine provision of monitored anaesthesia care outside the OR becoming functional. Most of the information could be retrieved from the departmental "morbidity and mortality" file. In addition, patients' medical records were reviewed to fill out any missing data.
This study was part of the ongoing quality assurance activity of our department. The overall incidence of cardiac arrest was calculated per 10,000 cases with 95% confidence intervals (CI). Incidence related to demographic and clinical variables was calculated in a similar way. Outcome in terms of death and survival to discharge was calculated as actual numbers for the different demographic variables.
Over the 15 years of the audit, 140,384 patients received anaesthetic care for various forms of noncardiac surgery at our university hospital. A total of 42 cardiac arrests occurred during the period from start of induction to PACU discharge or ICU admission. The overall incidence was 2.99 per 10,000 cases (95% CI: 2.90 to 3.08). Seventeen (40.5%) patients died during the arrest, 15 (35.7%) survived for more than one hour but died later in the hospital and 10 (23.8%) were discharged home. Incidence in terms of age, gender, urgency of surgery, ASA status and type of anaesthesia is given in Table 1, along with the outcome in terms of death and survival to discharge. There was no survival to discharge in patients falling into ASA physical status classification of IV and V (Table 2).
Thirty-four patients were undergoing open surgery, two were undergoing laparoscopic surgery and three were having endoscopic procedures. The remaining three were having closed procedures such as magnetic resonance imaging or electroconvulsive therapy. Twenty-five arrests occurred during the routine working hours of 0800 to 1700 hours, 12 occurred between 1700 hours and midnight and five between midnight and 0800 hours. Among those who arrested during routine working hours, eight survived to hospital discharge, whereas only two out of the 17 patients who suffered cardiac arrest after the routine working hours survived to discharge. Twenty-two patients were being monitored non-invasively (electrocardiogram, peripheral oxygen saturation, non-invasive blood pressure and end-tidal carbon dioxide) and invasive monitoring was being used in 20 patients (intra-arterial and central venous pressures). A record of core temperature was available in eight patients, out of whom four were hypothermic (<35[degrees]C), three were normothermic (36 to 37[degrees]C) and one had a temperature of 39[degrees]C.
Twenty cases of arrest occurred between 1992 and 1999 (3.33/10,000), of which four survived to hospital discharge. During the years 2000 to 2006, 22 patients arrested (2.73/10,000), of whom six were discharged. Cardiac arrests were deemed primarily anaesthesia-related in nine cases (0.64/10,000). The patient's primary condition was considered the main cause in 16 patients (1.13/10000), haemorrhage in seven (0.5/10000) and surgical complication in two patients (0.14/10,000). Anaesthesia was considered contributory to the event in eight patients (0.56/10000). The cause of primarily anaesthesia-related arrests was medication-related in four cases, airway-related in three cases, massive air embolism in one and under-replacement of fluids in one (Table 3). Six of the nine patients with anaesthesia-related cardiac arrests were discharged home. There were no survivors in those who arrested after massive haemorrhage (0/7) or mainly surgical reasons (0/2). There was one survival to discharge in arrests due to the patient's primary condition and three of the eight patients in whom anaesthesia was judged to be a contributory factor survived to hospital discharge. The arrest was considered to be avoidable in 26 (61.9%) cases and unavoidable in 14 (33.3%), while in two cases the reviewers were not sure in this regard.
A consultant anaesthetist was present in 30 cases at the time of the arrest and a resident of different levels in 12 cases. Seven of the 30 patients where the consultant was present survived and three of the 12 survived when an anaesthesia resident was the anaesthesiologist present. Eight arrests occurred at induction of anaesthesia, 16 during maintenance, five at emergence, two during shifting to PACU and 10 during the PACU stay. The timing was not documented for one patient.
Anaesthesia routinely involves intensive supervision of patients' vital signs. Thus, intraoperative cardiac arrest is likely to be diagnosed much earlier than that occurring in other situations, permitting prompt initiation of management. It is reasonable to believe that survival after cardiac arrest would be improved with quick diagnosis and prompt management. Over the 15 years of our audit, 42 perioperative cardiac arrests occurred in 140,384 non-cardiac anaesthetic cases (2.99 / 10,000) in a university teaching hospital serving as a tertiary referral centre. An important implication of our results is that there is a small number of arrests over a long period of time, but when we consider the results of other studies done on perioperative cardiac arrests, the reported overall incidence is 2.36 to 23.09 / 10,000 anaesthetics2,3, which is comparable to our results. The incidence of primarily anaesthesia-related cardiac arrests and mortality, 0.64 / 10,000 and 0.21 per 10,000 cases respectively, is also comparable with other such studies, 0.50 to 2.10 / 10,000 anaesthetics for cardiac arrests (2,6,9-11) and 0.12 to 1.40 / 10,000 anaesthetics (2,6,9,12) for mortality. Some of the authors have included a much larger overall number (10,13), and the time spans studied range from three to 18 years (2,4,7,10,11).
The higher incidence of cardiac arrest seen in males in other studies (2,9) was not reproduced in ours where the incidence was higher in females (3.77/10,000 vs. 2.35/10,000). We do not have an effective primary and secondary health care system, which results in tertiary care hospitals dealing with poorly optimised sick patients. Added to this is our social system in which most women are dependent upon men to seek medical help and this often leads to their disease being well advanced at the time of presentation. As seen in other studies2,9, a better survival to discharge in females was also reproduced in our results (29.1% vs. 16.6% in males). Our audit demonstrated a higher incidence of cardiac arrests in patients of ASA physical status III to V and there was no survival to discharge in patients with ASA physical status classification of IV or V. This has been demonstrated in other studies (2,9,12), ASA physical status, along with emergency surgery, has been termed an important predictive factor of mortality after cardiac arrest (2,12). The highest number of cardiac arrests in the paediatric patients was observed in children under one year (6/10). This has been noted by other authors (2,4,6,9,11,12). A recent study by Flick and colleagues4 found that the incidence of cardiac arrest in children undergoing non-cardiac procedures was very high for neonates (39.4 / 10,000) and much lower for all other age groups. They have concluded that children older than one year do not seem to be at increased risk of perioperative cardiac arrest compared with adults. Advanced age has also been seen to increase the risk of cardiac arrest during anaesthesia2,6,9. In our results too, the highest incidence was found in patients over 60 years (4.28/10,000) compared to those between 40 to 60 (2.39/10,000) and 18 to 40 years (1.76/10,000).
Emergency surgery has been reported as a risk factor for anaesthesia-related cardiac arrest (2,9,12). This was also reproduced in our study (6.48/10,000 vs. 2.25/10,000 in elective surgery). The reason for this may well be a lack of appropriate preoperative optimisation in patients presenting for emergency surgeries. Similar to other studies2,12, survival was better in patients who underwent elective surgery. A large majority of patients (34/42) were having open surgical procedures at the time of arrest. The reason for this, most probably, is that laparoscopic surgeries, although performed routinely at our centre, are mostly limited to simple gynaecological procedures and laparoscopic cholecystectomies. Therefore most patients having these procedures fall into ASA physical status of I to III. The overall number of other types of procedures was too small to enable further analysis or comments.
The incidence of cardiac arrest was found to be slightly higher in neuraxial anaesthesia (3.73/10,000) compared to general anaesthesia (2.98/10,000). As the number of procedures performed under neuraxial anaesthesia were very small, it would not be appropriate to comment on this finding, but the higher incidence seems to be different when compared to other authors who have found a greater incidence of perioperative cardiac arrests in patients receiving general anaesthesia (2,3). The reason for this could be that many high risk orthopaedic, urological and lower abdominal surgeries are performed under neuraxial anaesthesia at our centre. Added to this is the ineffective primary and secondary health care system resulting in tertiary care hospitals dealing with poorly optimised sick patients presenting in advanced stages of disease requiring urgent surgery. Kopp et al (5) have observed that cardiac arrest during neuraxial anaesthesia is associated with an equal or better likelihood of survival than that during general anaesthesia. No arrests were observed during nerve plexus blocks, which has also been observed by other authors (2,6), the reason suggested being the low likelihood of major cardiovascular and respiratory changes during plexus block. Our finding of a better survival during routine working hours was also seen by Sprung and colleagues (3). Increased availability of personnel and a more comprehensive response could well be the reason for this. The overall incidence of perioperative cardiac arrest showed a slightly declining trend when the data were divided into two periods, 1992 to 1999 and 2000 to 2006. The monitoring and reporting of critical incidents has improved considerably in recent years at our institution and a decreasing trend despite better reporting is an encouraging sign, hopefully indicating improved perioperative management. Sprung and colleagues (3) found a declining incidence over time in a single centre study, although they were not able to identify the factors responsible for this trend. Olsson and Hallen (13) also reported a declining incidence over time.
Six out of the nine patients with primarily anaesthesia-related arrests survived to discharge. Sprung and colleagues (3) found a 79.2% hospital survival after cardiac arrests attributable primarily to anaesthesia, whereas Newland et al (9) reported 80% mortality after such arrests. The cause of anaesthesia-related arrests (Table 3) was medication-related in four cases and airway-related in three cases. Braz et al (2) reported that 44.4% of the arrests were medication-related and 55.6% were due to respiratory causes and Sprung and colleagues3 found that 21% of the arrests were caused by airway misadventures and 54% were attributed to medication. In our study, as also seen by Braz et al (2), the predominant mechanism in medication-related cardiac arrests was cardiovascular depression, whereas the events were mainly related to neuromuscular blocking drugs in Sprung and colleagues' study (3). They suggest that practice-specific factors may be important determinants of primarily anaesthesia attributable cardiac arrests. Surgical specialty and phase of anaesthesia at the time of arrest were documented during data collection, but the small number of cases precluded further analysis and comments. Twenty-six of the 42 arrests at our centre were considered to be preventable. It is well accepted that inadequate preoperative preparation is a prominent contributing factor to adverse outcome (6,14). Preventable factors associated with poor outcomes include faulty anaesthetic technique (14), inadequate monitoring, postoperative respiratory and circulatory failure (14) and inadequate postoperative care14. Olsson and Hallen (13) have concluded that inadequate preoperative assessment, failure in the application of existing knowledge and neglecting to consult more experienced colleagues are the most important avoidable causes of intraoperative cardiac arrests.
Before concluding, it is important to highlight some of the limitations of this audit. Although our department has collected data prospectively for all morbidities since 1992, some data were missing in the department files and had to be retrieved from patients' medical records. The completeness of the information obtained was limited by the amount of documentation available in old medical records. The small number of cases prevented the application of statistical tests for univariate and multivariate analyses. A very narrow window of time, from the induction of anaesthesia to PACU discharge or ICU admission, was considered in this audit. We specified this period because we wanted to analyse the arrests during the time when the patient is under the direct care of the anaesthesia team, i.e. the anaesthetist has immediate control on the situation. Many anaesthetic complications, including anaesthesia-related cardiac arrests and mortality, are likely to occur in the time immediately following this period.
Most authors who have reported on anaesthesia-related cardiac arrests and anaesthesia mortality have studied a much longer duration of time, usually up to 24 hours after the end of surgery or longer (2,4,6,8). Therefore our use of the term 'anaesthesia-related' cardiac arrests and mortality does not correlate with the usually accepted definition of anaesthesia-related complications and we cannot claim that our results portray the true incidence of anaesthesia-related cardiac arrests and mortality at our centre. Our results are also not directly comparable to most other studies because of this difference in methodology, although Sprung and colleagues (3) have defined a similar time period.
There are also limitations regarding the applicability of results to other settings. In a developing country like ours, there is limited availability of some anaesthetic and resuscitation drugs, while these and standard monitoring equipment are lacking in many centres. In contrast, ours is a private university hospital with adequate facilities and the ASA recommended standards for intraoperative monitoring are adhered to at all areas of anaesthesia provision. Therefore our results would not be directly comparable to similar studies done in most other centres of our country. Despite the fact that the coordinators of morbidity and mortality meetings are very vigilant in collecting information regarding all perioperative morbidities, there is a possibility that some events were missed, leading to an underestimation of the incidence. Nevertheless, we believe that adherence to methodology in a single centre is a strength of this study.
In summary, as expected, our incidence of perioperative cardiac arrests was higher in patients with poor physical status undergoing emergency surgery. It was also higher in infants, patients above 60 years of age and in females. Anaesthesia-related cardiac arrests were mainly due to medication and airway-related causes. The majority of the cases were considered avoidable, indicating the importance of prevention strategies.
The authors would like to thank Mr Iqbal Azam, Assistant Professor, Community Health Sciences, Aga Khan University for his help and cooperation in data entry and analysis.
Accepted for publication on June 5, 2008.
(1.) Boelle PY, Garnerin P, Clergue F, Sicard JF, Bonnet F. Voluntary reporting system in anaesthesia: is there a link between undesirable and critical events? Qual Health Care 2000; 9:203-209.
(2.) Braz LG, Modolo NS, do Nascimento P Jr, Bruschi BA, Castiglia YM, Ganem EM et al. Perioperative cardiac arrest: a study of 53,718 anaesthetics over 9 yr from a Brazilian teaching hospital. Br J Anaesth 2006; 96:569-575.
(3.) Sprung J, Warner ME, Contreras MG, Schroeder DR, Beighley CM, Wilson GA et al. Predictors of survival following cardiac arrest in patients undergoing noncardiac surgery. Anesthesiology 2003; 99:259-269.
(4.) Flick RP, Sprung J, Harrison TE, Gleich SJ, Schroeder DR, Hanson AC et al. Perioperative cardiac arrests in children between 1988 and 2005 at a tertiary referral centre. Anesthesiology 2007; 106:226-237.
(5.) Kopp SL, Horlocker TT, Warner ME, Hebl JR, Vachon CA, Schroeder DR et al. Cardiac arrest during neuraxial anesthesia: frequency and predisposing factors associated with survival. Anesth Analg 2005; 100:855-865.
(6.) Biboulet P, Aubas P, Dubourdieu J, Rubenovitch J, Capdevila X, D'Athis F. Fatal and non fatal cardiac arrests related to anaesthesia. Can J Anesth 2001; 48:326-332.
(7.) Braz LG, Braz JRC, Modolo NSP, Nascimento P, Shuhama AP, Navarro LHC. Cardiac arrest during anesthesia at a tertiary teaching hospital. Prospective survey from 1996 to 2002. Rev Bras de Anestesiol 2004; 54:755-768.
(8.) Khan M, Khan FA. Anesthetic deaths in a developing country. Middle East J Anesthesiol 2007; 19:159-172.
(9.) Cohen MM, Duncan PG, Pope WD, Wolkenstein C. A survey of 112,000 anaesthetics at one teaching hospital (1975-83). Can Anaesth Soc J 1986; 33:22-31.
(10.) Kawashima y, Takahashi S, Suzuki M, Morita K, Irita K, Iwao Y et al. Anaesthesia-related mortality and morbidity over a 5-year period in 2,363,038 patients in Japan. Acta Anaesthesiol Scand 2003; 47:809-817.
(11.) Aroonpruksakul N, Raksakiatisak M, Thapenthai y, Wangtawesaup K, Chaiwat O, Vacharaksa K et al. Perioperative cardiac arrests at Siriraj Hospital between 1999-2001. J Med Assoc Thai 2002; 85 (Suppl 3):S993-999.
(12.) Morray JP, Geiduschek JM, Ramamoorthy C, Haberkern CM, Hackel A, Caplan RA et al. Anesthesia-related cardiac arrest in children: initial findings of the Pediatric Perioperative Cardiac Arrest (POCA) Registry. Anesthesiology 2000; 93:6- 14.
(13.) Olsson GL, Hallen B. Cardiac arrest during anaesthesia. A computer- aided study in 250,543 anaesthetics. Acta Anaesthesiol Scand 1988; 32:653-664.
(14.) Lee A, Lum ME. Measuring anaesthetic outcomes. Anaesth Intensive Care 1996; 24:685-693.
A. AHMED*, M. ALI ([dagger]), F. A. KHAN ([double dagger]), M. U. KHAN ([section])
Department of Anaesthesia, Aga Khan University Hospital, Karachi, Pakistan
* M.B., B.S., F.F.A.C.R.S.I., Senior Instructor.
([dagger]) M.B., B.S., F.C.P.S. (Pakistan), Assistant Professor.
([double dagger]) M.B., B.S., F.R.C.A., Professor.
([section]) M.B., B.S., F.C.P.S. (Pakistan), Associate Professor.
Address for reprints: Dr Aliya Ahmed, Department of Anaesthesia, Aga Khan University, PO Box 3500, Stadium Road, Karachi 74800, Pakistan.
TABLE 1 Incidence of cardiac arrest and outcomes according to age, gender, urgency of surgery, ASA physical status and anaesthetic technique Cardiac arrests Cardiac arrests Anaesthetics (n) n Incidence (per 10,000) Age group 0-<18y 20,216 10 4.95 18-<40 y 45,499 8 1.76 40-<60 y 41,927 10 2.39 60 and above 32,742 14 4.28 Gender Male 76,648 18 2.35 Female 63,736 24 3.77 Urgency Elective 115,684 26 2.25 Emergency 24,700 16 6.48 ASA status I and II 115,373 8 0.69 III, IV, V 25,011 34 13.59 Anaesthesia type General 134,307 40 2.98 Regional 5355 2 3.73 MAC 682 0 0.00 Nerve blocks 40 0 0.00 Cardiac arrests Outcome Outcome 95% CI Died (n) Discharged (n) Age group 0-<18y 1.88-8.01 7 3 18-<40 y 0.54-2.98 7 1 40-<60 y 0.91-3.86 7 3 60 and above 2.04-6.52 11 3 Gender Male 1.26-3.43 15 3 Female 2.26-5.27 17 7 Urgency Elective 1.38-3.11 17 9 Emergency 3.30-9.65 15 1 ASA status I and II 0.21-1.17 5 3 III, IV, V 9.03-18.16 27 7 Anaesthesia type General 2.06-3.90 30 10 Regional -1.44-8.91 2 0 MAC 0 0 0 Nerve blocks 0 0 0 CI=confidence interval, MAC=monitored anaesthesia care. TABLE 2 Incidence of cardiac arrests according to each level of ASA physical status (number and percentage), their immediate survival and survival to discharge Cardiac Cardiac Cardiac arrests Cardiac arrests ASA status arrests arrests Immediate Survival to n % survival discharge I 1 2.4 1 1 II 7 16.7 2 2 III 19 45.2 12 7 IV 12 28.5 8 0 V 3 7.1 2 0 TABLE 3 Cardiac arrests attributed primarily to anaesthesia: cause, survival and preventability S# Age ASA status Event (probable cause of arrest) 1 33 y III Asystolic arrest after suxamethonium (medication-related) 2 2 days III Under-replacement of fluid during surgery of a large sacrococcygealteratoma 3 8 months II Inadequate ventilation after extubation following cleft plate repair (airway-related) 4 58 y III Respiratory depression after intravenous sedation during combined spinal-epidural (medication-related). 5 6 days II Bradycardic arrest after suxamethonium (medication-related) 6 58y III Severe laryngospasm during difficult intubation (airway-related) 7 55 y II Massive air embolism during rapid fluid administration 8 53 y III Severe bronchospasm at extubation (airway-related) 9 1 y I Two vaporisers were opened simultaneously during magnetic resonance imaging (medication-related) S# Survival to Preventable discharge 1 Yes Yes 2 Yes Yes 3 No Yes 4 No Yes 5 Yes Yes 6 Yes Yes 7 Yes Yes 8 No Yes 9 Yes Yes
|Printer friendly Cite/link Email Feedback|
|Author:||Ahmed, A.; Ali, M.; Khan, F.A.; Khan, M.U.|
|Publication:||Anaesthesia and Intensive Care|
|Article Type:||Clinical report|
|Date:||Sep 1, 2008|
|Previous Article:||Dreaming and recall during sedation for colonoscopy.|
|Next Article:||How well do serum sTREM-1 measurements prognosticate in septic shock?|