Bleeding in cardiac surgery is still one of the major concerns for cardiac surgeons and anaesthesiologists. The present review addresses the main pathophysiological mechanisms underlying bleeding after cardiac operations, and analyzes possible therapeutic strategies.

Overview of different bleeding mechanisms and related diagnostic and therapeutic approaches, with specific respect to disseminated intravascular coagulopathy.

Biological bleeding after cardiac operations depends on the interaction of platelet activity, the procoagulant system, and the fibrinolytic system. A diagnostic and therapeutic approach based on a four-phase model of disseminated intravascular coagulation and haemostasis imbalance is presented. At each step, adequate diagnostic tools (thromboelastography and antithrombin activity) are required to make a diagnosis and guide the appropriate treatment.

Bleeding is not a fatality. Quite the contrary, it must be treated through a multi-system approach, and as early as possible, preferably at a biological phase before clinical bleeding. In addition, early monitoring and treatment adapted to each patient makes it possible to better follow a patient’s evolution and helps reduce the occurrence of later thromboembolic events.

The importance of monitoring the breathing pattern during sedation of children undergoing magnetic resonance scans is indicated in guidelines, but no appropriate magnetic resonance-compatible devices are available. We report preliminary findings from a technique referred to as remote pressure sensor respiratory plethysmography.

A data acquisition system was developed, enabling measurement of respiratory rate, plethysmogram amplitude, proportion of inspiratory time over cycle time, thoraco–abdominal phase shift and sigh rate. Correlation between plethysmogram amplitude and tidal volume was investigated on adult volunteers. Twenty-seven children undergoing sedation were monitored with remote pressure sensor respiratory plethysmography, in addition to SPO2 and PetCO2. Differences in monitoring parameters were searched for among three groups: patients who received chloral hydrate only (chloral succeeded, CS group), those who received a supplementation of sodium thiopental (chloral failed, CF group), and those who were sedated with sodium thiopental directly (no chloral, NC group). Correlations were searched for among monitoring parameters, and with total dose of thiopental. The long-term behaviour of respiratory rate, proportion of inspiratory time over cycle time and phase shift was studied.

Plethysmogram amplitude was found to correlate linearly with tidal volume (r > 0.92), with a slope varying up to 22%. While 11% of patients did not tolerate the capnometric probe and readings were discontinuous in 26%, all of them tolerated remote pressure sensor respiratory plethysmography belts. Sighs and non-respiratory movements of the torso could be distinguished on remote pressure sensor respiratory plethysmography waveforms. No significant inter-group differences were found in PetCO2, SPO2, respiratory rate and phase shift. Proportion of inspiratory time over cycle time was higher in the NC group when compared to the CS group (0.497 ± 0.03 vs. 0.463 ± 0.008; P = 0.02), the CF group being characterized by intermediate values (0.480 ± 0.008); when compared to the CS group, sigh rate was lower in the CF group (0.04 ± 0.04 vs. 0.14 ± 0.08; P = 0.04) and in the NC group (0.06 ± 0.05 vs. 0.14 ± 0.08, P = 0.03). A positive correlation was found between total dose of thiopental and proportion of inspiratory time over cycle time, with r = 0.4 and P = 0.04. A large baseline variability in phase shift was found. No long-term trends predictive of patient movement could be identified.

Breathing pattern monitoring is feasible through pneumatic devices, which are well tolerated. The resulting correlation with changes in tidal volume can be better when compared to visual inspection. Proportion of inspiratory time over cycle time and sigh rate convey information related to the state of the sedated patient. These results are not specific to the technology employed, and large-scale studies on the clinical usefulness of breathing pattern monitoring are motivated.

To evaluate the effects of total intravenous anaesthesia vs. volatile anaesthesia on cardiac troponin release in coronary artery bypass grafting with cardiopulmonary bypass, we performed a multicentre randomized controlled study to compare postoperative cardiac troponin release in patients receiving two different anaesthesia plans.

We randomly assigned 75 patients to propofol (intravenous anaesthetic) and 75 patients to desflurane (volatile anaesthetic) in addition to an opiate-based anaesthesia for coronary artery bypass grafting. Peak postoperative troponin I release was measured as a marker of myocardial necrosis.

There was a significant (P < 0.001) difference in the postoperative median (25th–75th percentiles) peak of troponin I in patients receiving propofol 5,5 (2,3–9,5) ng dL−1 when compared to patients receiving desflurane 2,5 (1,1–5,3) ng dL−1. The median (interquartile) troponin I area under the curve analysis confirmed the results: 68 (30.5–104.8) vs. 36.3 (17.9–86.6) h ng dL−1 (P = 0.002). Patients receiving volatile anaesthetics had reduced need for postoperative inotropic support (24/75, 32.0% vs. 31/75, 41.3%, P = 0.04), and tends toward a reduction in number of Q-wave myocardial infarction, time on mechanical ventilation, intensive care unit and overall hospital stay.

Myocardial damage measured by cardiac troponin release could be reduced by volatile anaesthetics in coronary artery bypass surgery.

The death of patients treated by ventricular assist device is usually related to multiorgan failure for which a disorder of splanchnic circulation is blamed. Gastric tonometry (measurement of gastric intra-mucosal pressure of CO2) has already been studied in many fields and especially in cardiac surgery. The aim of this study was to investigate the prognostic value of gastric tonometry monitoring after implantation of a ventricular assist device.

In this prospective study, all consecutive patients scheduled for a ventricular assist device were included. Gastric tonometry was added to standard monitoring. Data were collected (lactate, gastric CO2 (PgCO2) during cardiopulmonary bypass, at admission to ICU, 24 and 48 h later and when norepinephrine was stopped. Preoperative biologic and haemodynamic data were also collected. The primary endpoint was death.

Fifty-six patients (50 men and 6 women) were included. In 91% of the cases, the mechanical assistance was biventricular. The objective of the assistance was a bridge to transplant in 93% (n = 27). Twenty-seven deaths (48%) occurred during the study, 59% (n = 16) of them took place before the cardiac transplantation (mean time = 18 ± 16 days after assist device insertion). Many factors were found to be associated with death: weight (P = 0.018), red cells administration (P = 0.025), length of surgery (P = 0.016), PgCO2 on admission to ICU (P = 0.040) and norepinephrine dose at 24 h.

Gastric tonometry has a prognostic value in the early postoperative hours after the implantation of a ventricular assist device.

Insufficient blood flow and oxygenation in the intestinal tract is associated with increased incidence of postoperative complications after bowel surgery. High fluid volume administration may prevent occult regional hypoperfusion and intestinal tissue hypoxia. We tested the hypothesis that high intraoperative fluid volume administration increases intestinal wall tissue oxygen pressure during laparotomy.

In all, 27 pigs were anaesthetized, ventilated and randomly assigned to one of the three treatment groups (n = 9 in each) receiving low (3 mL kg−1 h−1), medium (7 mL kg−1 h−1) or high (20 mL kg−1 h−1) fluid volume treatment with lactated Ringer’s solution. All animals received 30% and 100% inspired oxygen in random order. Cardiac index was measured with thermodilution and tissue oxygen pressure with a micro-oximetry system in the jejunum and colon wall and subcutaneous tissue.

Groups receiving low and medium fluid volume treatment had similar systemic haemodynamics. The high fluid volume group had significantly higher mean arterial pressure, cardiac index and subcutaneous tissue oxygenation. Tissue oxygen pressures in the jejunum and colon were comparable in all three groups.

The three different fluid volume regimens tested did not affect tissue oxygen pressure in the jejunum and colon, suggesting efficient autoregulation of intestinal blood flow in healthy subjects undergoing uncomplicated abdominal surgery.

Although a considerable amount of promising experimental research has been performed on cardiopulmonary resuscitation, clinical data indicate an ongoing limited outcome in human beings. One reason for this discrepancy could be that experimental studies use healthy animals whereas most human beings undergoing cardiopulmonary resuscitation suffer from acute or chronic myocardial dysfunction. To overcome this problem, we sought to develop a new model of myocardial infarction, that is easy to perform in all kind of laboratories and compromises on the myocardial function significantly.

Following approval by the local authorities, 14 domestic pigs were instrumented for measurement of arterial, central venous, left atrial and left ventricular pressures. Myocardial infarction was induced in eight pigs by clipping the circumflex artery close to its origin from the left coronary artery (infarction group; n = 8). Six animals (no infarction group, n = 6) served as no-infarct controls. Following a 4-min period of cardiac arrest, internal cardiac massage was performed in these two groups, and haemodynamics were recorded during the first 30 min of reperfusion.

All animals were resuscitated successfully. Compared to the no-infarction group, the infarction group showed significantly decreased myocardial contractility, coronary perfusion pressure and cardiac index (30 min after restoration of spontaneous circulation: infarction group: 57 ± 7 and 89 ± 19 mL min−1 kg−1 in the no-infarction group; mean ± SD; P < 0.05) during reperfusion. Two animals from the infarction group (25%), but none of the animals in the no-infarction group, died during the reperfusion period.

These data demonstrate that clipping of the circumflex artery leads to a reduced myocardial performance after successful resuscitation, whereas the rate of restoration of spontaneous circulation is not reduced. Therefore, this set-up provides a reproducible model for future studies of post-resuscitation haemodynamics and treatment.

Tracheostomy is one of the most common procedures in intensive care units worldwide. In this study we aimed to compare three different tracheostomy techniques with respect to duration of procedure and complications.

One hundred and thirty patients requiring endotracheal intubation for more than 10 days due to acute respiratory distress syndrome, infections or cerebrovascular events were consecutively selected to undergo the percutaneous dilatational tracheostomy technique (PDT n = 44), the guide-wire dilating forceps technique group (GWDF n = 41) or the PercuTwist technique (n = 45). The time taken to perform the procedure (skin incision to successful placement of tracheostomy tube) and complications were recorded.

The operating times were found to be 9.9 ± 1.1, 6.2 ±1.4 and 5.4 ± 1.2 min in PDT, GWDF and PercuTwist groups, respectively. The duration of the procedure was significantly shorter in the PercuTwist group as compared to the percutaneous dilatational tracheostomy (P < 0.01) and guide-wire dilating forceps (P < 0.05) groups. During postoperative bronchoscopy, eight cases of longitudinal tracheal abrasion (four in the PDT group, two in the GWDF group and two in the PercuTwist group), two cases of posterior tracheal wall injury (one in PDT and one in GWDF) and one case of tracheal ring rupture in the PDT group were seen.

Percutaneous tracheostomy techniques have their own advantages and complications. PercuTwist, a new controlled rotating dilatation method, was associated with minimal complications, appears to be easy to perform and a practical alternative to percutaneous dilatational tracheostomy and guide-wire dilating forceps techniques.

The objective was to carry out a retrospective study of changes in regional cerebral oxygen saturation (rSO2) using the Somanetics Invos Cerebral Oximeter (SICO) in older patients undergoing prolonged major abdominal surgery. Since evidence is accumulating that detection and correction of falls in rSO2 may be associated with a reduced incidence of postoperative cognitive dysfunction, the study assessed the incidence and possible predisposing factors for significant falls in rSO2 and strategies for correction.

Data from 46 consecutive patients aged 55 yr or above undergoing major abdominal surgery were collected and studied. A SICO electrode was placed over the right forehead prior to commencement of anaesthesia and values of rSO2 were recorded automatically on a floppy disk at 20 s intervals throughout the procedure and until transfer of the patient to the postanaesthesia care unit. Anaesthesia and physiological data were routinely collected by the author on an anaesthetic record computer and transferred to an Excel spreadsheet for analysis.

Average duration of the surgery exceeded 7 h. Average blood loss was 1363 mL (interquartile range 500–2000). In 11 of the 46 patients the rSO2 drop exceeded 20%, and in six, there was a significant temporal association of rSO2 drop with ongoing major haemorrhage. In 23 of the 46 patients, a maximum drop in rSO2 occurred, which was about 15% or more. Fall in rSO2 in the 46 patients significantly correlated with blood loss (P < 0.05) and percentage fall in haemoglobin (Hb) (P = 0.01) but not with lowest Hb. Despite maintenance of conventional haemodynamic parameters such as systolic blood pressure (BP) in most patients, the fall in rSO2 seemed only reversible by blood transfusion. In five of the six patients who experienced the greatest decline in rSO2 during haemorrhage, there was no correlation between fall in rSO2 and systolic BP.

This retrospective study confirms that a significant reduction in rSO2 is a common accompaniment to prolonged major abdominal surgery in elderly patients, especially if associated with blood loss, and is correctable by blood transfusion. In most cases, these changes would have gone unnoticed with conventional monitoring.

Inadvertent perioperative hypothermia causes serious morbidity in surgical patients. However, recent reports suggest that patients might still be hypothermic after elective surgery. We thus surveyed intraoperative temperature monitoring and management practices in Europe.

Postal survey of 801 representative hospitals from 17 European countries on the same day. The questions addressed the number of surgical procedures and type of anaesthesia performed, mode and site of temperature monitoring and method of patient warming. Mean and standard error of the mean or count and percentage were calculated. The t-test or contingency table analysis with the Fisher’s exact test were used.

Eight thousand and eighty-three surgical procedures were assessed from 316 responding hospitals (39.4%). Overall, patient temperature monitored in 19.4% and 38.5% of the patients were actively warmed. Under general anaesthesia, body temperature was monitored in 25% and during regional anaesthesia in 6%, P = 0.0005. Nasopharyngeal temperature was most often taken under general anaesthesia, while tympanic temperature was preferred during regional anaesthesia. Under general anaesthesia, 43% of patients were actively warmed as compared to 28% with regional anaesthesia, P = 0.0005. Forced-air warming was the method of choice for both general and regional anaesthesia.

Intraoperative temperature monitoring is still uncommon and hence active patient warming is not a standard of care in Europe. Awareness of perioperative hypothermia is critical to its prevention, and thus temperature monitoring is a pre-requisite. The objective is to maintain normothermia in patients throughout surgery. A European practice guideline for perioperative patient temperature management is warranted.

Several studies showed that single analgesic modality management can attenuate perioperative stress, but little is known about the effect of multimodal analgesia on catecholamine responses to surgical trauma in children.

Fifty children (American Society of Anesthesiologists Grade I or II) were randomly allocated to one of two groups: one received general anaesthesia and a caudal block (control group), and one group was given general anaesthesia, caudal block and intravenous (i.v.) fentanyl 2 μg kg−1 (fentanyl group). Plasma epinephrine and norepinephrine concentrations were measured three times during the perioperative period: at induction time (T0), at the end of surgery (T1) and when the children were fully awake in the postanaesthesia care unit (T2).

There was a significant reduction in the catecholamine levels in the two groups when (T1) and (T2) were compared with T0. When plasma epinephrine levels (at T0, T1 and T2) between the two groups were compared, a statistically significant reduction at T2 was obtained in the fentanyl group, when compared with the control group. However, plasma norepinephrine levels showed no statistically significant difference between the two groups (at T0, T1 and T2).

These findings suggest that the multimodal analgesic approach of adding i.v. low-dose fentanyl to a caudal block may decrease the plasma epinephrine release in children undergoing inguinal herniotomy.

The purpose of this study was to determine whether brain oxyhaemoglobin–deoxyhaemoglobin coupling was altered by anaesthesia or intubation-induced stress.

This was a prospective observational study in the operating room. Thirteen patients (ASA I and II) undergoing spinal or peripheral nerve procedures were recruited. They were stabilized before surgery with mask ventilation of 100% oxygen. Anaesthesia was induced with 2 μg kg−1 fentanyl and 3 mg kg−1 thiopental. Laryngoscopy and intubation were performed 4 min later. After intubation, desflurane anaesthesia (FiO2=1.0) was adjusted to maintain response entropy of the electroencephalogram at 40–45 for 20 min. Prefrontal cortex oxyhaemoglobin and deoxyhaemoglobin were determined every 2 s using frequency domain near-infrared spectroscopy. Blood pressure, heart rate and response entropy were collected every 10 s.

Awake oxyhaemoglobin and deoxyhaemoglobin were 18.9 ± 2.3 μmol (mean ± SD) and 12.7 ± 0.8 μmol, respectively, and neither changed significantly during induction. Intubation increased oxyhaemoglobin by 37% (P < 0.05) and decreased deoxyhaemoglobin by 16% (P < 0.05), and both measures returned to baseline within 20 min of desflurane anaesthesia. Blood pressure, heart rate and electroencephalogram response entropy increased during intubation, and the increase in heart rate correlated with the increase in brain oxygen saturation (r = 0.48, P < 0.05).

Intubation-related stress increased oxyhaemoglobin related to electroencephalogram and autonomic activation. Stress-induced brain stimulation may be monitored during anaesthesia using frequency domain near-infrared spectroscopy.

Mortality prediction systems have been calculated and validated from large mixed ICU populations. However, in daily practice it is often more important to know how a model performs in a patient subgroup at a specific ICU. Thus, we assessed the performance of three mortality prediction models in four well-defined patient groups in one centre.

A total of 960 consecutive adult patients with either severe head injury (n = 299), multiple injuries (n = 208), abdominal aortic aneurysm (n = 267) or spontaneous subarachnoid haemorrhage (n = 186) were included. Calibration, discrimination and standardized mortality ratios were determined for Simplified Acute Physiology Score II, Mortality Probability Model II (at 0 and 24 h) and Injury Severity Score. Effective mortality was assessed at hospital discharge and after 1 yr.

Eight hundred and fifty-five (89%) patients survived until hospital discharge. Over all four patient groups, Mortality Probability Model II (24 h) had the best predictive accuracy (standardized mortality ratio 0.62) and discrimination (area under the receiver operating characteristic curve 0.9), but Simplified Acute Physiology Score II performed well for patients with subarachnoid haemorrhage. Overall calibration was poor for all models (Hosmer–Lemeshow Type C-values between 20 and 26). Injury Severity Score had the worst discrimination in trauma patients. All models over-estimated hospital mortality in all four patient groups, and these estimates were more like the mortality after 1 yr.

In our surgical ICU, Mortality Probability Model II (24 h) performed slightly better than Simplified Acute Physiology Score II in terms of overall mortality prediction and discrimination; Injury Severity Score was the worst model for mortality prediction in trauma patients.

In most studies of cardiac output changes after spinal anaesthesia, the time-resolution is limited. The aim of this study was to demonstrate cardiac output changes with high time-resolution during onset of spinal anaesthesia in elderly patients.

We investigated 32 patients aged ⩾60 yr scheduled for elective lower limb surgery. Fourteen received concurrent cardiovascular medication. Cardiac output was measured every 10 s using a pulse wave algorithm derived from the radial artery pressure curve, after calibration with lithium chloride (LiDCO™plus). Data collection ended when the patients were ready for surgery, or if ephedrine was given to raise the mean arterial pressure.

Cardiac output increased initially reaching a maximum after a mean of 7 min. The average increase was 1.1 L min−1 (P < 0.0001). This occurred when mean arterial pressure was reduced 14 mmHg on average. At the end of data collection, cardiac output decreased 0.5 L min−1 from baseline (P = 0.02). Mean arterial pressure decreased progressively in all patients, and only minimal changes in heart rate were found.

Using this high time-resolution method, we detected biphasic changes in cardiac output during onset of spinal anaesthesia. Initially, cardiac output increased. Subsequently, it was significantly reduced from baseline, although this decrease was of minor clinical importance.

When intrathecally or epidurally administered, α2-adrenoceptor agonists produce potent antinociception by affecting the activity of primary afferent fibres and spinal cord neurons. Recent reports have indicated that in dorsal root ganglion neurons, tetrodotoxin-resistant Na+ channels play important roles in the conduction of nociceptive sensation. We therefore investigated the effects of α2-adrenoceptor agonists on tetrodotoxin-resistant Na+ currents.

Using the whole-cell patch-clamp technique, we recorded tetrodotoxin-resistant Na+ currents from rat dorsal root ganglion neurons.

Both clonidine and dexmedetomidine reduced the peak amplitude of the tetrodotoxin-resistant Na+ current concentration- and use-dependently. The concentration required for a half-maximal effect was significantly lower for dexmedetomidine (58.0 ± 10.2 μmol) than for clonidine (257.2 ± 30.9 μmol) at holding potential −70 mV. The current inhibitions induced by these agonists were not prevented by 1 μmol yohimbine, an α2-adrenoceptor antagonist. Both clonidine and dexmedetomidine shifted the inactivation curve for the tetrodotoxin-resistant Na+ current in the hyperpolarizing direction. The combinations clonidine with lidocaine and dexmedetomidine with lidocaine produced an additive blockade-type interaction on the tetrodotoxin-resistant Na+ current.

The results suggest that a direct inhibition of tetrodotoxin-resistant Na+ channels may contribute to the antinociceptive effects of clonidine and dexmedetomidine when used as additives to regional anaesthesia.

We surveyed delegates at the Group of Anaesthetists in Training (UK) meeting to investigate evidence of a training-gap (number of fibreoptic intubations believed to bestow competence vs. number actually performed).

Questionnaires were distributed to and collected from delegates in person. Questions covered six areas, including experience of fibreoptic intubation and cricothyrotomy, fibreoptic intubation as a specialist skill and ethical issues.

We received 221 replies (76%). All trainees believed competence to be achievable with 10 intubations (interquartile range (IQR) 10–20); the median number performed was 2 (IQR 0–4). This was statistically significant for the groups senior house officers, 1st and 2nd year registrars and 3rd and 4th year registrars; P < 0.0001. Many final year trainees (12/20, 60%) also failed to achieve their competency target. Few trainees had seen or performed any cricothyrotomies (medians 0, IQRs 0–1 and 0–0). Most (195/208, 94%) believed that fibreoptic intubation was a core skill and 199/212 (94%) believed that all should be competent by completion of training. Ten percent (n = 208) felt it unethical to perform an awake training intubation with full consent and 10% believed it acceptable without explanation. Most (82.7%) would fibreoptically intubate an asleep patient (requiring intubation) without consent.

Trainees reported a gap between their perception of competence and achievement in awake fibreoptic intubation. Simple and complex simulations and structured training programmes may help. Anaesthetists must address the ethics of clinical training in advanced airway management.

The postoperative monitoring and treatment of the patient undergoing aortic valve replacement is a complex challenge. Echocardiography is the only method which provides dynamic and real-time bedside imaging of the heart. Focused assessed transthoracic echocardiography has been shown to provide a usable window for cardiac imaging in a mixed ICU population. The aim of this study was to evaluate the feasibility of perioperative imaging of the heart and pleura according to the focused assessed transthoracic echocardiography protocol in patients scheduled for aortic valve replacement.

Thirty-five adult patients scheduled for aortic valve replacement were followed perioperatively with focused assessed transthoracic echocardiography examinations. A Vivid-7 echo-machine and a 2.5 MHz matrix transducer with second-harmonic imaging were used for data acquisition. The image quality for the cardiac window was graded 1–5 (1 = no image, 2 = poor and unusable image quality, 3 = usable image quality, 4 = good image quality and 5 = perfect image quality). A score ⩾3 equalled an image quality judged to be of sufficient quality to be interpreted and thereby to contribute to clinical decision-making.

All patients had at least one usable window preoperatively. At least one usable window was obtained in 88% of patients on the first postoperative day, and in 97% at discharge. The image quality changed over time, with the poorest quality being observed on the first postoperative day. The apical view with the patient in the left lateral position provided the best cardiac window on all occasions. The presence of drains did not significantly affect the achievability of a satisfactory examination. The number of patients with pleural effusion was relatively high. On the first postoperative day, 10 subjects had unilateral and one subject had bilateral pleural effusions. At discharge, 14 patients had unilateral and four patients had bilateral effusions.

We conclude that the image quality of the heart and pleura, according to the focused assessed transthoracic echocardiography concept, is sufficient to undergo interpretation and thereby contribute to the perioperative clinical decision-making in patients with aortic stenosis.