Rapid reversal of coagulopathy in warfarin-related intracranial haemorrhages with prothrombin complex concentrates.

We report our initial experience using Profilnine SD, a 3-Factor prothrombin complex concentrate (PCC) in combination with fresh frozen plasma and vitamin K in seven patients admitted to our neurointensive care unit with oral anticoagulation therapy-related intracranial haemorrhage over a six-month period, to achieve rapid normalisation of the international normalised ratio (INR) and allow surgical evacuation when indicated. Four patients presented with subdural haematomas while three had intracerebral haematomas. Six of seven patients had admission INR in the appropriate therapeutic range for oral anticoagulation therapy. The median dose of PCC administered was 28.5 IU/kg body weight (interquartile range 21.3 to 38.5 IU/kg). All four patients with subdural haematoma underwent surgical evacuation once INR was less than 1.5. Median time from computed tomography diagnosis to surgery was 275 minutes (range 102 to 420 minutes). The median time to INR normalisation post-PCC administration was shorter, at 85 minutes (range 50 to 420 minutes) for the four patients who survived, versus 10 hours (range 9 to 44 hours) in the three patients who died. Two of the three patients who died had haematoma increase, worsening midline shift and subfalcine herniation, leading to withdrawal of therapy. Prothrombin complex concentrates should be considered for use in the urgent reversal of INR in oral anticoagulation therapy-related intracranial haemorrhage, potentially halting haematoma expansion and expediting urgent neurosurgical intervention, although data from randomised controlled trials is still lacking. The literature supporting the use of PCC is reviewed and a protocolised emergent treatment algorithm is proposed, which may help achieve earlier consistent normalisation of the INR.

Anaesthetic management of planned or emergency endovascular stent graft surgery for thoracic aneurysms and dissections.

We reviewed the intraoperative anaesthetic management and outcome of seven consecutive cases of endovascular stent graft surgery for thoracic aortic aneurysms or dissections over a period of 20 months in our institution. Seven males (median age 63) underwent endovascular stent graft surgery of the thoracic aorta under general anaesthesia. Four were emergency procedures for acute dissection or leaks of thoracic aneurysms. The duration of the procedures ranged from 120 to 300 minutes. Intraoperative stent migration occurred in one patient. Induced hypotension and immobilization were critical during stent deployment. Six out of the seven patients were discharged home between three to 20 days postoperatively. There was no 30-day mortality. One patient died sixty days postoperatively. Anaemia and respiratory complications were the most common postoperative problems encountered. None of the patients sustained spinal cord ischaemia, acute myocardial infarction or renal impairment. Endovascular stent graft surgery of the thoracic aorta is a relatively new alternative to conventional open surgery. Our experience suggests satisfactory short-term outcome even for patients with acute dissection. However, long-term survival and results require further evaluation.

Acute intraoperative cerebral oedema: are current therapies evidence based?

Acute intraoperative ischaemic cerebral oedema following torrential haemorrhage from the left intracranial internal carotid artery occurred during resection of a recurrent middle cranial fossa meningioma. A series of immediate anaesthetic interventions was effective in reducing brain oedema, allowed for surgical haemostasis, and resulted in no permanent sequelae to patient outcome. A review of the literature indicates that direct evidence for the efficacy of extremely early interventions as described in this case report is lacking and must be extrapolated from other brain injury models.

Advances in neurointensive care.

PURPOSE: To highlight recent advances in neurological and neurosurgical intensive care. DATA SOURCES: A MEDLINE search was conducted from January 1980 to August 2000. Keywords included intensive care, head injury, subarachnoid haemorrhage, status epilepticus, myasthenic crisis, Guillain-Barre syndrome and stroke. All articles in English were considered for review. Additional articles were identified from the references of the retrieved articles and cross-referencing selected articles. DATA EXTRACTION: All clinical studies, review articles and abstracts were reviewed. DATA SYNTHESIS: Rapid advances in neurological and neurosurgical intensive care in the last decade have led to the development of specialised neurointensive care units with joint ventures between neurology and neurosurgery. Work in these units have contributed immensely to our understanding of the pathophysiology and management of acute brain injury. The principles of intensive care management include amelioration or reversal of brain injury and preservation of normal neural tissue. Treatment algorithms are possible with the aid of intense clinical and neurophysiologic monitoring. Ongoing clinical and basic science research may provide new treatment options for the intensivist in the acute phase of brain injury. CONCLUSION: Specialised neurointensive care units provide the best environment for the patient with acute brain injury. Outcome is frequently enhanced the clinicians skilled towards dealing with the whole spectrum of neurologic insults.

Decision analytic approach to severe head injury management.

Severe head injury management in the intensive care unit is extremely challenging due to the complex domain, the uncertain intervention efficacies, and the time-critical setting. We adopt a decision analytic approach to automate the management process. We document our experience in building a simplified influence diagram that involves about 3000 numerical parameters. We identify the inherent problems in structuring a model with unclear domain relationships, numerous interacting variables, and real-time multiple inputs. We analyze the effectiveness and limitations of the decision analytic approach and present a set of desiderata for effective knowledge acquisition in this setting. We also propose a semi-qualitative approach to parameter elicitation.

Outcome of patients with traumatic brain injury managed on a standardised head injury protocol.

A standardised protocol in the management of severe head injury in our hospital enables pre-determined critical care-paths and consistent treatment regimes to be instituted. In Singapore there has been no previously reported data on the outcome of severely head injured patients. Over a 6-month period, 48 consecutive patients who were enrolled in our severe head injury protocol were prospectively studied. In addition to demographic and outcome data, physiologic measurements obtained from a computerised patient information system (Carevue Hewlett-Packard 9000) were analysed to determine the mean cerebral perfusion pressure (CPP) and intracranial pressure (ICP) achieved throughout the protocol period. Median Glasgow Coma Score for all patients on admission to the protocol was 6 (range 4 to 8). The mean age was 34.46 +/- 15.03 years with a male to female ratio of 43:5. The average duration of treatment on the protocol was 110.73 hours. Initial ICP measured was 25.5 +/- 19.68 mmHg. Outcome was measured at 6 months post-injury using the Glasgow Outcome Score. Favourable outcome (GOS 4-5) was seen in 29 of 48 patients (60.4%) while 12 out of 48 (25%) had an unfavourable outcome. There was a mortality of 14.6% (7 of 48 patients). Patients who survived had a higher mean CPP (P = 0.00005), a lower initial ICP and a mean ICP (P = 0.007 and 0.0009). The use of a protocol with standardised treatment goals in the management of traumatic brain injury allows for the optimal use of limited resources and provides consistency in treatment. Good outcome is related to early aggressive resuscitation to prevent hypotension and hypoxia, prompt evacuation of surgical mass lesions and the maintenance of an adequate cerebral perfusion pressure. Our results are comparable with that reported in other established neurotrauma systems.

Diabetes insipidus in neurosurgical patients.

Diabetes insipidus (DI) is an uncommon but important complication in the neurosurgical population. This retrospective study aimed to determine the incidence, profile and outcome of patients admitted to an 18-bedded neurosurgical intensive care unit who developed DI. The overall incidence was 3.7% (29/792 admissions). Aetiologies included subarachnoid haemorrhage (12/29), severe head injury (11/29), post-surgical excision of craniopharyngioma or pituitary adenoma (5/29) and acute haemorrhagic stroke (1/29). All patients were treated with a regime of fluid replacement, electrolyte correction, parenteral or intranasal desmopressin (DDAVP), or parenteral pitressin. Overall mortality was 72.4%. There were no deaths in the patients who underwent excision of tumours. Complications included acute pulmonary oedema, hypernatremia and hypokalaemia. The development of DI was found to be associated with impending brain death and mortality in the majority of patients with subarachnoid haemorrhage and severe head injury. However, careful diagnosis and management of DI after hypothalamo-neurohypophyseal surgery did not result in any permanent neurological sequelae.

Foregoing life support in medically futile patients.

Advances in medical technology have rendered the ability to provide prolonged physiologic support of incurable or terminally ill patients commonplace in the intensive care unit. In tandem, there has been a global shift in the intensivist's mindset from solely pursuing an unrelenting course of aggressive therapy, to a recognition of the limitations of intensive care and the appropriate discontinuance of nonbeneficial therapy. Underpinning this shift remains the physician's adherence to the ethical principles of beneficence, nonmaleficence, and disclosure; the patient's right to autonomy and self determination; and the community's right to just distribution of medical resources. When the doctor assumes the role of patient advocate, and assesses illness severity and evaluate recovery, or lack of, to a quality consistent with the patient's own life philosophy, he is able to communicate to the family a course of action that is in the patient's best interest. A consensus on withholding or withdrawal of care is often then achieved. The process of foregoing or withdrawing life-sustaining therapy itself, must be carried out with sensitivity and empathy, with the primary goal of providing comfort and reducing suffering.

A case of pseudohyperkalaemia and thrombocytosis.

An unusual cause of "hyperkalaemia" was observed in a neurosurgical patient admitted to our intensive care unit. The cause of the hyperkalaemia was not known initially and treatment with cation ion exchange resin was initiated to lower the elevated serum potassium level. The concurrent occurrence of thrombocytosis and hyperkalaemia raised the possibility of psuedohyperkalaemia associated with thrombocytosis. Simultaneous measurement of plasma and serum potassium with the Hitachi 917 Analyzer (indirect ion selective electrode, coefficient of variation = 1% to 2%) confirmed the diagnosis. Correlation between thrombocytosis and pseudohyperkalaemia was found to be highly significant (r = 0.54; P = 0.014). It is estimated that for for every 100 x 10(9)/L of platelets, an increase of 0.07 to 0.15 mmol/L of potassium is expected. In thrombocytosis, plasma rather than serum potassium should be measured.

Predicting the size of a double-lumen endobronchial tube based on tracheal diameter.

UNLABELLED: We assessed whether using the tracheal diameter to predict the correct size of the left double-lumen endobronchial tube (DLT) could be used for our generally smaller sized Asian patients. Sixty-six consecutive adult patients under anesthesia for elective surgery requiring the use of a DLT were studied. The size of the left-sided DLT used was based on the width of patients' trachea measured from the preoperative posterior-anterior chest radiograph. The placement of the DLT was standardized and confirmed with fiberoptic bronchoscopy. The correct size of the DLT was the largest size tube inserted into the left bronchus with a small air leak detectable when the endobronchial cuff was deflated but not exceeding the recommended resting volume when inflated for lung isolation. Using this method of choosing our DLT, we found that an oversized DLT was often chosen especially among our female Asian patients. The overall positive predictive values for the male and female patients were 77.3% and 45.5%, respectively. We postulate that this could be due to our criteria for correct DLT size or that our local Asian patients, especially the females, were smaller and shorter. IMPLICATIONS: This study assessed whether the correct double-lumen endobronchial tube size could be predicted from tracheal diameter measurements taken from the chest radiograph. We found that this method of choosing the double-lumen endobronchial tubes was not always reliable.

The role of preoperative cervical spine X-rays in rheumatoid arthritis.

The cervical spine is frequently involved in rheumatoid arthritis and yet there exists no consensus on the need to screen for cervical spine subluxations preoperatively. We reviewed retrospectively 77 patients who underwent 132 operations under general or regional anaesthesia over a 44-month period. We found that while the majority of patients had received preoperative X-ray screening for cervical spine instability, a third of the X-ray examinations done had been inadequate. Many anaesthetists did not repeat cervical spine X-rays if there were previously performed views available. We showed that a complete X-ray examination of the cervical spine should include flexion and extension stress views in addition to frontal views of the odontoid and entire cervical spine. Anterior atlantoaxial subluxation was the most common subluxation encountered in our study population. The detection of cervical spine instability was found to significantly affect anaesthetic management, favouring techniques that avoided unprotected manipulations of the neck under anaesthesia.

Tracheostomy in a neuro-intensive care setting: indications and timing.

A retrospective review was made of 49 survivors who were mechanically ventilated for more than 48 hours in the neurosurgical ICU. Thirty-two patients (Gp I) were successfully extubated, 9 patients (Gp II) underwent tracheostomy after one or more failed extubations, and 8 patients (Gp III) underwent elective tracheostomy. Glasgow Coma Scale (GCS) scores at extubation were 11.3 +/- 2.8 (mean (SD) for Gp I vs 7.8 +/- 2.7 for Gp II (P = n.s.) and at elective tracheostomy (Gp III) was 5.4 +/- 2.3. Incidence of ventilator-associated pneumonia were 35% in Gp I vs 100% of patients in Gp II and III (P < 0.05). Reasons for reintubation in 7 of 9 patients (Gp II) were upper airway obstruction and tenacious tracheal secretions while 14 of 17 patients were weaned off the ventilator within 48 hours of tracheostomy. The length of stay in ICU was 16.8 +/- 7.1 days in Gp II vs 11.7 +/- 2.9 days in Gp III (P < 0.05). In our study, elective tracheostomy for selected patients with poor GCS scores and nosocomial pneumonia has resulted in shortened ICU length of stay and rapid weaning from ventilatory support.

Candida mediastinitis and septic shock following occult esophageal perforation in a patient with posttraumatic quadriplegia.

Unexplained septic shock was ultimately shown to be caused by Candida mediastinitis after perforation of the cervical esophagus by a dislodged cervical methylmethacrylate construct in a 25-year-old patient with traumatic quadriplegia. Communication between the prevertebral abscess and pleural space further led to the formation of a esophageal-pleural-cervical fistula. Despite antibiotics, surgical removal of the construct, and drainage of the esophagus and mediastinum, the patient died from refractory shock and respiratory failure.

Venous air embolism during cesarean section: more common than previously thought.

To confirm the incidence of venous air embolism (VAE), we simultaneously monitored expired nitrogen concentration (FEN2) and precordial Doppler in 30 patients undergoing cesarean delivery during general anesthesia. Patients were randomized into two groups and the effect of a 10 degrees reverse Trendelenburg tilt versus the horizontal position was evaluated. Forty-two episodes of VAE, defined by an increase in FEN2 of 0.1%, were detected in 97% (29/30) of patients. Doppler ultrasound failed to detect 9 of the episodes and 23 (41%) changes in Doppler tones were not associated with an increase in FEN2. These spurious Doppler signals were synchronous with compression of retroperitoneal structures, suggesting turbulent venous return to be the cause. The reverse Trendelenburg position did not reduce the incidence of VAE. Compared with the preinduction baseline, mean arterial blood pressure decreased by 31.5 +/- 10.1 mm Hg in this position and 22.6 +/- 13.4 mm Hg in the supine position after hysterotomy. We conclude that VAE in cesarean delivery during general anesthesia occurs very frequently, and that changes in Doppler tones may not be reliable indicators of this complication. Measures to reduce the size and effect of air emboli therefore should be applied routinely in all patients.

Delivery of constant air-oxygen mixtures using a closed circle absorber system.

The aim of this study was to devise and validate a technique to deliver constant air-oxygen mixtures from a standard anaesthetic machine using only oxygen as the compressed gas source. The common gas outlet was modified to allow measured quantities of ambient air to be insufflated via a three-way attachment into a closed circle absorber system with a double-circuit collapsible bellows ventilator. During positive pressure ventilation, leakages of between 50-150 ml.min-1 occur from the circuit and nomograms of the minimal air and oxygen flow rates needed to maintain constant oxygen concentrations in the presence of the leaks were then mathematically derived. The accuracy of the nomograms was tested on three different anaesthetic machines using test lung models. There were no differences observed among the mean oxygen concentrations using the three machines. Pooled mean values (SD) of 30.65% (0.77), 51.07% (1.04) and 70.4% (0.73) were obtained for predicted inspired concentrations of 30, 50 and 70% respectively. Next, the technique was studied on 18 patients who underwent isoflurane or propofol anaesthesia (duration 40-210 min) for various surgical procedures. Pooled mean values (SD) obtained were 29.3% (1.86), 40.95% (1.65) and 50.06% (1.41) respectively for predicted oxygen concentrations of 30, 40 and 50% respectively. We conclude that this technique can be used to deliver constant air-oxygen mixtures accurately during inhalational or total intravenous anaesthesia when N2O is contraindicated but a source of compressed air is not readily available.