Drugs for preventing postoperative nausea and vomiting in adults after general anaesthesia: an abridged Cochrane network meta-analysis.

Postoperative nausea and vomiting is a common adverse effect of anaesthesia. Although dozens of different anti-emetics are available for clinical practice, there is currently no comparative ranking of efficacy and safety of these drugs to inform clinical practice. We performed a systematic review with network meta-analyses to compare, and rank in terms of efficacy and safety, single anti-emetic drugs and their combinations, including 5-hydroxytryptamine3 , dopamine-2 and neurokinin-1 receptor antagonists; corticosteroids; antihistamines; and anticholinergics used to prevent postoperative nausea and vomiting in adults after general anaesthesia. We systematically searched for placebo-controlled and head-to-head randomised controlled trials up to November 2017 (updated in April 2020). We assessed how trustworthy the evidence was using the Grading of Recommendations Assessment, Development and Evaluation (GRADE) and Confidence In Network Meta-Analysis (CINeMA) approaches for vomiting within 24 h postoperatively, serious adverse events, any adverse event and drug class-specific side-effects. We included 585 trials (97,516 participants, 83% women) testing 44 single drugs and 51 drug combinations. The studies' overall risk of bias was assessed as low in only 27% of the studies. In 282 trials, 29 out of 36 drug combinations and 10 out of 28 single drugs lowered the risk of vomiting at least 20% compared with placebo. In the ranking of treatments, combinations of drugs were generally more effective than single drugs. Single neurokinin-1 receptor antagonists were as effective as other drug combinations. Out of the 10 effective single drugs, certainty of evidence was high for aprepitant, with risk ratio (95%CI) 0.26 (0.18-0.38); ramosetron, 0.44 (0.32-0.59); granisetron, 0.45 (0.38-0.54); dexamethasone, 0.51 (0.44-0.57); and ondansetron, 0.55 (0.51-0.60). It was moderate for fosaprepitant, 0.06 (0.02-0.21) and droperidol, 0.61 (0.54-0.69). Granisetron and amisulpride are likely to have little or no increase in any adverse event compared with placebo, while dimenhydrinate and scopolamine may increase the number of patients with any adverse event compared with placebo. So far, there is no convincing evidence that other single drugs effect the incidence of serious, or any, adverse events when compared with placebo. Among drug class specific side-effects, evidence for single drugs is mostly not convincing. There is convincing evidence regarding the prophylactic effect of at least seven single drugs for postoperative vomiting such that future studies investigating these drugs will probably not change the estimated beneficial effect. However, there is still considerable lack of evidence regarding safety aspects that does warrant investigation.

Influence of xenon on pulmonary mechanics and lung aeration in patients with healthy lungs.

BACKGROUND: The anaesthetic xenon shows potent organ-protective properties. Due to high density and dynamic viscosity, peak inspiratory pressure (Pmax) increases during xenon application. Thus, barotrauma may counteract organ protection. Accordingly, we investigated the influence of xenon on lung mechanics and lung aeration in patients with normal and reduced thoracic wall compliance. METHODS: After registration and ethical approval, 20 patients free of pulmonary disease undergoing routine xenon-based anaesthesia were mechanically ventilated. The primary outcome variable transpulmonary pressure (Ptp) was determined from plateau pressure and intraoesophageal pressure before and after xenon wash-in. We recorded Pmax, and calculated airway resistance (RAW), and static (Cstat) and dynamic (Cdyn) respiratory compliances. Finally, lung aeration was quantified by electrical impedance tomography-derived centre of ventilation index (CVI) and global inhomogeneity index (GI) in the awake state, before and during xenon. RESULTS: Xenon increased Pmax [20.8 (SD 3) vs 22.6 (3) cm H2O, P<0.001] and RAW [0.9 (0.2) vs 1.4 (0.3) cm H2O litre-1 s, P<0.001], without affecting Ptp [1.5 (4) vs 2.0 (4) cm H2O, P=0.15]. While Cstat remained unchanged, Cdyn was reduced [33.9 (7) vs 31.2 (6) ml (cm H2O)-1, P<0.001). A ventral tidal volume shift after anaesthesia induction [CVI 0.53 (0.03) vs 0.59 (0.04), P<0.001] was unaltered during xenon [CVI 0.59 (0.04), P=0.29]. Homogeneity of lung aeration was also unchanged during xenon [GI 0.37 (0.03) vs 0.37 (0.03), P=0.99]. There were no clinically meaningful differential BMI-related effects. CONCLUSIONS: Xenon increases calculated airway resistance and peak inspiratory pressure without affecting transpulmonary pressure, independent of BMI. CLINICAL TRIAL REGISTRATION: NCT02682758.

The hip fracture surgery in elderly patients (HIPELD) study to evaluate xenon anaesthesia for the prevention of postoperative delirium: a multicentre, randomized clinical trial.

BACKGROUND: Postoperative delirium occurs frequently in elderly hip fracture surgery patients and is associated with poorer overall outcomes. Because xenon anaesthesia has neuroprotective properties, we evaluated its effect on the incidence of delirium and other outcomes after hip fracture surgery. METHODS: This was a phase II, multicentre, randomized, double-blind, parallel-group, controlled clinical trial conducted in hospitals in six European countries (September 2010 to October 2014). Elderly (≥75yr-old) and mentally functional hip fracture patients were randomly assigned 1:1 to receive either xenon- or sevoflurane-based general anaesthesia during surgery. The primary outcome was postoperative delirium diagnosed through postoperative day 4. Secondary outcomes were delirium diagnosed anytime after surgery, postoperative sequential organ failure assessment (SOFA) scores, and adverse events (AEs). RESULTS: Of 256 enrolled patients, 124 were treated with xenon and 132 with sevoflurane. The incidence of delirium with xenon (9.7% [95% CI: 4.5 -14.9]) or with sevoflurane (13.6% [95% CI: 7.8 -19.5]) were not significantly different (P=0.33). Overall SOFA scores were significantly lower with xenon (least-squares mean difference: -0.33 [95% CI: -0.60 to -0.06]; P=0.017). For xenon and sevoflurane, the incidence of serious AEs and fatal AEs was 8.0% vs 15.9% (P=0.05) and 0% vs 3.8% (P=0.06), respectively. CONCLUSIONS: Xenon anaesthesia did not significantly reduce the incidence of postoperative delirium after hip fracture surgery. Nevertheless, exploratory observations concerning postoperative SOFA-scores, serious AEs, and deaths warrant further study of the potential benefits of xenon anaesthesia in elderly hip fracture surgery patients. CLINICAL TRIAL REGISTRATION: EudraCT 2009-017153-35; ClinicalTrials.gov NCT01199276.

Predictors for postoperative nausea and vomiting after xenon-based anaesthesia.

BACKGROUND: In contrast to volatile anaesthetics, xenon acts by antagonism at N-methyl-d-aspartate receptors and antagonizes 5-hydroxytryptamine type 3 receptors that mediate nausea and vomiting. Therefore, it is unknown whether the same risk factors for postoperative nausea and vomiting (PONV) after volatile anaesthetics apply to xenon-based anaesthesia. METHODS: With ethics committee approval and written informed consent, 502 consecutive patients undergoing xenon-based anaesthesia were included in a multicentre prospective observational study. Antiemetic prophylaxis was administered at the discretion of the attending anaesthetists. Postoperative nausea and vomiting and need for antiemetic rescue medication were assessed for 24 h after anaesthesia. Multivariate logistic regression analysis was performed to quantify risk factors for PONV and need for rescue medication. RESULTS: Four hundred and eighty-eight subjects were available for the final analysis. The incidence of PONV in subjects without prophylaxis was lower than expected according to the Apfel Score (28% observed; 42% expected, P<0.001). Independent predictors for PONV were (adjusted odds ratio; 95% confidence interval) female sex (1.76; 1.08-2.89), younger patient age (0.82 per 10 yr; 0.69-0.97), and longer duration of anaesthesia (1.36 per hour; 1.17-1.59). CONCLUSIONS: The incidence of PONV was significantly lower than predicted by the Apfel Score. Female sex, younger age, and longer duration of anaesthesia are risk factors for PONV after xenon-based anaesthesia. CLINICAL TRIAL REGISTRATION: German Federal Institute for Drugs and Medical Devices number AL-PMS-01/07GER.

Cardiovascular stability and unchanged muscle sympathetic activity during xenon anaesthesia: role of norepinephrine uptake inhibition.

BACKGROUND: Intraoperative hypotension is associated with increased risk of perioperative complications. The N-methyl-d-aspartate (NMDA) receptor (NMDA-R) antagonist xenon (Xe) induces general anaesthesia without impairment of cardiac output and vascular resistance. Mechanisms involved in cardiovascular stability have not been identified. METHODS: Muscle sympathetic activity (MSA) (microneurography), sympathetic baroreflex gain, norepinephrine (NE) plasma concentration (high-performance liquid chromatography), anaesthetic depth (Narcotrend(®) EEG monitoring), and vital parameters were analysed in vivo during Xe mono-anaesthesia in human volunteers (n=8). In vitro, NE transporter (NET) expressing HEK293 cells and SH-SY5Y neuroblastoma cells were pre-treated with ketamine, MK-801, NMDA/glycine, or vehicle. Subsequently, cells were incubated with or without Xe (65%). NE uptake was measured by using a fluorescent NET substrate (n=4) or [(3)H]NE (n=6). RESULTS: In vivo, Xe anaesthesia increased mean (standard deviation) arterial pressure from 93 (4) to 107 (6) mm Hg and NE plasma concentration from 156 (55) to 292 (106) pg ml(-1), P<0.01. MSA and baroreflex gain were unaltered. In vitro, ketamine decreased NET activity (P<0.01) in NET-expressing HEK293 cells, while Xe, MK-801, and NMDA/glycine did not. Xe reduced uptake in SH-SY5Y cells expressing NET and NMDA-Rs (P<0.01). MK-801 (P<0.01) and ketamine (P<0.01) also reduced NET activity, but NMDA/glycine blocked the effect of Xe on [(3)H]NE uptake. CONCLUSIONS: In vivo, Xe anaesthesia does not alter sympathetic activity and baroreflex gain, despite increased mean arterial pressure. In vitro, Xe decreases the uptake of NE in neuronal cells by the inhibition of NET. This inhibition might be related to NMDA-R antagonism and explain increased NE concentrations at the synaptic cleft and in plasma, contributing to cardiovascular stability during Xe anaesthesia.

Ventilation with low tidal volumes during upper abdominal surgery does not improve postoperative lung function.

BACKGROUND: Prolonged postoperative decrease in lung function is common after major upper abdominal surgery. Evidence suggests that ventilation with low tidal volumes may limit the damage during mechanical ventilation. We compared postoperative lung function of patients undergoing upper abdominal surgery, mechanically ventilated with high or low tidal volumes. METHODS: This was a double-blind, prospective, randomized controlled clinical trial. One hundred and one patients (age ≥ 50 yr, ASA ≥ II, duration of surgery ≥ 3 h) were ventilated with: (i) high [12 ml kg(-1) predicted body weight (PBW)] or (ii) low (6 ml kg(-1) PBW) tidal volumes intraoperatively. The positive end-expiratory pressure was 5 cm H(2)O in both groups and breathing frequency adjusted to normocapnia. Time-weighted averages (TWAs) of forced vital capacity (FVC) and forced expiratory volume in 1 s (FEV(1)) until 120 h after operation were compared (P<0.025 considered statistically significant). Secondary outcomes were oxygenation, respiratory and non-respiratory complications, length of stay and mortality. RESULTS: The mean (sd) values of TWAs of FVC and FEV(1) were similar in both groups: FVC: 6 ml group 1.8 (0.7) litre vs 12 ml group 1.6 (0.5) litre (P=0.12); FEV(1): 6 ml group 1.4 (0.5) litre vs 12 ml group 1.2 (0.4) litre (P=0.15). FVC and FEV(1) at any single time point and secondary outcomes did not differ significantly between groups. CONCLUSIONS: Prolonged impaired lung function after major abdominal surgery is not ameliorated by low tidal volume ventilation.

Prospective observational study for perioperative volume replacement with 6% HES 130/0,42, 4% gelatin and 6% HES 200/0,5 in cardiac surgery.

BACKGROUND: the constantly growing amount of different kinds of colloid fluids necessitates comparative investigations with regards to the safety and effectivity in clinical use of these preparations. Hence we compared three colloid fluids in an observational study. The objective was the exploration of the influence of these three colloids on blood coagulation, hemodynamics and renal function of the cardiac surgical patient. METHODS: we included 90 patients undergoing an elective open-heart surgery with the use of the heart-lung machine and observed them consecutively. Group 1 [gelatin 4% (n = 30)], Group 2 [HES 200/0,5 (n = 30)] and Group 3 [HES 130/0,42 (n = 30)]. We measured the perioperative volume replacement, the administration of blood- and coagulation-products, the application of catecholamines, the renal function, blood gas and the platelet aggregation using multiplate electrode analyzer (Multiplate, Dynabyte medical, Munich, Germany). RESULTS: the gelatin-group needed significantly more norepinephrine than the HES 130/0.42 group. The responsible surgeon considered the blood coagulation in the HES 200/0.5 group most frequently as impaired. Furthermore we saw a significant decrease in platelet function in the HES 200/0.5 group when performing the multiplate-analysis (ADP-and COL-test). HES 130/0.4 as well as gelatin 4% showed no significant change in platelet function. The gelatin-group and the HES 200/0.5 needed significantly more aprotinine than the HES 130/0.4 group. We saw no significant difference with regards to administration of blood and coagulation products between the three groups. The urinary excretion during the intervention was significantly higher in the HES 200/0.5 group and in the gelatin group than in the HES 130/0.4 group. CONCLUSIONS: our results confirm the lower stabilizing effect of gelatin on circulation during fluid resuscitation. The blood coagulation was mostly impaired due to HES 200/0.5 confirmed by the multiplate®-analysis as well as by different clinical findings.

Effects of desmopressin on platelet function under conditions of hypothermia and acidosis: an in vitro study using multiple electrode aggregometry*.

SUMMARY: Hypothermia and acidosis lead to an impairment of coagulation. It has been demonstrated that desmopressin improves platelet function under hypothermia. We tested platelet function ex vivo during hypothermia and acidosis. Blood samples were taken from 12 healthy subjects and assigned as follows: normal pH, pH 7.2, and pH 7.0, each with and without incubation with desmopressin. Platelet aggregation was assessed by multiple electrode aggregometry. Baseline was normal pH and 36 degrees C. The other samples were incubated for 30 min and measured at 32 degrees C. Acidosis significantly impaired aggregation. Desmopressin significantly increased aggregability during hypothermia and acidosis regardless of pH, but did not return it to normal values at low pH. During acidosis and hypothermia, acidosis should be corrected first; desmopressin can then be administered to improve platelet function as a bridge until normothermia can be achieved.

Prospective study comparing skin impedance with EEG parameters during the induction of anaesthesia with fentanyl and etomidate.

OBJECTIVE: Sympathetic stimulation leads to a change in electrical skin impedance. So far it is unclear whether this effect can be used to measure the effects of anaesthetics during general anaesthesia. The aim of this prospective study is to determine the electrical skin impedance during induction of anaesthesia for coronary artery bypass surgery with fentanyl and etomidate. METHODS: The electrical skin impedance was measured with the help of an electro-sympathicograph (ESG). In 47 patients scheduled for elective cardiac surgery, anaesthesia was induced with intravenous fentanyl 10 mug/kg and etomidate 0.3mg/kg. During induction, the ESG (Electrosympathicograph), BIS (Bispectral IndeX), BP (arterial blood pressure) and HR (heart rate) values of each patient were recorded every 20 seconds.The observation period from administration of fentanyl to intubation for surgery lasted 4 min. - RESULTS: The ESG recorded significant changes in the electrical skin impedance after administration of fentanyl and etomidate(p <0.05). During induction of anaesthesia, significant changes of BIS, HR and blood pressure were observed as well (p <0.05). CONCLUSIONS: The electrical skin impedance measurement may be used to monitor the effects of anesthetics during general anaesthesia.

Whole blood coagulation and platelet activation in the athlete: a comparison of marathon, triathlon and long distance cycling.

INTRODUCTION: Serious thrombembolic events occur in otherwise healthy marathon athletes during competition. We tested the hypothesis that during heavy endurance sports coagulation and platelets are activated depending on the type of endurance sport with respect to its running fraction. MATERIALS AND METHODS: 68 healthy athletes participating in marathon (MAR, running 42 km, n = 24), triathlon (TRI, swimming 2.5 km + cycling 90 km + running 21 km, n = 22), and long distance cycling (CYC, 151 km, n = 22) were included in the study. Blood samples were taken before and immediately after completion of competition to perform rotational thrombelastometry. We assessed coagulation time (CT), maximum clot firmness (MCF) after intrinsically activation and fibrin polymerization (FIBTEM). Furthermore, platelet aggregation was tested after activation with ADP and thrombin activating peptide 6 (TRAP) by using multiple platelet function analyzer. RESULTS: Complete data sets were obtained in 58 athletes (MAR: n = 20, TRI: n = 19, CYC: n = 19). CT significantly decreased in all groups (MAR -9.9%, TRI -8.3%, CYC -7.4%) without differences between groups. In parallel, MCF (MAR +7.4%, TRI +6.1%, CYC +8.3%) and fibrin polymerization (MAR +14.7%, TRI +6.1%, CYC +8.3%) were significantly increased in all groups. However, platelets were only activated during MAR and TRI as indicated by increased AUC during TRAP-activation (MAR +15.8%) and increased AUC during ADP-activation in MAR (+50.3%) and TRI (+57.5%). DISCUSSION: While coagulation is activated during physical activity irrespective of type we observed significant platelet activation only during marathon and to a lesser extent during triathlon. We speculate that prolonged running may increase platelet activity, possibly, due to mechanical alteration. Thus, particularly prolonged running may increase the risk of thrombembolic incidents in running athletes.

[Implantation of cardioverter-defibrillators. How much anesthesia is necessary?]. / Implantation von Kardioverter-Defibrillatoren. Wie viel Anästhesie ist nötig?

Updated cardiologic guidelines constitute the background for an extended spectrum of indications for the implantation of automatic implantable cardioverter defibrillators (AICDs) and lead to an increasing number of operative implantations of AICDs. Moreover, during implantation of devices for cardiac resynchronization therapy the anesthesiologist is responsible for the most critically ill patients with the longest duration of surgery. As a result anesthesiologists face an increasing number of critically ill patients, whose management contributes to perioperative outcome. Automatic implantable cardioverter defibrillators can be implanted either during general anesthesia, local anesthesia or during a combination of local anesthesia combined with deep conscious sedation accomplished by an anesthesiologist. Besides economic aspects there is an increasing demand for anesthesia with the least cardiovascular side effects and rapid recovery in the often seriously ill patient with preexisting limitations of cardiac and pulmonary functions. Accordingly procedure and anesthesia-associated risks are reviewed and an algorithm for anesthesia management is suggested.

Alterations in forearm vascular reactivity in patients with septic shock.

Patients with septic shock are haemodynamically unstable and suffer from vasodilation. Studying the human forearm vascular bed in patients with septic shock, we tested the hypothesis that the responses to regionally infused endothelium-(in)dependent vasodilators and vasoconstrictors are uniformly impaired. Forearm blood flow (FBF, venous occlusion plethysmography) and brachial arterial pressure were determined to calculate forearm vascular resistance (FVR) in eight consecutive sedated, mechanically ventilated patients with septic shock (APACHE II Score range 21-34, SOFA Score 11-16) and 11 healthy volunteers. Despite increased baseline FBF in patients with septic shock (6.1 (SD 1.5) ml x min(-1) x (100 ml of tissue)(-1) compared to 4.7 (1.4) in volunteers) the significant decreases in FVR seen in response to exogenous nitric oxide (nitroprusside) and acetylcholine did not differ between groups. However, compared to volunteers, mitigation of endogenous nitric oxide production by a low dose of N(G)-methyl-L-arginine acetate (L-NMMA) caused a significant increase (+6.7 mmHg x min x ml(-1)) in septic patients. Regional vasoconstriction in response to phenylephrine (FVR: +9.9 vs +30.7 mmHg x min x ml(-1) in controls) and angiotensin II (FVR: +9.0 vs +67.4 mmHg x min x ml(-1)) was markedly impaired. In contrast, vasopressin, in dosages evoking no vasoconstriction in volunteers, induced a significant increase in FVR in septic patients (+10.0 mmHg x min x ml(-1)). In the forearm of patients with septic shock, vasoconstriction by alpha1- and angiotensin II receptor agonists is selectively impaired, whereas the vasoconstrictor response to vasopressin is exaggerated. These findings exclude a generalised impairment of vasomotor activity in patients with septic shock and provide a rationale for vasopressin administration.

Off pump coronary artery bypass grafting - midterm results.

OBJECTIVE: Off pump coronary artery bypass (OPCAB) grafting is still discussed controversially in the cardiac surgical community. Early perioperative results are encouraging. Only few reports have focused on mid-term recurrence of angina and freedoms from death or re-intervention. - METHODS: 107 OPCAB patients (mean age 63 +/- 1 years, 77 male, log EuroScore 5.6 +/- 0.7, number of distal anastomoses 2.0 +/- 0.1), operated on between January 1999 and December 2003, were systematically followed up comparing pre- and post-op NYHA- and CCS-classifications and assessing freedom from death and re-intervention. 52 of 107 patients underwent postoperative angiography or multi-slice computed tomography (MSCT); 6 of the latter 52 patients were symptomatic, 3 with unstable angina, the others underwent follow-up studies having given their informed consent. - RESULTS: The 30 day mortality was 2%. Freedom from death or re-intervention at 5.5 years was 91% and 80%, respectively. Only three patients required re-intervention in an OPCAB-related vessel. CCS classification was 2.8 +/- 0.1 before surgery and 1.8 +/- 0.2 (p<0.01) at follow-up (3.3 +/- 0.3 years). NYHA classification was 2.7 +/- 0.1 and 2.2 +/- 0.1 (p<0.01), respectively. Out of 107 patients, 52 underwent coronary angiography or MSCT (6 for cardiac symptoms) at a mean follow-up of 2.2 +/- 0.3 years. Left internal thoracic artery was patent in 91%, venous graft patency rate was 83%. - CONCLUSIONS: In this small but consecutive OPCAB population with a considerable perioperative risk according to the EuroScore, freedom from death and re-intervention at 5.5 years is acceptable and graft patency rate at 2.2 +/- 0.3 years is in the expected range. Significant reduction in both CCS and NYHA classification indicate sustained clinical improvement at mid-term.

Impact of intraoperative myocardial cellular damage on early hemodynamics after off-pump versus on-pump coronary artery bypass surgery.

BACKGROUND: Due to the surgical trauma a small amount of myocardial cellular damage is inherent during coronary artery bypass grafting (CABG). The purpose of the present study was to assess the degree of myocardial cellular damage after off-pump (OPCAB) and on-pump CABG (ONCAB) as measured by cardiac troponin I (cTnI), creatine kinase (CK), its MB isoenzyme (CK-MB) and myoglobin (Myo) and to examine its impact on early hemodynamics after surgery. METHODS: Ninety-nine consecutive OPCAB patients, operated between 01/1999 and 01/2004, were enrolled in the present study and compared to 99 ONCAB patients operated during the same period of time, who were matched for baseline data and mean number of grafts per patient. Early hemodynamics, cTnI, CK/CK-MB and Myo were measured preoperatively and at 1, 6, 12, 24 and 48 hours (h) postoperatively. Perioperative inotropic support, clinical data and potoperative outcome were recorded prospectively. RESULTS: The two groups were similar concerning preoperative characteristics. The mean number of distal grafts/patient was 2.1 +/- 1.0 in OPCAB and 2.1 +/- 0.8 in ONCAB patients (mean +/- SD). There was no significant difference among the groups regarding early hemodynamics in terms of cardiac index (CI), systemic vascular resistance index (SVRI), and left ventricular stroke work index (LVSWI), and inotropic support. However, cTnI, CK/CK-MB but not Myo levels were significantly lower in OPCAB compared to ONCAB patients at 1, 6, 12, 24, 36 and 48 h postoperatively (P<0.05). CONCLUSIONS: Off-pump surgery results in equal early hemodynamics despite a significantly lower release of cTnI and CK, suggesting a reduced myocardial cell damage as compared to ONCAB surgery.

Antiphospholipid syndrome in cardiac surgery-an underestimated coagulation disorder?

OBJECTIVE: Antiphospholipid syndrome (APS) is a rare coagulation disorder associated with recurrent arterial and venous thrombotic events. We analysed our experience with five APS patients who underwent cardiac surgery. In three of them the diagnosis of APS had been established before surgery, two patients were diagnosed after surgery. METHODS: From March 1999 to March 2004 five patients with APS underwent cardiac surgery using cardiopulmonary bypass (CPB). We retrospectively reviewed their clinical data, operative and postoperative courses, and the long-term results. RESULTS: Procedures performed were heart and lung transplantation (patient 1), endoventriculoplasty and CABG (patient 2), biventricular resection of endoventricular fibrosis and thrombus (patient 3), mitral valve repair repair and coronary artery bypass grafting (CABG, patient 4), and mitral valve replacement with closure of a patent foramen ovale (patient 5). There were three perioperative deaths (patients 1, 2 and 3), two of three patients in whom the diagnosis was known before surgery, survived (patients 4 and 5). In these patients, only half the dose of protamin (patient 4) and no protamin at all (patient 5) was applied to reduce the probability of postoperative thromboembolic complications. At 1 year follow up, only patient 4 had survived, patient 5 had died of the complications of intestinal thromboembolism. CONCLUSIONS: Patients with APS undergoing cardiac surgery belong to a high risk subgroup. Thus, though rare, APS can be a critical issue in cardiac surgery. Some of the cardiac patients with unexplained perioperative thromboembolic complications, such as graft occlusion, may turn out to have an undiagnosed APS.