These results are different from the published results on which the NICE guidelines were based; however, the evidence base in children is small. There is currently insufficient evidence to support the use of ultrasound guidance for central venous catheterization in children.

We considered that, with modern perfusion equipment and mildly hypothermic cardiopulmonary bypass, protracted post-operative ventilation in an intensive care unit (ITU) is no longer required after most cardiac operations. We used a three-bedded cardiac recovery area (CRA) within the operating suite for 1,000 patients between January 1990 and June 1991. Forty-five patients with special needs were managed in the ITU. The time to extubation (T50%; range) for coronary bypass, aortic valve, mitral valve, and double-valve patients was 2.0 (0-42), 2.5 (0-12), 3.0 (0-15), and 3.0 (1-36) hours, respectively. Recovery beds were re-used allowing 5-6 operations daily. The difference in nursing staff complement for a CRA versus ITU bed was 4.5/7.8. Patient management was by nurse specialists supported by cardiac surgeons. Intervention by cardiac anaesthetists or intensivists was limited to specific ventilatory problems or renal failure. The early extubation policy failed in ten patients (five coronary, three aortic, one mitral and one double-valve patient) through poor pre-operative respiratory function, left ventricular failure or intra-operative events. The overall mortality in CRA was 1.4%. The mean duration of post-operative stay was 7 days (range 5-12). We conclude that a CRA staffed by nurse practitioners provides a safe and effective alternative to the anaesthetist-managed ITU. A rapid turnover of CRA beds removes the constraints of ITU bed availability.

Objective

SummaryPlasma levels of D-dimer have been found useful as a marker of deep venous thrombosis in those patients in whom thrombosis is suspected, but their usefulness in postoperative screening is less clear. We have investigated the relationship of D-dimer to deep venous thrombosis in 90 patients after total hip and knee arthroplasty. From the first postoperative day the D-dimer levels were found to be highly significantly raised in patients with deep venous thrombosis. A combined result over the first six postoperative days in excess of 1200ng/ml correlated with thrombosis with a specificity of 100%, sensitivity of 45%, positive predictive value of 60%, and negative predictive value of 100%. However, individual estimations were not discriminatory.

Heparin infusion may cause heparin resistance and may affect monitoring by measurement of the activated coagulation time (ACT), making the assessment of anticoagulation difficult, with the risk of over- or undertreatment, especially during cardiac surgery. We studied two groups of patients undergoing cardiopulmonary bypass (CPB): patients on heparin infusions (group H) and heparin-naive controls (group C). All patients received heparin 300 IU kg(-1) before CPB and a further dose of 5000 IU if the ACT 5 min after commencing bypass was less than 400 s. Measurements of ACT, heparin concentration, antithrombin-3, thrombin-antithrombin complex, prothrombin fragment F(1+2) and D-dimers were made before and 5 and 20 min after start of CPB. A second dose of heparin was given to eight out of 18 patients in group C and 10 out of 24 in group H. Antithrombin-3 in group H was significantly less than in group C at 5 min [59 (14) vs 52 (9)%, P<0.05]. ACT was significantly lower in group H than group C at 20 min [387 (64) vs 431 (67) s, P<0.05]. Despite ACTs of less than 400 s in both groups, no coagulation was seen, suggesting that 300 IU kg(-1) heparin is a safe dose for anticoagulation in CPB even after heparin therapy.

Three techniques of intermittent venous occlusion (an oscillotonometer cuff, pneumatic leggings and an automatic arterial pressure monitoring cuff (Sentinel] were applied to one upper limb of seven healthy male volunteers. Fibrinolytic activity was assessed by the measurement of the euglobulin clot lysis time during a control period followed by 1 h of intermittent venous occlusion. Although no statistically significant differences were found with any of the methods, between the control and experimental periods, there was a trend towards fibrinolytic enhancement with intermittent pneumatic compression.

The "preclose" technique employing two Perclose (P) devices is well established for large-bore artery (LBA) hemostasis. Occasionally, only one Perclose deploys successfully during the initial preclose because of arterial calcification necessitating the use of the crossover balloon technique to achieve hemostasis at the LBA. We sought to determine if the combined use of one Perclose and either one Angioseal or one Mynx vascular closure device (VCD) is a safe alternative closure technique large-bore arteriotomy closure. In total, 40 patients underwent high-risk percutaneous coronary intervention (HRPCI) with Impella support, of whom 38 had common femoral artery (CFA) arteriotomies and 2 underwent percutaneous axillary arteriotomy (AA). Prior to Impella insertion, one Perclose device was predeployed. At the end of HRPCI, Impella was removed and a 0.035″ wire was inserted through the Impella sheath. This sheath was then withdrawn over the wire, and partially deployed Perclose was fully deployed. A 6-Fr sheath was advanced over a 0.035″ wire into the CFA or AA, achieving hemostasis and reducing the LBA to a 6-Fr size. The 6-Fr arteriotomy was closed with a 6-Fr Mynx or Angioseal VCD. Patients were followed at day 1 and day 30. Hybrid closure was successful in 38 of 40 cases. In one case of Mynx balloon rupture, hemostasis was achieved with heparin reversal and manual compression. In the case of Perclose failure, crossover balloon tamponade at arteriotomy site and external manual compression achieved hemostasis. Patients were free of complications at day 1 and day 30. Hybrid closure with one Perclose and either one Mynx or one Angioseal VCD is safe and effective for LBA closure.

At operation a malignant bronchial carcinoid that had invaded the left atrium produced a carcinoid crisis in a 63 year old woman. A somatostatin infusion was required to resuscitate the heart and circulation and to allow subsequent resection of the carcinoid under cardiQpulmonary bypass. Case report A 63 year old white female non-smoker presented with a nine month history of persistent distressing cough. For a month she had also had epigastric pain and flushing of the face. She was not taking methyldopa. There had been no diarrhoea or weight loss or other features of the carcinoid syndrome. Physical examination showed no abnormality. There were no cardiac murmurs. A chest radiograph (figure) showed a 5 cm diameter mass at the right hilum with a further large paratracheal mass in the mediastinum above the azygos vein. A computerised tomogram suggested that this was a solitary enlarged lymph gland. A provisional diagnosis of bronchogenic carcinoma was made based on these findings. Bronchoscopy, with the rigid Stortz scope, showed external compression of the lateral wall of the right bronchus intermedius and of the right tracheal wall but no intraluminal tumour. Mediastinoscopy was performed to obtain a histological diagnosis. The smooth tumour was easily located and a biopsy specimen was taken without event. Histology showed this to be a metastatic bronchial carcinoid; however, despite a history suggestive of hormone secretion, urinary hydroxyindole acetic acid estimations proved negative. Biochemical and haematological screening tests were normal. Liver ultrasound and bone scans were unremarkable.Because there was no evidence of extrathoracic spread surgical resection was considered to be the treatment of choice. The first approach was by right thoracotomy which confirmed the preoperative findings. The mediastinal deposit measuring 10 cm by 8 cm was first mobilised and then excised without difficulty from the mediastinum. No other signs of tumour spread were seen and there were no adverse haemodynamic consequences. However, the pulmonary mass was situated centrally, between the superior and inferior pulmonary veins and seemed to invade through the pericardium at this point. Involvement of the hilum including the right main pulmonary artery above the fissure meant that pneumonectomy was required. However, an attempt to mobilise the lesion provoked profound hypotension (systolic < 30 mm Hg) and there was