Case Discussions for ICM training

Dr PB Sherren

ICM and Anaesthetic Specialist trainee

2012

ContentsCase 1: Blunt traumatic pericardial rupture and cardiac herniation

Case 2: Pulmonary interstitial emphysema

Case 3: Traumatic brain injury and multimodal monitoring

Case 4: Acute Traumatic coagulopathy and damage control resuscitation

Case 5: Sudden cardiac death and commotio cordis

Case 6: Renal replacement therapy and dosing

Case 7: Hyperbaric oxygen therapy in carbon monoxide poisoning

Case 8: Guillain Barré Syndrome and immunomodulation

Case 9: Extra Corporeal Membrane Oxygenation in Respiratory Failure

Case 10: New and novel strategies for managing propranolol overdose

Case 1: Blunt traumatic pericardial rupture and cardiac herniation

Introduction

Cardiac herniation is a significant and potentially fatal complication of blunt traumatic pericardial rupture (BTPR). Despite 60 years of experience, it is an injury that frequently results in pre/early hospital death and diagnosis at autopsy, probably owing to a combination of diagnostic difficulties, lack of familiarity and associated polytrauma. Of those who make it to hospital, and are later diagnosed with BTPR, the survival rate is 36.4% - 42.9%. The common issue is that of a missed or delayed diagnosis. Potentially, with increasing awareness of the injury and improved use and availability of imaging modalities, survival rates will improve.

Clinical problem and relevant management

A 21-year-old male was admitted to a district general hospital emergency department following a high speed motorbike accident. On admission he was resuscitated according to standard ATLS principles. Initially he was noted to have a high alveolar-arterial gradient but was cardiovascularly stable. He was conscious with a GCS of 15/15. He was intubated and ventilated due to respiratory distress and high alveolar-arterial gradient. Following intubation, he become progressively more haemodynamically unstable and was transferred to our trauma centre. By the time of arrival the patient's condition deteriorated; on arrival in our department, he was found to be on a FiO2 of 1.0 with PaO2 around 10 kPa and requiring high dose norepinephrine and epinephrine to sustain his mean arterial pressure. He was re-trauma called at this stage and plain radiographs were obtained to further ascertain and clarify his injuries (Figure 1).

Figure 1. Plain supine AP chest radiograph showing a prominent, right-sided cardiac silhouette ('boot shaped'), bilateral pulmonary contusions, rib fractures, and endotracheal and tube thoracostomies.

The finding of dextrocardia had been noted previously at the district general hospital and was not thought to be pathological at this stage. A further tube thoracostomy did not improve the hemodynamic status of the patient. The patient was transferred for CT scan where the following images were obtained (Figure 2).

Figure 2. Axial chest CT demonstrating multiple parenchymal lung contusions, collapsed bilateral haemopneumothoraces, tube thoracostomies, surgical emphysema, large left-sided pneumopericardium, and displacement of the heart into the right hemithorax.

The CT showed a multitude of head and thoracic injuries, as well as a number of rib fractures and bilateral haemopneumothacaces. The presence of pericardial air with herniation of the heart into the right hemithorax was also causing concern. At this stage the patient's condition had not improved and it was agreed to take the patient to theatre to investigate his thoracic injuries. The patient underwent a clamshell thoracotomy where a 10 cm tear in the right of the pericardium was noted with a cardiac herniation through the defect. The heart was relocated and the pericardium repaired with interrupted non-absorbable sutures.

There was an almost immediate reduction in inotrope requirement and the patient was transferred to ICU. His post-operative care was complicated by a chest infection and frequent episodes of fast atrial fibrillation secondary to a myocardial contusion, requiring DC cardioversion. He was discharged from ICU after 14 days and made a full recovery.

Discussion

Cardiac herniation can occur when there is a significant defect within the pericardial sac. Pericardial tears may involve either the superior, left or right pleuropericardium, or the diaphragmatic pericardium [1]. The defect can allow cardiac luxation and, in the case of diaphragmatic pericardial tear, herniation of abdominal contents into the pericardial sac. Defects of the pleuropericardium usually occur vertically along the phrenic nerve and if the tear is large enough, approaching 8-12 cm, the heart can sublux through the defect [1]. The resulting torsion of the great vessels can lead to a form of obstructive cardiogenic shock and cardiovascular instability [1].

As seen with our own experience and that of others, there is often a delay in diagnosis of BTPR and cardiac herniation, which is a real concern given that, once recognised, the treatment is simple and effective [2]. Road traffic collisions and sudden decelerations are the most common mechanisms of injury, particularly those involving a vector of injury from the left side of the chest [1]. The following pattern of associated injuries should also arouse suspicion of BTPR [1]: cardiac contusions and dysrhythmias (28%), multiple rib fractures, haemopneumothoraces, pulmonary contusions, abdominal injuries (27%), pelvic/long bone fractures (49%), spinal cord and traumatic brain injuries (32%).

Given the severity of associated injuries, patients usually require invasive ventilation early on. However, if the patient is conscious, they may report symptoms of palpitations, shortness of breath and chest pain as well as angina type pains as a result of coronary obstruction following herniation [3-5].

The main clinical signs, which may be subtle include:

Signs similar to that of tamponade; in particular that of hypotension, pulsus paradoxus and raised jugular venous pressure (JVP). This may occur early or late depending on the timing of herniation. This haemodynamic compromise may manifest itself despite fluid administration and inotropic support.

Fluctuating haemodynamic parameters, sometimes to the extent of sudden cardiac arrest (often as a result of change in patient's position), should evoke a high index of suspicion of BTPR.

Tachycardia and dysrhythmias may also be seen, such as the atrial tachyarrhythmias noted in our case.

Displaced and heaving apex beat.

A splashing murmur "bruit de Moulin" as a result of the heart moving in a haemopneumopericardium.

There are a multitude of investigations available to most hospitals that can assist in the diagnosis including electrocardiogram, chest radiograph, echocardiography, computerised tomography (CT) and magnetic resonance imaging. Along with its increasing availability and use in the multiply injured trauma patients, CT is also more sensitive for identifying cardiac axis changes and pericardial discontinuity than

plain radiographs [6,7]. Characteristic changes indicating a pericardial rupture include [6,7]:

Focal pericardial dimpling and discontinuity.

Pneumopericardium.

Interposition of lung between: aorta and pulmonary artery; or heart and diaphragm; or right atrium and right ventricular outflow tract.

Characteristic changes for a cardiac herniation include.

"Empty pericardial sac" sign, air outlining the empty pleuropericardium as a result of cardiac luxution into the hemithorax.

"Collar" sign is the result of compression of the cardiac contour as a result of constriction by the pericardial band caused by the defect.

Associated signs include dilated inferior vena cava (IVC), reflux of contrast into IVC and deformed ventricular silhouette, as well as, secondary signs of tamponade periportal lymphoedema, pericholecystic fluid and ascites.

Once BTPR and cardiac herniation has been diagnosed, treatment is simple and effective. It has even been suggested that, as it is such a rapidly reversible cause of sudden cardiac arrest, there may be a role for post-arrest emergency thoracotomy for select patient groups with blunt chest trauma and positional cardiovascular instability [5]. The treatment of choice for tears of the diaphragmatic pericardium, right pleuropericardium, and moderate/large left pleuropericardium defects, is surgical closure [1,5]. Closure of moderate-sized pericardial defects is best achieved by interrupted non-absorbable sutures and larger ones with a mesh prosthesis [1].

Learning points

BTPR and cardiac herniation is a complex and often fatal injury that usually presents under the umbrella of polytrauma. Patients with blunt chest trauma and any of the following signs are exceptionally high risk for BTPR and the need for an urgent operative intervention should be considered:

Cardiovascular instability with no obvious cause. This instability may be labile and mimic cardiac tamponade, particularly with changes in patient position. A bedside TTE in this setting is a vital tool for exclusion of differential pathology.

A prominent, possibly displaced, cardiac silhouette and asymmetrical large volume pneumopericardium. These signs may show varying degrees of prominence on the plain chest radiograph, if there is uncertainty and the patient's condition allows, a chest CT should be sought as it has been shown to better delineate the injuries.

References

1. Clark DE, Wiles III CS, Lim MK, et al: Traumatic rupture of the pericardium. Surgery 1983: 93; 495-503.

2. Janson JT,Harris DJ, Pretorius J, et al. Pericardial rupture and cardiac herniation after blunt chest trauma Ann Thorac Surg 2003: 75; 581-582

3. Wright MP, Nelson C, Johnson AM, Mcmillan AKR. Herniation of the heart. Thorax 1970: 25; 656-666.

4. Chughtai T, Chiavaras MM, Sharkey P, et al. Pericardial rupture with cardiac herniation. Can J Surg. 2008: 51(5); E101–E102

5. Wall MJ Mattox KL, Wolf DA. The Cardiac Pendulum: Blunt Rupture of the Pericardium with Strangulation of the Heart. The Journal of Trauma Injury, Infection and Critical care. 2005: 59(1); 136-142.

6. Nassiri N, Yu A, Statkus N, et al. Imaging of Cardiac herniation in Traumatic pericardial rupture. Journal of Thoracic Imaging. 2009: Vol 24(1); 69-72.

7. Wielenberg AJ, Demos TC, Luchette FA, et al. Cardiac Herniation Due to Blunt Trauma: Early Diagnosis Facilitated by CT. AJR 2006: 187; W239-W240

Case 2: Pulmonary Interstitial Emphysema and Acute Respiratory Distress Syndrome

Introduction

Pulmonary interstitial emphysema (PIE) is a barotrauma-related life-threatening condition, not uncommon to the neonatologist caring for pre-term babies. For the intensivist, despite being confronted by significant compliance issues resulting from the fibroproliferative phase of adult respiratory distress syndrome (ARDS) on a daily basis, PIE in the critically ill adult is an extremely rare occurrence.

Clinical problem and relevant management

An 87-year-old Caucasian British woman presented to our emergency department with a three-day history of shortness of breath, pyrexia and non-productive cough. Her only significant past medical history was well-controlled hypertension. She was independent in her daily activities, did not smoke cigarettes and reported a good cardiorespiratory reserve prior to the onset of symptoms. The diagnosis of community-acquired multilobar pneumonia was made with a CURB-65 score of three.

She was admitted to the high dependency unit with type 1 respiratory failure and a high alveolar-arterial oxygen gradient. She received intravenous antibiotics (piperacillin/tazobactam and erythromycin) and non-invasive high-flow continuous positive airway pressure (CPAP). By the fourth day, the patient had deteriorated, with a chest radiograph showing bilateral alveolar and interstitial infiltrates and a ratio of partial pressure of arterial oxygen to the fraction of inspired oxygen (PaO2/FiO2) of < 26.6 kPa, which required invasive ventilation. The criteria for ARDS were met and the cause was thought to be a combination of the direct pneumonic pulmonary injury and extrapulmonary severe sepsis. With protective lung ventilation, low dose methylprednisolone, antibiotic therapy and a negative fluid balance, gradual improvement was made; over the following five days respiratory parameters weaned sufficiently to allow an uncomplicated percutaneous tracheostomy to be performed.

On day 10, a period of desaturation required recruitment manoeuvres with a Mapleson C circuit that resulted in notable surgical emphysema. The cause was thought to be a tracheal injury sustained at the time of tracheostomy insertion. An adjustable flange tube was positioned under bronchscopic guidance just proximal to the carina in an attempt to limit any further tracking of air through the potential tracheal defect. Despite these measures, a high alveolar-arterial oxygen gradient and peak airway pressures persisted. A chest radiograph showed more homogenous central pulmonary alveolar shadowing. An upper airway bronchoscopy showed no obvious tracheal wall injury and computed tomography (CT) of her chest showed extensive surgical emphysema and a small anterior left sided pneumothorax. On further review of the CT scan, it was felt that the perivascular and peribronchial emphysema was consistent with a diagnosis of pulmonary interstitial emphysema

(Figure 1). Over the following days, despite protective ventilatory strategies and intercostal tube thoracostomy, lung compliance along with oxygenation deteriorated. By day 13, the deteriorating respiratory parameters along with inotropic requirements resulted in a decision to limit therapy and the patient died on day 14.

.

Figure 1. Axial CT image of the chest demonstrating classic findings of tracking perivascular and peribronchial emphysema

Discussion

In ARDS the inflammatory injury to the lungs begins with an early exudative phase characterised by an exaggerated inflammatory response. The alveolar macrophages secrete cytokines, interleukins and tumour necrosis factor-α, which initiate chemotaxis and activate neutrophils. The influx of pro-inflammatory mediators results in the disruption of the alveolar basement membrane and the influx of proteinaceous fluid into the alveoli space with impairment of surfactant function. If the inflammation is allowed to continue unchecked, a fibroproliferative stage develops and results in type II pneumocyte and fibroblast proliferation with associated collagen deposition. The late fibrotic phase of ARDS is categorised by embryonic mesenchymal cell proliferation, neovascularisation and intra-alveolar fibrosis.

Acute Lung injury (ALI) and ARDS was first defined and stratified by the American-European Consensus Conference (AECC) in the 1990s. Issues pertaining to the reliability and validity of this definition have arisen since and this has resulted in a revised Berlin definition being developed 2011. The revised Berlin Definition for ARDS was developed in an attempt to address these limitations based on a consensus process involving a panel of experts [1].

Table 1. The differences between the AECC ALI/ARDS and the revised Berlin definition for ARDS [1].

Table 2. The Berlin definition of ARDS [1].

The incidence of PIE in pre-term infants requiring ventilation for respiratory distress syndrome to be around 19.5%, with a mortality rate of 24% [2]. PIE almost exclusively occurs as a result of intermittent positive pressure ventilation (IPPV) with peak airway pressures exceeding 30 cm H2O [3]. When high airway pressures and dramatic shearing forces are applied to a non-compliant lung unit, the result is alveolar duct rupture, usually at the terminal bronchiole-alveolar junction [4]. This allows air to escape into the connective tissue of the peribronchovascular sheaths, interlobular septa and visceral pleura, occasionally migrating into the lymphatic and venous circulation [4].

Pre-term babies are particularly prone to PIE, because of the high shearing forces and airway pressures required to re-recruit lung units with collapsing pressure exceeding their functional residual capacity, secondary to reduced surfactant levels [4]. Other postulated risks for PIE include increased amount of pulmonary connective tissue or a sudden reduction in extravascular lung water, which may offer a degree of protection against tracking interstitial emphysema.

The poor lung compliance associated with ARDS was pivotal in our case, but a reduction in extravascular lung water also perhaps had a role to play in the development of PIE. Interstitial emphysema has a number of potentially detrimental sequelae [1,5]. These include: compression atelectasis of adjacent healthy lung and resulting intrapulmonary shunt which is worsened by recruitment manoeuvres; compression of surrounding pulmonary vasculature; and decompression of interstitial blebs into surrounding spaces, potentially resulting in pneumomediastinum, pneumothorax, pneumopericardium, pneumoperitoneum and surgical emphysema. Although all the above can be very difficult to manage in a critically ill patient, the addition of a pneumothorax to PIE alone doubles the mortality [1].

Chest radiograph findings are often very subtle, and identification, given the frequent presence of overlying dense alveolar shadowing as a result of the lung injury and exudative processes, make the diagnosis difficult. However, the following findings may sometimes be distinguishable [5-7]: parenchymal stippling; lucent mottling and streaking extending to the mediastinum; perivascular halos (from perivascular air collections); subpleural cysts; lucent bands; and parenchymal cysts or bullae. CT is a more sensitive tool for delineating the pathology, and the classic findings of tracking perivascular and peribronchial emphysema [8] were both demonstrated in our case (Figure 1).

The chosen treatment will, to an extent, depend on the distribution of the disease along with the severity and complications. The mainstay of treatment is to achieve adequate oxygenation with lower mean and peak airway pressures, hence minimizing interstitial leak through the defects [9]. This technique of protective lung ventilation and permissive hypercapnia is a familiar one to intensivists trying to avoid the many ramifications of volutrauma and barotrauma. There are a number of other therapeutic options that may be considered. Lateral decubitus positioning with the affected lung in the dependent position can be tried as an early conservative

approach, encouraging plugging of the dependent lung. This is only of benefit when the disease is localized. Selective main bronchial intubation and occlusion can be useful, although only for unilateral disease. High-frequency ventilation (high frequency jet ventilation or high frequency oscillatory ventilation) and extracorporeal membrane oxygenation have all been used effectively [1,3-5,9].

Beyond these mainstays of treatment, there have been some case reports and series regarding the use of steroids and surgical resection for persistent localized disease, but such therapies have not established a good evidence base as yet.

Learning points

The development of PIE is a rare but real risk when caring for patients with ARDS and worsening lung compliance.

When undertaking recruitment manoeuvres and interpreting peak airway pressures, it is important to remember the differential lung time constants encountered throughout the diseased lungs.

These variations result in an uneven distribution of pressure across the alveoli, and areas of lung with long time constants are at high risk of barotrauma.

It is these areas that are at particular risk of developing PIE when exposed to the shearing forces experienced during IPPV.

References

1. ARDS Definition Task Force, Ranieri VM, Rubenfeld GD, Thompson BT, et al. Acute respiratory distress syndrome: the Berlin Definition. JAMA. 2012 Jun 20;307(23):2526-33.

2. Greenough A, Dixon AK, Roberton NRC. Pulmonary Interstitial Emphysema. Arch Dis Child 1984, 59(11):1046-1051.

3. Woodring JH. Pulmonary interstitial emphysema in the adult respiratory distress syndrome. Crit Care Med 1985, 13(10):786-791.

4. Plenat F, Vert P, Didier F, et al. Pulmonary interstitial emphysema. Clin Perinatol 1978, 5(2):351-75.

5. Kemper AC, Steinberg KP, Stern EJ. Pulmonary interstitial emphysema: CT findings. AJR 1999, 172(6):1642.

6. Unger JM, England DM, Bogust JA. Interstitial emphysema in adults: recognition and prognostic implications. J Thorac Imaging 1989, 4(1):86-94.

7. Westcott JL, Cole SR. Interstitial pulmonary emphysema in children and adults: roentgenographic features. Radiology 1974, 111(2):367-378.

8. Satoh K, Kobayashi T, Kawase Y, et al. CT appearance of interstitial pulmonary emphysema. J Thorac Imaging 1995, 11(2):153-154.

9. Frantz ID, Stark AR, Westhammer J. Improvement in pulmonary interstitial emphysema with high frequency ventilation. Pediatr Res 1981, 15:719.

Case 3: Traumatic brain injury and multimodal monitoring.

Introduction

Traumatic brain injury remains a common and often debilitating event across the world, producing significant burdens upon health and social care. Effective neurocritical care coupled with timely and appropriate neurosurgical intervention can produce significant improvements in patient outcome.

Clinical problem and relevant management

A 39 year old male was admitted to the Royal London Hospital following an assault and isolated closed head injury. His initial Glasgow Coma Score (GCS) with the pre-hospital team (HEMS) prior to rapid sequence intubation was 6/15 (E1V2M4). He had no lateralising or pupillary signs on initial examination by the HEMS. On arrival in the emergency department he had no cardiorespiratory issues on his primary survey. When reassessing his disability on admission it was noted his right pupil was enlarged (5mm) and unreactive with a normal left pupil. Adequate sedation and analgesia was achieved with an infusion of propofol 1% and intermittent boluses of Fentanyl. A bolus of 200ml of 20% Mannitol was administered and standard neuroprotective measures were implemented prior to transfer for a CT scan. The CT head showed a large right-sided subdural haematoma with 8mm of midline shift. He had an immediate craniotomy and evacuation of haematoma where it was noted there was significant underlying parietal contusions. A Licox catheter (Integra Neurosciences) was placed in the surrounding healthy parenchyma and the patient was transferred to the ICU. During the initial 48 hours the intracranial pressures (ICP) were extremely labile despite adequate PaO2, PaCO2, cerebral perfusion pressures (CPP), sedation, analgesia and anticonvulsants. A repeat CT scan showed multiple haemorrhagic contusions with patent basal cisterns but no surgically amenable lesion. Consideration was given to a hemispheric craniectomy but instead a decision was taken to continue maximal medical management and target a PbtO2 of greater than 20mmHg on the Licox while accepting higher ICPs. After 5 days the ICP had stabilised and remained stable following a CO2 challenge. The patient was woken and extubated uneventfully. On discharge to the neurosurgical ward he had a GCS of 14/15 (E3V5M6) with no gross neurological deficit. However, he did have a mild cognitive deficit with short term memory issues.

Discussion

A sustained elevation in ICP is known to adversely affect patient outcomes [1]. Measurement usually requires the presence of an invasive indwelling catheter within the subdural space, parenchyma, or ventricular system. However, other none-invasive methods such as near infra-red spectroscopy (NIRS), transcranial doppler (TCD) and ultrasound (US) assessment of the optic nerve sheath diameter (ONSD) have shown promise as screening tools for extra-axial haematomas [2] and raised ICP [3-5].

The importance of ICP (<20-25mmHg) and CPP (60-70mmHg) control and their incorporation into evidence based guidelines for the management TBI is well-established [6,7]. However, guidelines are designed for a homogenous population which does not translate well to the diversity seen in TBI.

The incidence and duration of impaired cerebral pressure autoregulation (CPA) in TBI is variable [8,9]. Peterson et al showed, contrary to the popular belief, that static CPA was intact in 75.7% of patients with severe TBI [8]. The role of static CPA and perfusion CT scans in the management of severe TBI remains to be elucidated. Dynamic CPA relates to the cerebrovasculature response to transient fluctuations in CPP [10] and can be assessed with pressure reactivity index (PRx) [10], pressure volume index (PVI) [11] or the use of TCD to assess the mean flow velocity index

[12]. Uncertainty still remains over the clinical use of CPA and whether there is an association between impaired CPA and mortality [13,14]. Johnson et al suggested that patients with impaired CPA and lower targeted CPP (<60mmHg) had better outcomes [15]. This interesting study, although thought-provoking, was limited by its small sample size and that it was a non-randomised controlled study. The future of individualised targeting of CPP on the basis of CPA in TBI is continuously evolving and is now awaiting multicentre randomised control trials.

With the use of magnetic resonance or positron emission tomography imaging, it has been identified that significant secondary neuronal injury can occur in the absence of any pathological changes in ICP or CPP. Hence there is an increasing interest in multimodal monitoring to identify early markers of secondary cerebral insults. When addressing raised ICP, optimisation of the CBF, cerebral metabolism, CBF-metabolism coupling, immunological and biochemical markers may help mitigate any secondary cerebral insult. A variety of monitoring devices are commercially available to assist, including Laser Doppler flowmetry, thermal diffusion flowmetry, NIRS, brain tissue oxygen tension (pbtO2), jugular venous oximetry, TCD and cerebral microdialysis (MD). In TBI, the use of pbtO2 (LICOX) and MD have proved the most promising in recent literature.

The brain parenchyma is dependent on a continuous supply of oxygen to maintain aerobic metabolism and cellular integrity. In TBI the oxygen supply and demand varies dramatically and cerebral hypoxia is common. To measure pbtO2 in vivo, a mini Clark electrode can be inserted into the parenchyma. The electrode can be placed in normal white matter on the side of maximal pathology near the injury, or in the non-dominant frontal lobe in diffuse injury. The threshold for critical ischemia is usually considered to be 10mmHg [15]. The integration of PbtO2 guided therapies into existing CPP/ICP based protocols has been shown to reduce mortality rates and improve patient outcomes following severe TBI [15]. Spiotta et al demonstrated that the use PbtO2 guided management (PbtO2 >20mmHg) enabled clinicians to tolerate higher ICPs and avoid detrimental side effects of ICP and CPP management [15].

While the efficacy of early craniotomy and evacuation of extra-axial bleeds with mass effect is well established, the use of decompressive craniectomy remains

controversial. Decompressive craniectomy may be performed prophylactically at the time of removal of mass lesions or as a rescue technique when maximal medical therapy has failed; although this reduces ICP [16], as yet no high-quality evidence demonstrates an improved clinical outcomes.

The recent DECRA trial examined the role of early (within 72 hours) decompressive craniectomy in patients with severe TBI and a refractory intracranial hypertension (greater than 20mmHg) in excess of 15 minutes [17]. In this large, multicentre trial, patients were randomly assigned to either standard care or bifronto-temporo-parietal craniectomy. In keeping with previous literature, patients undergoing decompression exhibited significantly lower ICPs and reduced ICU length of stay. However, functional outcome was unfavourable. There are limitations to the generalisation of these observations. Only 155 patients were recruited over a seven year period from a potential 3478 patients, suggesting a highly specific sample population. It could also be argued that an inclusion threshold of refractory intracranial hypertension lasting 15 minutes was too short.

A number of these concerns may be addressed when the results of the ongoing RESCUEicp trial are made available [18]. This randomised controlled study is evaluating the place of decompressive craniectomy in the management of refractory intracranial hypertension defined as an ICP greater than 25 mmHg for 1 to 12 hours at any point following injury. Until then, it appears the role of surgery in the management of brain ischaemia is limited to focal lesions.

Learning points

In TBI there is a variable incidence of impaired CBF autoregulation.

Multimodal neurological monitoring and individualised management strategies should be employed in TBI.

Individualisation of cerebral perfusion targets against brain-tissue oxygenation parameters and markers of anaerobic metabolism may allow clinicians to accept higher intracranial pressure limits.

The evidence base for the use of PbtO2 is growing. In conjunction with traditional monitoring and appropriate therapeutic interventions, PbtO2 may improve outcomes in TBI.

Further trials are required to delineate the role of decompressive craniectomy in TBI.

References

1. Metzger JC, Eastman AL, Pepe PE. Year in review 2008: Critical Care--trauma. Crit Care. 2009;13(5):226. Epub 2009 Oct 21.

2. Leon-Carrion J, Dominguez-Roldan JM, Leon-Dominguez U, et al. The Infrascanner, a handheld device for screening in situ for the presence of brain haematomas. Brain Injury, September 2010; 24(10): 1193–1201.

3. Rajajee V, Vanaman M, Fletcher JJ, et al. Optic Nerve Ultrasound for the Detection of Raised Intracranial Pressure. Neurocrit Care. 2011 Jul 19. [Epub ahead of print]

4. Major R, Girling S, Boyle A. Ultrasound measurement of optic nerve sheath diameter in patients with a clinical suspicion of raised intracranial pressure. Emerg Med J. 2011 Aug;28(8):679-81. Epub 2010 Aug 15.

5. Sharma D, Souter MJ, Moore AE, et al. Clinical experience with transcranial Doppler ultrasonography as a confirmatory test for brain death: a retrospective analysis. Neurocrit Care 2010. [Epub ahead of print]

6. Farahvar A, Gerber LM, Chiu YL, et al. Response to intracranial hypertension treatment as a predictor of death in patients with severe traumatic brain injury. J Neurosurg. May 2011;114:1471–1478.

7. Zeng T, Gao L. Management of patients with severe traumatic brain injury guided by intraventricular intracranial pressure monitoring: a report of 136 cases. Chinese Journal of Traumatology. 2010; 13(3):146-151.

8. Peterson E, Chesnut RM. Static autoregulation is intact in majority of patients with severe traumatic brain injury. J Trauma. 2009 Nov;67(5):944-9.

9. Sviri GE, Aaslid R, Douville CM, et al. Time course for autoregulation recovery following severe traumatic brain injury. J Neurosurg. 2009 Oct;111(4):695-700.

10.Consonni F, Abate MG, Galli D, et al. Feasibility of a continuous computerized monitoring of cerebral autoregulation in neurointensive care. Neurocrit Care. 2009;10(2):232-40. Epub 2008 Oct 16.

11.Lavinio A, Rasulo FA, De Peri E, et al. The relationship between the intracranial pressure–volume index and cerebral autoregulation. Intensive Care Med 2009; 35:546–549

12.Sorrentino E, Budohoski KP, Kasprowicz M, et al. Critical thresholds for transcranial doppler indices of cerebral autoregulation in traumatic brain injury. Neurocrit Care. 2011 Apr;14(2):188-93.

13.Zweifel C, Lavinio A, Steiner LA, et al. Continuous monitoring of cerebrovascular pressure reactivity in patients with head injury. Neurosurg Focus. 2008;25(4):E2.

14.Johnson U, Nilsson P, Ronne-Engstro E, et al. Favorable Outcome in Traumatic Brain Injury Patients With Impaired Cerebral Pressure Autoregulation When Treated at Low Cerebral Perfusion Pressure Levels. Neurosurgery. 2011 Mar;68(3):714-21; discussion 721-2.

15.Spiotta AM, Stiefel, MF, Gracias VH, et al. Brain tissue oxygen–directed management and outcome in patients with severe traumatic brain injury. J Neurosurg Sep 2010;113:571–580.

16.Taylor A, Butt W, Rosenfeld J, et al. A randomized trial of very early decompressive craniectomy in children with traumatic brain injury and sustained intracranial hypertension. Child's Nervous System. 2001 Feb. 23;17(3):154–162.

17.Cooper DJ, Rosenfeld JV, Murray L, et al. Decompressive Craniectomy in Diffuse Traumatic Brain Injury. N Engl J Med. 2011 Apr. 21;364(16):1493–1502.

18.Hutchinson PJ, Menon DK, Kirkpatrick PJ. Decompressive craniectomy in traumatic brain injury - time for randomised trials? Acta Neurochir. 2004 Nov. 2;147(1):1–3.

Case 4: Acute traumatic coagulopathy and damage control resuscitation

Introduction

Uncontrolled haemorrhage is the most common cause of potentially preventable death in trauma patients. The aetiology of trauma induced coagulopathy is complicated but awareness and timely intervention with damage control resuscitation could help improve patient outcomes.

Clinical problem and relevant management

A 40 year old male was admitted to our trauma centre following a high speed motor bike collision. The physician-staffed helicopter emergency medical service (HEMS) attended the patient initially. The injuries noted by HEMS on arrival included a right complete traumatic forequarter amputation with significant exsanguination, right pneumothorax, pelvic injury and closed head injury with a GCS 7/15 (E1V2M4). The patient was in cardiorespiratory extremis and required an emergent rapid sequence induction, right open thoracostomy, direct wound compression, sam pelvic sling, 1g of tranexamic acid, 250ml 7.5% saline and 500ml of hartmans solution to maintain radial pulses. Despite these interventions there were persistent volume issues and a massive transfusion pre-alert was ordered. The primary survey on arrival in the trauma centre showed a stable airway and breathing system, but despite on-going haemostatic resuscitation with a Level 1 rapid transfusor, the haemodynamic instability persisted with a severe lactic acidaemia (pH 6.8 and lactate 16) and negative E-FAST scan. Other investigations of note on arrival in the ED included an INR and APTTR of 2.6 and 2.1 respectively. Given the ongoing transfusion requirements and haemodynamic instability the patient went straight to theatres for surgical haemostasis. During the damage control surgery a massive transfusion was required targeting systolic blood pressure of approximately 80mmHg. A helical CT scan from the head through to pelvis was undertaken prior to transfer to the ICU. On arrival in the ICU the patient had received :

23 units of packed red blood cells, 16 units of fresh frozen plasma, 3 pooled units of platelets and cryoprecipitate, 2g tranexamic acid, 3g Calcium chloride, 2000ml compound sodium lactate

Figure 1. Chest radiograph as part of the primary survey.

Discussion

Hypothermia, acidaemia and coagulopathy or the ‘lethal triad’, is a well described entity in the trauma population and is associated with significant mortality [1]. A traumatic insult, coupled with systemic hypoperfusion results in a decreased oxygen delivery, a shift to anaerobic metabolism, lactate production and metabolic acidaemia. Common instigators of hypothermia in trauma include exposure, massive cold fluid resuscitation and impaired endogenous heat production as a result of anaerobic metabolism .Traditionally the aetiology of a trauma induced coagulopathy was thought to be multifactorial and involve the lethal triad, dilutional coagulopathy, pre-existing bleeding diathesis and disseminated intravascular coagulation (Figure 2).

Figure 2. A diagram showing some of the mechanisms leading to coagulopathy in the injured.

Interestingly in the case described here, a significant early coagulopathy was evident on arrival in the ED despite a short pre-hospital time, minimal fluid resuscitation and normothermia. In 2003, Brohi et al showed that around 25% of severely injured trauma patients present to hospital with a significant coagulopathy which is unrelated to fluid administration [2,3]. The early coagulopathy has become known as the Acute Traumatic coagulopathy (ATC) or Acute Coagulopathy of Trauma Shock (ACoTS). It is associated with an increase in transfusion requirements, injury severity scores, organ dysfunction and mortality rates [2-5].

ATC is an impairment of haemostasis involving a dynamic interaction between endogenous anticoagulants and fibrinolysis that is initiated immediately after an injury [5]. ATC is driven by an endothelial injury and hypoperfusion, which results in in increased thrombomodulin expression and activation of protein C (Figure 3). The inhibitory effect of activated protein C on clotting factors V/VIII and plasminogen activator inhibitor-1 (PAI-1) would appear key in the development of ATC [5,6].

Figure 3. Expression of thrombomodulin following a traumatic injury results in increased activation of protein C with resulting impairment of clotting factors V/VIII and reduction in thrombin generation. Activated Protein C also has an inhibitory effect on PAI-1 which results in unregulated tPA activity and fibrinolysis.

Damage control resuscitation (DCR) describes a package of care for the haemorrhaging trauma patient. It involves early damage control surgery, haemostatic resuscitation and permissive hypotension. DCR aims to control haemorrhage early while aggressively targeting the ATC and lethal triad. DCR has emerged as the accepted standard of care and some observational studies have suggested a survival benefit [6].

The priority for any haemorrhaging trauma patient is good haemostasis. Unstable patients with major trauma do not tolerate prolonged definitive surgery and hence the emergence of damage control surgery. The aim of damage control surgery is to normalise physiology at the expense of anatomy. The priorities are:

• Stop haemorrhage (Packing, clamping, resection +/- interventional radiology)• Minimise contamination • Limb saving procedures• Good wash out of cavities• Drains and low threshold for Laparostomy• Definitive surgery another day

Haemostatic resuscitation describes the aggressive early use of packed red blood cells, clotting products and coagulation adjuncts in an attempt to mitigate the effects of the ATC and lethal triad in major trauma patients. The exact PRBC:FFP ratio remains unclear, but should ideally be less than 2:1 [7]. In massive transfusions along with appropriate FFP, platelet and fibrinogen supplementation, consideration should be given to early adjunctive therapies such as tranexamic acid [8] while maintaining ionised calcium levels greater than 1.0 mmol/L [9].

Permissive hypotension involves titrated volume resuscitation, which targets a subnormal end point that maintains organ viability until haemorrhage is controlled. By avoiding overzealous fluid resuscitation which targets normotension, the hope is to preserve the first and often best clot. Although permissive hypotension is frequently employed in traumatic haemorrhage, there is really only robust evidence that it is advantageous in penetrating trauma [10]. In blunt trauma there is a relative paucity of good evidence to guide practice, while strong evidence exists for maintaining cerebral perfusion pressures when there are associated head injuries. The end points for resuscitation will depend on age, premorbid autoregulatory state and acute pathology. ‘Rule of thumb’ resuscitation end points include:

Penetrating trauma - maintain cerebration or central pulse or SBP~60mmHg Blunt trauma – maintain radial pulse or SBP >80mmHg Head injury – maintain temporal pulse or SBP >100mmHg Spinal cord injury – Spinal cord perfusion can be improved with

SBP>90mmHg, but no functional outcome data as yet.

DCR is an ever evolving concept, and potential future strategies that are as yet unproven include:

Increasing use of thromboelastometry (TEG/ROTEM) to guide haemostatic resuscitation.

Prothrombin complex concentrate (FII, VII, IX and X) in non-warfarin patients

Fibrinogen complex concentrate (fibrinogen and FXIII) over cryoprecipitate. Alkalising agents such as Tris-hydroxymethyl aminomethane (THAM) in

massive transfusion with severe acidaemia. Novel hybrid resuscitation strategies. High flow/low pressure resuscitation – endothelial resuscitation and

microvascular washout. Suspended Animation. Platelet functional assessment using platelet mapping and aggregometry vs

traditional PF-100.

Learning points

Early coagulation dysfunction is common in trauma patients with haemorrhagic shock.

Tailored management of the ‘lethal triad’ and ATC is essential.

DCR is an emerging standard of care, however, some of its components are pushing the boundaries of what is good evidence based medicine.

References

1. Moore EE. Staged laparotomy for the hypothermia, acidosis, and coagulopathy. Am J Surg 1996;172:405-410.

2. Brohi K, Singh J, Heron M, et al. Acute Traumatic coagulopathy. J Trauma. 2003;54:1127-1130.

3. Davenport R, Manson J, De’Arth H, et al. Functional definition and characterization of acute traumatic coagulopathy. Crit Care Med. 2011;39(12):2652-2658.

4. Maegele M, Lefering R, Yucei N, et al. Polytrauma of the German Trauma Society (DGU). Early coagulopathy in multiple injury: an analysis from the German Trauma Registry on 8724 patients. Injury. 2007 Mar;38(3):298-304.

5. Firth D, Davenport R, Brohi K. Acute traumatic coagulopathy. Curr Opin Anaesthesiol. 2012 Apr;25(2):229-34.

6. Cotton BA, Reddy N, Hatch QM, et al. Damage control resuscitation is associated with a reduction in resuscitation volumes and improvement in survival in 390 damage control laparotomy patients. Ann Surg. 2011 Oct;254(4):598-605.

7. Davenport R, Curry N, Manson J, et al. Hemostatic effects of fresh frozen plasma may be maximal at red cell ratios of 1:2. J Trauma. 2011 Jan;70(1):90-5; discussion 95-6.

8. CRASH-2 collaborators, Roberts I, Shakur H, Afolabi A, et al. The importance of early treatment with tranexamic acid in bleeding trauma patients: an exploratory analysis of the CRASH-2 randomised controlled trial. Lancet. 2011 Mar 26;377(9771):1096-101, 1101.e1-2.

9. Dawes R, Thomas GO. Battlefield resuscitation. Curr Opin Crit Care. 2009 Dec;15(6):527-35

10.Bickell WH, Wall MJ Jr, Pepe PE, et al. Immediate versus delayed fluid resuscitation for hypotensive patients with penetrating torso injuries. N Engl J Med. 1994 Oct 27;331(17):1105-9.

Case 5: Sudden cardiac death and commotio cordis

Introduction

Sudden cardiac death in adolescents and young athletes is a rare occurrence with a large number of differential diagnoses. Early recognition and management is essential to a successful resuscitation and outcome.

Clinical problem and relevant management

A 19 year old gentleman with no previous medical history was brought by ambulance to the Emergency Department resuscitation room following an impact to the central chest in a heavy rugby tackle.

According to bystanders, he was knocked backwards, sustaining a low impact injury to the occiput, and was unconscious from this point. Paramedics arrived at the scene and three-point immobilisation of the cervical spine was instituted. His airway was of concern due to his GCS of 4/15, therefore an oropharyngeal airway was inserted. There was good air entry throughout both lung fields. He was hypotensive with a blood pressure of 64/35 and had a heart rate of 35. ECG monitoring revealed complete heart block. En route to hospital, his rhythm degenerated to ventricular fibrillation, and CPR was started. His rhythm defibrillated to sinus rhythm after one biphasic shock given at 200J, and he began to regain consciousness.

On arrival to hospital, both cardiac arrest and trauma teams were in attendance. At that point, his GCS had increased to 6/15 and he was intubated. Lung fields remained clear and he was cardiovascularly stable with a blood pressure of 135/86, a heart rate of 98 and an ECG showing normal sinus rhythm. Blood glucose level was 7.4mmol/L. Chest and pelvic radiographs did not reveal an abnormality, and a fastscan was negative. After completion of the primary survey, he was transferred to the CT scanner for a CT of the head, neck, chest, abdomen and pelvis, and was subsequently transferred to ICU, where he remained ventilated.

The following day, he was extubated and a secondary survey was performed with no injuries found. He was stepped down to CCU for cardiac monitoring and had no significant elevation in his cardiac enzymes.

In hospital he underwent an echocardiogram revealing a structurally normal heart with normal LV function and negative electrophysiological investigations. On balance it was felt given the minimal trauma, negative investigations and rapid improvement following defibrillation this represented a case of commotio cordis.

He was subsequently discharged from hospital with no cardiological or neurological sequelae.

Discussion

There are a number of differential diagnoses in this clinical scenario, which should include causes of sudden cardiac death (SCD) in young athletes. SCD in young

athletes has an estimated incidence of 1-3/100 000 person years [1]. The risk in males is approximately nine times higher than that in females [1].

Commotio cordis, a disruption to the cardiac rhythm, occurring from a direct impact to the anterior chest wall, although rare, is high in the list of differentials for this patient [2]. The energy transmitted to the myocardium alters ventricular dissipation during repolarisation and triggers ventricular depolarisation, which could result in re-entry and ventricular fibrillation (Figure 1). The phenomenon usually causes ventricular fibrillation in patients with structurally normal hearts even after only a modest force [2,3]. In an experimental model, a single impact over the heart, timed at 10-30ms at the vulnerable point of the repolarisation before the T-wave peak can induce ventricular fibrillation [4]. This is the same principle that governs electrical cardioversion of tachyarrhythmias, and why the direct current must be synchronised to the R wave to avoid inadvertent conversion to ventricular fibrillation via the ‘R on T’ phenomena. Shortening of the cardiac cycle due to tachycardia during exercise may render the heart more susceptible to this phenomenon. It is thought that a blow of 50J is low enough to induce cardiac arrest, and many sporting injuries involve injuries of much higher energy.

Figure 1. Variables and mechanism involved in the development of commotio cordis [8].

Around 10-20 cases are added to the Commotio Cordis Registry per year. It is most common in adolescence but is also seen in older individuals. It must be differentiated

from contusio cordis, a condition of blunt cardiac trauma causing structural cardiac damage, such as myocardial contusions [5].

The majority of those affected by commotio cordis die unless defibrillated within three minutes [6]. It was fortunate in this case that the patient’s initial rhythm was complete heart block, a rhythm that has been described previously in this setting clinically and experimentally, and that his subsequent cardiac arrest was witnessed and acted upon quickly during transfer [4,7]. The increasing prevalence of automated external defibrillators, particularly at large scale sporting events may help to prevent deaths due to this condition.

Alternate causes of sudden cardiac death should be sought and aggressively excluded prior to a diagnosis of commotio cordis being made. Alternate causes of SCD could include undiagnosed coronary artery disease, cardiomyopathies, myocarditis, congenital coronary/aortic anomalies and ion channelopathies including Brugada syndrome. Cross et al published a comprehensive review on sudden cardiac death in adolescents in 2011 [8]. The table below details the incidence of the various aetiologies (Table 1).

Table 1. Underlying heart disease of sudden cardiac death in various studies [8].

The use of implantable cardioverter-defibrillators is controversial in survivors of suspected commotio cordis as it is unlikely to occur twice, but may be prudent if the exact cause of cardiac arrhythmia is uncertain. A low threshold could be wise when you consider a certain number of channelopathies are difficult to diagnose and there may be many variants still to be discovered.

Prevention of blunt chest trauma with chest protectors is controversial, as some believe that it does not dissipate the energy inflicted in cases of commotio cordis. Newer forms of chest protection are under trial [9]. Educating athletes in avoidance of direct chest wall trauma may also help in the prevention of this condition occurring primarily.

Learning points

There are a large number of potential causes of SCD in young healthy individuals.

Commotio cordis and resulting ventricular fibrillation is an extremely rare occurrence that requires immediate defibrillation to maximise the potential for a positive outcome.

References

1. Borjesson M, Pelliccia A. Incidence and aetiology of sudden cardiac death in young athletes: an international perspective. Br J Sports Med. 2009 Sep;43(9):644-8.

2. Maron BJ, Gohman TE, Kyle SB, et al. Clinical profile and spectrum of commotion cordis. JAMA 2002;287:1142–6.

3. Maron BJ, Poliac LC, Kaplan JA, et al. Blunt impact to the chest leading to sudden death from cardiac arrest during sports activities. N Engl J Med 1995;333:337–42.

4. Link MS, Wang PJ, Pandian NG, et al. An experimental model of sudden death due to low-energy chest-wall impact (commotio cordis). N Engl J Med 1998;338:1805–11.

5. Commotio cordis: ventricular fibrillation triggered by chest impact-induced abnormalities in repolarization. Circ Arrhythm Electrophysiol. 2012;5(2):425-32.

6. Maron BJ, Ahluwalia A, Haas TS, et al. Global epidemiology and demographics of commotio cordis. Heart Rhythm. 2011;8(12):1969-71. Epub 2011 Jul 18.

7. Thakar S, Chandra P, Pednekar M, et al. Complete heart block following a blow on the chest by a soccer ball: a rare manifestation of commotio cordis. Ann Noninvasive Electrocardiol. 2012;17(3):280-2.

8. Cross BJ, Estes NA 3rd, Link MS. Sudden cardiac death in young athletes and nonathletes. Curr Opin Crit Care. 2011 Aug;17(4):328-34.

9. Maron BJ, Estes NA 3rd. "Medical Progress: Commotio cordis". N Engl J Med 2010; 362 (10): 917–27.

Case 6: Renal replacement therapy and dosing

Introduction

Renal replacement therapy (RRT) is the artificial extracorporeal or intracorporeal blood purification when intrinsic renal homeostasis is impaired or lost. This type of organ support is often over simplified on the ICU despite a complexity that exceeds ventilatory support. The fact acute kidney injury (AKI) is still associated with high mortality rates could lead us to hypothesise there is still room for improvement on current practice [1].

Clinical problem and relevant management

A 58 year old gentleman, with no significant past medical history, was admitted with a severe community acquired pneumonia secondary to a streptococcal pneumonia (CURB-65 score 4/5). He presented with a direct pulmonary severe ARDS (PaO2/FiO2 <100mHg with PEEP>5cmH2O), septic shock and MODS. He was ventilated with protective lung ventilator strategies but despite high PEEP, reverse I:E ratios and permissive hypercapnea, he deteriorated and required a trial of high frequency oscillatory ventilation. His intravascular volume was augmented with the appropriate use of fluid resuscitation, which was guided by a LiDCO and optimisation of stroke volume responsiveness and SVV/PPV. He was started on norepinephrine, L-argipressin (0,04u/min) and dobutamine to maintain a mean arterial pressure greater than 65mmHg.

In spite of his improved oxygen delivery and aggressive antimicrobial therapy he remained profoundly acidaemic. The origin of his metabolic acidaemia was thought to be overwhelming sepsis with microvascular dyfunction and a lactic acidaemia (Type B1) mixed with an AKI and uraemic acidaemia, His AKI was thought to be as a result of pre-renal hypoperfusion and intra-renal microvascular dysfunction as a result of his sepsis. The end result was that of an ischaemic acute tubular necrosis (ATN). Early continuous venovenous haemofiltration (CvvHF) was instigated to improve the metabolic status. In addition to improving his metabolic status, it was hoped CvvHF would help gain some control over his significant inflammatory humoral response, vasoplegic shock and allow optimisation of his fluid balance/extravascular lung water. Much discussion was had over the best renal replacement therapy modality to utilise and the possible benefit of high intensity renal dosing.

By day 3 of the ICU admission the patient deteriorated and became vasopressor unresponsive and developed disseminated intravascular coagulation (DIC). Multiple renal replacement filters suffered from premature coagulation dysfunction despite increased pre-filter dilution and various anticoagulants. The patient succumbed to his pneumococcal sepsis on day 4 of their ICU admission.

Discussion

Currently there is a great deal of debate and variations in practice when it comes to appropriate RRT mode, dose and timing of initiation [2]. There are a multitude of indications for initiating RRT, some with more robust evidence than others. A consensus statement issued by the Acute Dialysis Quality Initiative (www.adqi.org) in 2001 listed some absolute indications based on expert opinions and best practice:

Absolute renal indications:

Symptomatic Uraemia (encephalopathy, pericarditis and bleeding) Nephrogenic Pulmonary Oedema Severe unresponsive hyperkalaemia (>6 mmol/L and/or ECG

abnormalities. Severe metabolic academia (pH<7.15) Urine output less than 200ml/24 hours Extreme creatinine and urea

Relative non-renal indications:

SIRS/sepsis Fluid balance Rhabdomyolsis Overdose/Drug accumulation (Haemoperfusion) Renal protection pre/post contrast, against contrast-induced

nephropathy Temperature control Plasmapheresis/Exchange (immune complexes) Mg at least 4 mmol/l and/or anuria/absent deep tendon reflexes. Severe acute liver failure with molecular adsorbent re-circulating

system (MARS, PROMETHEUS) as bridge to transplant

Once the decision to initiate RRT has been undertaken, the first decision to be made is on the mode of RRT and whether that should be delivered intermittently or continuously:

Intermittent RRT

HD most commonly (IHD) Peritoneal dialysis

CRRT

SCUF - Slow Continuous Ultrafiltration Ultrafiltration - fluid removal

CvvHF - Continuous Veno-Venous Hemofiltration Convection - Small, medium and some large size molecules MW

<30000 Daltons

CvvHD - Continuous Veno-Venous Hemodialysis Diffusion - Small molecules <500 Daltons

CvvHDF - Continuous Veno-Venous Hemodiafiltration Diffusion and Convection- small and medium sized molecules

Niche techniques

Plasmapheresis/exchange Haemoperfusion

CRRT is intended to substitute for impaired renal function over an extended period of time and applied for 24 hours a day. Patients with acute kidney injury as part of a multi-organ dysfunction syndrome (MODS) are less likely to tolerate fluid shifts, cardiovascular instability and any further secondary renal insult. Correspondingly in patients with MODS, CRRT is certainly better tolerated in patients with haemodynamic instability and raised intra cranial pressure [3-5]. A great deal of equipoise currently exists on the ideal timing and mode of RRT for critically ill patients, with a lack of mortality benefit of CRRT over IHD [6,7]. It is important to realise that lack of coarse outcome data doesn’t mean that good individualised RRT doesn’t have a role or tangible impact on an individual’s ICU course. As such physicians should endeavour to time and utilise the appropriate RRT for patients based on underlying pathology, haemodynamic status, fluid balance, biochemical derangement, acid-base disturbance and local resource availability and protocols.

RRT dose is the term for the amount of dialysis required in order to achieve a certain level of blood purification. It is commonly measured in either ml/kg/hour in continuous RRT versus urea filtration fraction or clearance in IHD. Appropriate dosing of RRT in the critically ill depends upon the patient’s clinical status (starting point of solute to be cleared and degree of metabolic imbalance), the molecular weight of the solute to be cleared (Figure 1), and the target level for the desired solute.

Figure 1. Molecular weights of various solutes relevant to RRT.

The dose of RRT is dependent on timing schedule, modality, membrane type, extracorporeal blood flow and dialysis/replacement flow. RRT dose can be assessed by direct measurement of effluent fluid solute content; however, it is easily estimated in CRRT because of a few simple assumptions. In CvHF small solute clearance is considered equal to the ultrafiltration rate. Increasing the pre-membrane dilution will increase the filtrate formation but reduces the diffusion gradient and hence solute clearance. During CvvHDF the urea concentration in the dialysate will equilibrate with that in the plasma, and clearance can be approximated by the dialysate flow rate. When calculating the total renal dose in CvvHDF, it is important to include the ultrafiltration rate to the dialysate rate. The above approximations have been shown to acceptably correlate with a more formal set of measurements of urea clearance [8].

Ten years ago Ronco et al published one of the first randomised control trials to

suggest high intensity or augmented dose RRT (35 vs 20ml/kg/hr) had a mortality

benefit in the critically ill [9]. However, two landmark multicentre randomised control trials in America (ATN study) and Australasia (RENAL study) showed that a high renal dosing regime in RRT conferred no mortality benefit [7,10]. The RENAL study compared 25 to 40 ml/kg/hr CvvHDF, while the ATN study compared 20 ml/kg/hr CvvHDF or three IHD per week to 35 ml/kg/hr CVVHDF or daily IHD. These methodologically sound RCTs have certainly answered the question regarding the benefit of high intensity of RRT for the general ICU population.

However, as with all multicentre RCTs one must always exercise caution before applying to individual units and the heterogeneous ICU population. The first thing to consider when talking about renal dosing is the discrepancy often seen between the

intended prescribed renal dose and the actual delivered dose. In an audit completed at The Royal London ICU, we prospectively observed 59 patients requiring RRT with a total of 365 filters uses between July 2010 and January 2011. Fifty-nine filters were purposefully discontinued, leaving 306 filters that unexpectedly failed due to early coagulation or access issues. The mean filter lifespan was 21.21 hours (SD 18.96) and the mean time taken for filter changes was 3.05 hours (SD 1.24).

These results are by no means unusual and are comparable to previously published work by Uchino et al [11]. This ‘downtime’ highlights the difficulties of achieving true continuous RRT. Accordingly, physicians should be aware of the consequences of ‘downtime’ on the true renal dose delivered when prescribing RRT on the ICU. The ideal prescribed dose for CRRT is not universally agreed upon, however, 35 ml/kg/hr of filtrate production is recommended to achieve a delivered dose of 20-25ml/kg/hr for CvvH (post-filter dilution) and CvvHDF [12]. Even though the ATN and RENAL trials showed no mortality benefit to high intensity renal dosing in the general ICU population, physicians should not assume there is not a role for individualised prescribing in RRT. Given the variety of pathologies and patients encountered on the general ICU, it is imperative to try and optimise individuals RRT.

Although mortality is an important outcome, it is vital to consider other incidental benefits of high intensity RRT, including the potential attenuation of SIRS/sepsis according to the humoral theory of sepsis. The RENAL trial did show a non-significant trend towards lower mortality in the septic patient subgroup (OR 0.84, 95% CI 0.62-1.12), however, this was a post hoc analysis and the trial was not powered to irrefutably disprove this hypothesis [10]. The potential haemodynamic and organ dysfunction benefits are well described in animal studies, but would need further human trials to delineate its role on the adult ICU [13].

Learning points

Early initiation of RRT in AKI should be considered when the underlying pathology is likely to persist.

CRRT is better tolerated in the critically ill; however, there is no mortality benefit to any one particular RRT modality.

There is no clear evidence for high intensity renal dosing in AKI for the general ICU population.

High intensity renal dosing may have a role in inflammatory mediator clearance and adsorption in unresponsive septic shock.

References

1. Rhodes A, Moreno RP, Metnitz B, et al. Epidemiology and outcome following postsurgical admission to critical care. Intensive Care Med 2011; 37:1466-1472.

2. Basso F, Ricci Z, Cruz D, et al. International survey on the management of acute kidney injury in critically ill patients: year 2007. Blood Purif 2010;30:214-220

3. Davenport A, Will EJ, Davidson AM, et al. Improved cardiovascular stability during continuous modes of renal replacement therapy in critically ill patients with acute hepatic and renal failure. Crit Care Med 1993; 21(3): 328-338.

4. Augustine JJ, Sandy D, Seifert TH, et al. Randomised controlled trial comparing intermittent with continuous dialysis in patients with ARF. Am J Kidney Dis 2004; 44(6): 1000-1007.

5. Honore PM, Jamez J, Wauthier M, et al. Prospective evaluation of short-term , high volume isovolemic hemofiltration on the hemodynamic course and outcome in patients with intractable circulatory failure resulting from septic shock. Crit Care Med 2000. Vol 28(11) 3581-3586.

6. Vinsonneau C, Camus C, Combes A, et al. Continuous venovenous haemodiafiltration versus intermittent haemodialysis for acute renal failure in patients with multiple-organ dysfunction syndrome: a multicentre randomised trial. Lancet 2006; 368:379-385

7. VA/NIH Acute Renal Failure Trial Network. Intensity of renal support in critically ill patients with acute kidney injury. N Engl J Med 2008; 359:7-20.

8. Ricci Z, Salvatori G, Bonello M, et al. In vivo validation of the adequacy calculator for continuous renal replacement therapies. Crit Care 2005;9:R266-R273.

9. Ronco C, Bellomo R, Homel P, et al. Effects of different doses in continuous veno-venous haemofiltration on outcomes of acute renal failure: a prospective randomised trial. Lancet 2000; 356:26-30.

10.RENAL Replacement Therapy Study Investigators. Intensity of continuous renal-replacement therapy in critically ill patients. N Engl J Med 2009;361:1627-1638.

11.Uchino S, Fealy N, Baldwin I, et al. Continuous is not continuous: the incidence and impact of circuit "down-time" on uraemic control during continuous veno-venous haemofiltration. Intensive Care Med. 2003 Apr;29(4):575-8. Epub 2003 Feb 8.

12.Standards and Recommendations for the provision of renal replacement therpy on intensive care units in the UK. Intensive Care Society standards and Safety. 01/2009.

13.Bellomo R, Kellum JA, Gandhi CR, et al. The effect of intensive plasma water exchange by hemofiltration on hemodynamics and soluble mediators in canine endotoxemia. Am J Respir Crit Care Med 2000; 161:1429-1436

Case 7: Hyperbaric oxygen therapy in carbon monoxide poisoning

Introduction

Airway injuries as a result of house fires are thankfully not a common occurrence but can have serious morbidity and mortality implications. Early priority should be given to securing the airway prior to airway obstruction following a direct airway burn. Once this is achieved, it is important to actively seek and manage the other respiratory and systemic complications of smoke inhalation, carbon monoxide and cyanide poisoning for example. Isolated bronchial or parenchymal injury due to soot related reactive pneumonitis requires meticulous ICU care and due to its relative frequency compared to massive cutaneous thermal injuries, should fall within the remit of most intensivists.

Clinical problem and relevant management

A 59 year old gentleman with a background of hypertension was assessed in the Emergency Department (ED) resuscitation room with injuries sustained in a house fire. The local fire crew found the patient obtunded on the floor of his bedroom which was filled with a significant amount of noxious smoke. When assessed on the scene by paramedics he was noted to have a GCS of 5/15 with equal and reactive pupils.

On assessment at the Emergency Department (ED), he had audible stertor and stridor, dysphonia, with evidence of soot in his nostrils and oropharynx. With pulse oximetry his saturations on 15L/min oxygen via a Hudson non-rebreath mask were 98%, however the arterial blood gas revealed a PaO2 of 8.8kPa, PaC02 of 5.8kPa, a carboxyhaemoglobin level of 34%, a pH of 7.23, BE -7 and a lactate level of 1.9. He was haemodynamically stable and GCS had increased to 9/15. There were no burn or traumatic injuries.

He was intubated in the ED without complication, however, soft tissue laryngeal oedema was noted. Following a normal CT scan of his head he was subsequently transferred to the ICU. A bronchoscopy revealed significant carbonaceous material throughout the bronchial tree and an erythematous, friable mucosa with occasional bleeding despite minimal point contact. Multiple therapeutic washouts were undertaken and protective lung ventilation was instigated. Pressure control ventilation was utilised targeting a tidal volume of 6ml/kg of the ideal body weight and PEEP of 12cmH2O due to extensive basal atelectasis. Additionally the local ‘inhalation protocol’ was started with regular chest physiotherapy. Part of the ‘inhalational protocol’ was a triple nebulised therapy including salbutamol, unfractionated heparin and acetylcysteine. Oxygenation and lung compliance improved dramatically within 72 hours and he was extubated successfully, neurologically intact.

Given the lack of significant thermal injuries, the high admission carboxyhaemoglobin level and the low GCS at the scene, a debate occurred on

admission as to whether this gentleman would be appropriate for hyperbaric oxygen therapy. Due to the initial instability and equipoise regarding the clinical benefit, a decision was taken not to transport despite local bed availability.

Discussion

Carbon monoxide (CO) is a colourless, odourless, non-irritating gas which is formed by incomplete hydrocarbon combustion. It can therefore be formed and inhaled in accidental fires, and can also be produced by defective gas fires and heating systems. Rarely, methylene chloride, an industrial solvent can be ingested and metabolised to CO by the liver.

Physiologically, the affinity between haemoglobin and CO is approximately 230 times stronger than that with oxygen, so haemoglobin preferentially binds to carbon monoxide and consequently impairs oxygen transport causing hypoxaemia.

CO exposure and poisoning often initially goes unnoticed due to the nature of the gas, and patients may be rendered unconscious before noticing exposure. Patients may complain of headache, nausea, dizziness and general malaise, although symptoms of poisoning are highly variable. In severe cases, seizures, syncope, coma may occur, as well as cardiorespiratory and metabolic manifestations, such as cardiac arrhythmias, myocardial ischaemia, pulmonary oedema and lactic acidosis. Cardiac enzymes may be elevated and LVEF may be reduced, although most myocardial dysfunction resolves after treatment [1]. Delayed neurological sequelae, such as cognitive deficit, movement disorder and focal neurology can present up to 240 days after exposure to CO [2].

Diagnosis is largely based on history, examination and carboxyhaemoglobin percentage. Pa02 can occasionally reflect p02 levels dissolved in plasma and therefore may not be reduced. True co-oximetry haemoglobin oxygen saturations are usually a more sensitive indicator of oxygen carriage. The severity of CO poisoning maybe graded and deemed a mild (<20%), moderate (20-40%), severe (>40%) or fatal toxicity (>60%) according to the carboxyhaemoglobin percentage.

CO causes endothelial cell release of nitric oxide, and the formation of oxygen free radicals including peroxynitrite. These oxygen free radicals result in mitochondrial dysfunction, capillary leakage and leukocyte sequestration [3]. The final common pathway and mechanism for neurological injury is thought to be degradation of unsaturated fatty acids and brain lipid peroxidation, which causes delayed demyelinisation of white matter, oedema and cellular apoptosis [3].

Clinical suspicion of cyanide poisoning, evidenced by an acidosis and highly elevated lactate level, should be high in this clinical scenario, although this patient had a normal lactate, mild metabolic acidosis and normal ScvO2, leaving the probability of cyanide poisoning and histotoxic hypoxia unlikely [4].

Standard management of CO toxicity in the acute setting involves high flow oxygen therapy to competitively bind with haemoglobin displacing the carbon monoxide. The

half-life of carboxyhaemoglobin decreases from 300 minutes to 90 minutes with high flow oxygen therapy.

In recent years, the use of hyperbaric oxygen therapy (HBOT) as a treatment for acute CO toxicity has gained much interest in clinical practice due to the potential for mitigating neurological sequelae. In theory, a higher pressure of oxygen allows a more rapid dissociation of CO from haemoglobin, but more importantly has a potential impact on cytochrome oxidase function and brain lipid peroxidation.

There have been a multitude of conflicting results from various randomised control trials looking at the use of HBOT in CO poisoning [5-9]. Two out of the five RCTs suggested a potential benefit to HBOT, two were equivalent and one suggested a negative impact on morbidity [5]. Weaver et al in 2002 showed benefit following three hyperbaric oxygen treatments within 24 hours of presentation, with a reduced risk of cognitive sequelae at 6 and 12 weeks post-acute toxicity [6]. This double-blinded trial randomised patients into two groups of 76 patients: one group received one treatment of normobaric oxygen followed by two treatments of normobaric air within the chamber in a 24 hour period, and the other group received three sessions of hyperbaric oxygen treatments in the chamber over the same time frame. There was a statistically significant reduction in neurological sequelae in the patients receiving three hyperbaric oxygen treatments. A subsequent Cochrane review in 2005 suggested that the two groups in this trial did not have appropriately matched baseline variables, and concluded that there is no evidence to promote the use of hyperbaric oxygen in the prevention of neurological sequelae [10].

Thus the evidence for the use of HBOT to prevent ongoing neurological sequelae, which themselves may have a delayed presentation after acute toxicity, is currently limited. A well-designed multicentre randomised controlled trial would be needed to conclusively delineate any benefit of HBOT in CO poisoning.

Preliminary evidence supporting the use of C02 supplementation to allow normocarbic hyperventilation and increase CO displacement is promising, but as yet no randomised trial exists [11].

Learning Points

There is limited evidence for the use of hyperbaric oxygen therapy in the treatment of acute CO toxicity, and further trials are needed to elucidate the efficacy of its use.

Other methods to increase minute volume may prove useful in the future.

References

1. Kalay N, Ozdogru I, Cetinkaya Y, et al. Cardiovascular effects of carbon monoxide poisoning. Am J Cardiol. 2007 Feb 1;99(3):322-4. Epub 2006 29.

2. Kwon OY, Chung SP, Ha YR, et al; Delayed postanoxic encephalopathy after carbon monoxide poisoning. Emerg Med J 2004 Mar;21(2):250-1

3. Hardy KR, Thom SR. Pathophysiology and treatment of carbon monoxide poisoning. Journal of Toxicology. Clinical Toxicology 1994;32 (6): 613–629.

4. Baud FJ, Borron SW, Mégarbane B, et al. Value of lactic acidosis in the assessment of the severity of acute cyanide poisoning. Crit Care Med. 2002 Sep;30(9):2044-50.

5. Scheinkestel CD, Bailey M, Myles PS, et al. Hyperbaric or normobaric oxygen for acute carbon monoxide poisoning: a randomized controlled clinical trial. MJA March 1999;170 (5): 203–210.

6. Thom SR, Taber RL, Mendiguren II, et al. Delayed neuropsychologic sequelae after carbon monoxide poisoning: prevention by treatment with hyperbaric oxygen. Annals of Emergency Medicine. 1995;25 (4): 474–480.

7. Raphael JC, Elkharrat D, Jars-Guincestre MC, et al. Trial of normobaric and hyperbaric oxygen for acute carbon monoxide intoxication. Lancet. August 1989; 8660: 414–419.

8. Ducasse JL, Celsis P, Marc-Vergnes JP, et al. Non-comatose patients with acute carbon monoxide poisoning: hyperbaric or normobaric oxygenation? Undersea & Hyperbaric Medicine. March 1995; 22(1):9–15.

9. Weaver LK, Hopkins RO, Chan KJ, et al. Hyperbaric oxygen for acute carbon monoxide poisoning. N Engl J Med. 2002 Oct 3;347(14):1057-67

10.Juurlink D. Hyperbaric oxygen for carbon monoxide poisoning. Cochrane Database Syst Rev 2005;(1):CD002041

11.Takeuchi A. A simple "new" method to accelerate clearance of carbon monoxide. Am J Respir Crit Care Med 2000 Jun;161(6):1816-9.

Case 8: Guillain Barré Syndrome

Introduction

Guillain-Barré syndrome (GBS) is the most frequent cause of acute areflexic paralysis worldwide. The pathological process commonly involved is a multifocal inflammation of the spinal roots and peripheral nerves, especially their myelin sheath and rarely the axon themselves. Beyond supportive immediate care and meticulous respiratory care, the precise role of plasmapheresis (PP) over intravenous immunoglobulins (IVIG) remains contentious.

Clinical problem and relevant management

A 68 year old gentleman was admitted with an unsteady gait and progressive symmetrical weakness. On further questioning the patient had recovered from a five day upper respiratory tract infection a month previously. On examination there was a 3/5 weakness below the elbow and knees with relative sparing of sensation. There was no cardiovascular abnormality and initial lung function tests were normal. Investigations of note included a lumbar puncture which showed an elevated protein level, no white or red cells and a negative gram stain. A provisional diagnosis of GBS was made and the patient was admitted to a medical ward where IVIG and serial lung function tests were undertaken.

By day two of his admission his weakness had progressed and his lung function test had deteriorated. Given a FVC of 1.4L, FEV1 of 0.9L and progressive type 2 respiratory failure, the patient was intubated and ventilated. He completed a five day course of IVIG (2g/kg total) following which he had a percutaneous tracheostomy. He made a gradual neurological improvement over the following month despite a bout of sepsis and moderate ARDS secondary to a ventilator associated pneumonia. However, between day 38 and 43 a worsening of his peripheral and respiratory muscle weakness was noted. This was not felt to be in keeping with a critical illness neuro-myopathy and more likely reflected a deterioration in his GBS. Following discussion with various neurologists at a tertiary referral centre the decision was taken to complete a second course of IVIG over a course of PP.

Following a prolonged respiratory wean the patient was transferred to a long term weaning unit where he was eventually decannulated after 8 months. He was discharged home but required significant assistance from a zimmer frame to mobilise.

Discussion

GBS is an acute ascending polyneuropathy which can result in a progressive motor deficit, parasthesia and dysthesia. The muscle weakness tends to progress from the limbs centrally to the trunk over a period of 12 hours to 28 days, resulting in a symmetrical paralysis, respiratory muscles weakness and potentially a significant autonomic dysreflexia. Two thirds of cases with this demyelinating condition are

thought to have had exposure to a viral antigen, with upper respiratory tract infections or diarrhea being common prodromal complaints. A number of viruses have been implicated and include Campylobacter jejuni (30%), cytomegalovirus (10%), Epstein Barr virus, Varicella zoster virus, Haemophilus influenzae and Mycoplasma pneumonia [2]. GBS covers a whole spectrum of subtypes that include:

Acute inflammatory demyelinating polyneuropathy (AIDP)

o Facial variant: Facial diplegia and paresthesia

Acute motor axonal neuropathy (AMAN)

o More and less extensive forms

Acute motor–sensory axonal neuropathy (AMSAN) Acute motor-conduction-block neuropathy

o Pharyngeal–cervical–brachial weakness

Miller Fisher syndrome

o Incomplete forms

Acute ophthalmoparesis (without ataxia) Acute ataxic neuropathy (without ophthalmoplegia)

o Central nervous system variant - Bickerstaff’s brain-stem encephalitis

A thorough history and neurological assessment is vital to establishing a diagnosis. Additional investigations which may assist in the diagnosis include a lumbar puncture, nerve conduction studies and antiganglioside antibody. Cerebrospinal fluid may demonstrate an albuminocytologic dissociation in 50% of GBS during the first week of illness or pleocytosis [2]. An absent H response, abnormal F wave and abnormal upper extremity sensory nerve action potentials (SNAP) combined with a normal sural SNAP are characteristic of early GBS [3].

Despite increasing availability of immunomodulation therapies, GBS is still associated with a 5% mortality rate and 20% of patients will go on to have a permanent neurological disability [1]. In developed countries with early access to invasive ventilation the majority of the deaths are as a result of a medical complication such as superimposed sepsis, immobility related pulmonary emboli and autonomic dysfunction related sudden cardiac arrest.

Intubation and ventilation should be considered in patients who fulfill one major or two minor criteria [4]:

Major criteria

Hypercarbia - PaCO2 >6.4 kPa Hypoxaemia – PaO2 (on FiO2 of 0.21) <7.5 kPa Vital capacity < 15 ml/kg ideal body weight

Minor criteria

Inefficient cough – No accepted quantitative definition. Peak expiratory cough flow (PECF) <250L/min = Ineffective cough <160L/min = risk of repeated infections. Forced expiratory volume in one minute FEV1 <1L/min [5]

Impaired or unsafe swallow – early assessment is essential and nasogastric tube insertion should be considered.

Clinical or radiographic signs of atelectasis

Two recent Cochrane reviews have shown that PP or IVIG are effective in hastening GBS recovery if administered within two weeks of symptom onset [6,7]. The indication for initiation of immunomodulation therapy is the inability to walk independently [2]. Corticosteroids, interferon β-1a, brain-derived neurotrophic factor and cerebrospinal fluid filtration have shown to be of no significant benefit in GBS [6].

In PP, similar equipment to haemofiltration is utilised but with a more porous filter designed to remove the plasma and high molecular weight pathogenic material (IgG/M, paraproteins etc). Once the pathogenic material is removed, an equal volume of substitute fluid (human albumin solution or fresh frozen plasma) is added to the blood cells and returned to the patient. A typical regime would include five exchanges involving the replacement of five plasma volumes [2]. A cochrane review has shown that PP is more effective than supportive care alone in GBS [6].

IVIG is thought to exert its effect by neutralising pathogenic antibodies and inhibiting autoantibody-mediated complement activation which results in reduced myelin injury [2]. A typical course of IVIG in GBS would involve 0.4g/kg/day for five days [2].Treatment with IVIG is as effective as PP and is probably the treatment of choice given its greater safety, convenience and availability [2]. The combination of IVIG and PP has not been shown to be significantly more effective than either alone [7]. Limited evidence exists for a second course of IVIG in GBS patients that show no initial response or who deteriorate following a successful treatment [8].

The cornerstone to GBS management is thorough respiratory care, immunomodulation therapy and rehabilitation, however, the intensivist must also ensure that thromboembolic prophylaxis, early enteral nutrition, good bowel management and chronic pain issues are all addressed.

Learning points

Given the protracted timeline of the disease these patients require holistic multidisciplinary care that extends well beyond their acute care received on the intensive care.

Early use of PP or IVIG have been shown to be equally efficacious in reducing the recovery from GBS.

IVIG is probably the treatment of choice given its greater safety profile, convenience and availability.

References

1. Hughes RAC, Swan AV, Raphaël JC, et al. Immunotherapy for Guillain-Barré syndrome: a systematic review. Brain 2007;130:2245-57.

2. Yuki N, Hartung HP. Guillain-Barré syndrome. N Engl J Med. 2012 Jun 14;366(24):2294-304.

3. Gordon PH, Wilbourn AJ. Early Electrodiagnostic Findings in Guillain-Barré Syndrome. Arch Neurol. 2001;58(6):913-917.

4. Burakgazi AZ, Höke A. Respiratory muscle weakness in peripheral neuropathies. J Peripher Nerv Syst 2010;15:307-13.

5. Howard RS, Davidson C. Long term ventilation in neurogenic respiratory failure. Neurol Neurosurg Psychiatry. 2003 Sep;74 Suppl 3:iii24-30.

6. Hughes RA, Pritchard J, Hadden RD. Pharmacological treatment other than corticosteroids, intravenous immunoglobulin and plasma exchange for Guillain Barré syndrome. Cochrane Database Syst Rev. 2011 Mar 16;(3):CD008630.

7. Hughes RA, Swan AV, van Doorn PA. Intravenous immunoglobulin for Guillain-Barré syndrome. Cochrane Database Syst Rev. 2012 Jul 11;7:CD002063.

8. Farcas P, Avnun L, Frisher S, et al. Efficacy of repeated intravenous immunoglobulin in severe unresponsive Guillain-Barré syndrome. Lancet 1997;350:1747.

Case 9: Extra Corporeal Membrane Oxygenation in Respiratory Failure

Introduction

Extra Corporeal Membrane Oxygenation (ECMO) in is emerging as a useful component of an intensivist’s armamentarium when caring for patients with significant respiratory and cardiovascular insults. In patients where conventional ventilation and adjuncts have failed, ECMO is the next logical consideration. However, benefits may also extend to limiting over distension of the heterogeneously injured lungs and hopefully mitigate the effects of ventilator associated lung injury (VALI).

Clinical problem and relevant management

Paramedics were asked to attend a ten year old suffering from an asthma exacerbation. The boy was a known brittle asthmatic with one previous PICU admission, regular inhaled steroids and leukotriene antagonist orally. On arrival the boy was tachypnoeic with a silent chest, hypoxic on room air and was somnolent. He was given high flow oxygen, 0.25mcg IM epinephrine, nebulised salbutamol and ipratropium bromide. Despite these measures the patient became apnoeic in transport necessitating bag valve mask ventilation and bilateral needle thoracocentesis without obvious release of a tension pneumothorax. Within a few minutes bradycardia progressed to a full cardiorespiratory PEA arrest and was managed using standard advanced life support algorithms. The patient arrived in the ED nine minutes later. Intubation, bilateral open thoracostomies and a further 15 minutes of CPR were required prior to return of spontaneous circulation (ROSC).

Following ROSC the main problem encountered was hypoxia and persistent severe bronchospasm with respiratory acidaemia (pH 6.67). To address the bronchospasm a myriad of therapies were trialled and failed, including nebulisers, epinephrine/ketamine infusions, magnesium sulphate, hydrocortisone and inhaled volatile anaesthetics. Given the failure of all bronchodilator therapies and the need for neuroprotection following the cerebral insult sustained during the cardiac arrest, a decision was taken to place the child on venovenous ECMO.

21F right femoral vein (drainage line) and 19F right internal jugular vein (return line) cannulas were sited and systemic anticoagulation was achieved with unfractionated heparin. Attempts to achieve adequate flow was hampered by persistent sucking on the drainage cannula and further fluid resuscitation was required. Once adequate blood flow was achieved there was a rapid improvement in oxygenation and gas exchange, with a pH greater than 7.25 within 30 minutes of initiation.

As part of a specialist retrieval service (doctor, paramedic and perfusionist), I was tasked with managing the transfer of this patient to a paediatric ECMO centre. Greater Sydney area HEMS are the designated service for transfer of any patients requiring intra-aortic balloon pump or ECMO support. The patient was optimised pre-

transfer by increasing the blood flow to 2.3 L/min and sweep gas to 5L/min to ensure adequate oxygenation, normocarbia and neuroprotection. Ketamine, midazolam, fentanyl, heparin and epinephrine infusion were adjusted and volatile anaesthetics were weaned. Ultra-protective lung ventilation with a TV of 3ml/kg, PEEP of 10 cmH2O, RR 6 and a FiO2 of 0.4 was employed. Targeted temperature management was started and a temperature of 34°C was attained within two hours of the arrest. The patient was transferred in a specialist retrieval road ambulance uneventfully.

Subsequent rapid improvement in the patient’s bronchial smooth muscle tone and lung compliance, allowed the ECMO to be weaned off in 48 hours. Following discontinuation of sedation, the patient remained in a minimal conscious state with an MRI consistent with significant anoxic brain injury. The patient was still an inpatient at the time of completing this case summary.

Figure 1. ECMO retrieval stretcher with bridge that holds the maquet ECMO pump, circuit, hand crank, Lifepak 15 monitor and oxylog 3000+ ventilator

Figure 2. View of stretcher and bridge in the ambulance

Discussion

ECMO is a form of extracorporeal life support where an outside artificial circulation conveys venous blood from the patient to a gas exchange device (oxygenator) where blood becomes enriched with oxygen and carbon dioxide is removed. The use of ECMO in patients where adequate oxygenation and gas exchange cannot be maintained despite optimal conventional respiratory care would seem intuitive, but it may also allow ultra-protective lung ventilation to limit shearing forces and further VALI. While the following discussion will concentrate on peripheral ECMO, in practice the clinician may come across central ECMO (Figure 3) and AV extracorporeal CO2 removal (interventional lung assist, Novalung) depending on the clinical situation and local preferences.

Figure 3. Central VA ECMO

The modality of peripheral ECMO utilised will depend performed will depend on the patient’s underlying cardiac function. Venovenous (VV) ECMO is usually undertaken for isolated respiratory failure, while venoarterial (VA) ECMO is instigated for combined cardiac and respiratory failure.

VV ECMO involves removal of venous blood from the patient’s central veins via a wire reinforced access/drainage cannula. The blood then passes through a pump, typically a centrifugal rather than a roller pump to limit haemolysis [1]. After the pump the blood flows through the oxygenator and returned to the venous system via a return line near the right atrium. Oxygenation and carbon dioxide excretion can be optimised by altering the blood flow and oxygen sweep gas respectively. The required blood flow in VV ECMO is classically 2/3 of the patient’s cardiac output (60-70ml/kg/min) while the sweep gas flow rate will usually be double that of the blood flow rate. If attempts to improve oxygenation by increasing blood flow fail, the possibility of recirculation between the drainage and return cannula should be considered.

Other possible set ups seen in VV ECMO include a second venous drainage cannula in high flow VV ECMO or use of a single Avalon (dual lumen) cannula.

Figure 4. VV ECMO circuit

VA ECMO involves venous blood from the patient being accessed from the large central veins and returned to a major artery after it has passed through the pump and oxygenator. It provides support for severe respiratory and cardiac failure. Low flow VA ECMO is a temporary form of ECMO support in which small cannulae are inserted percutaneously. This is often referred to as ECMO-CPR (E-CPR) or Extracorporeal lifesupport (ECLS) and is merely a temporising circuit set up in the emergent situation.

All forms of ECMO need anticoagulation to limit circuit thrombosis and complications. With newer heparin bonded lines the anticoagulation requirements are lower and typically an ACT of 180-220 should be targeted.

Figure 5. VA ECMO circuit

ECMO is indicated for potentially reversible, life-threatening forms of respiratory and cardiac failure, which has failed to respond to conventional therapy. Indications include:

Pathology amenable to VV ECMO

Common indications

Severe pneumonia ARDS Acute lung (graft) failure following transplant Pulmonary contusion

Common indications

Alveolar proteinosis Smoke inhalation Status asthmaticus Airway obstruction Aspiration syndromes

Pathology amenable to VA ECMO

Common indications

Cardiogenic shock, myocardial infarction and complications, refractory to conventional therapy including IABP

Post cardiac surgery: unable to wean safely from cardiopulmonary bypass using conventional supports

Drug overdose with profound cardiac depression Myocarditis Early graft failure: post heart / heart-lung transplant Septic cardiomyopathy

Uncommon indications

Pulmonary embolism Cardiac or major vessel trauma Massive haemoptysis / pulmonary haemorrhage Pulmonary trauma Acute anaphylaxis Peri-partum cardiomyopathy Bridge to transplant

Early RCTs were unable to demonstrate any survival benefit with ECMO [2,3]. No survival benefit coupled with a worryingly high incidence of uncontrolled bleeding events whilst anticoagulated, limited the uptake of the technique [3]. However, over the last decade there has been a growing body of evidences to suggest that ECMO may be beneficial in well selected patient groups [4-8]. This survival benefit could be

explained by the greater application of protective lung ventilation strategies and adjuncts as well as the improved ECMO technology.

The CESAR trial published in 2009, was a RCT that showed patients with severe respiratory failure (Murray score >3 and pH<7.2) that were managed in an ECMO capable institution had a 16% absolute risk reduction of death or severe disability at six months (RR=0.69, CI 0.05-0.97, p=0.03). This adequately powered study randomised 180 patients in a 1:1 ratio to either a control arm of local institution management or transfer to a specialist ECMO facility. Of the patients assigned to transfer to the specialist ECMO facility; 68 (75%) received ECMO, 17 had conventional management, 3 died before transport and 2 died in transit. Encouragingly, this true intention to treat analysis, use of traditional roller pumps and high dose anticoagulation may have actually diluted the potential benefits of ECMO.

A few potential concerns do exist with the trial. The major concern is that patients in the control arm were managed across a number of centres with no specific management protocol. The importance of protective ventilation strategies is well known [9], however, only 70% of patients in the control arm were ventilated in accordance with this evidence [7]. This control arm were also not directly compared to patients receiving ECMO, as only 75% of patients in the intervention group received ECMO [7]. The intervention group were managed using a standardised protocol which utilised protective lung ventilation strategies and then potentially prone ventilation, HFOV, inhaled NO and ECMO [7]. So potentially some of the perceived benefit could be related to the standardised protocol treatment and not purely ECMO.

The novel influenza (H1N1) pandemic in 2009 highlighted the potential benefits of ECMO in severe viral respiratory failure [8,10]. The reported hospital mortality rates of 21-27.5% are far lower than any previously published data [8,10]. From the 4400 patients described in the extracorporeal life support registry report we know the survival rates from viral pneumonia are high comparatively to other ECMO indications [11]. This in combination with a young median age of 34.4 and low rates of MODS would probably explain the high survival rates. However, this trial is one of the largest recent observational ECMO cohort and it may reflect also reflect an improvement in technology. The Greater Sydney HEMS were involved in the transfer of 40 patients with novel influenza [12]. Following ECMO initiation in the referring institution, there were no deaths during transfer by our specialist retrieval service, compared to the two deaths in the CESAR trial [7,12].

Emergent, low flow VA ECMO, often referred to as ECMO-CPR (E-CPR) or extracorporeal life support (ECLS) is a potential new use for ECMO in refractory cardiac arrest. In a prospective, observational trial with propensity-score matched groups there was a significant improvement in one year survival (18.6% vs 9.7%, p=0.006) between E-CPR and conventional resuscitation [13]. Clearly careful patient selection is vital here to ensure desirable survival rates, while limiting poor neurological outcomes.

Learning points

ECMO technology is continuously evolving and improving.

There may be a survival benefit when ECMO is employed in a well selected subgroup of patients with severe respiratory and cardiac failure.

References

1. Lawson DS, Ing R, Cheifetz IM et al. Hemolytic characteristics of three commercially available centrifugal blood pumps. Pediatr Crit Care Med 2005;6:573-77.

2. Zapol WM, Snider MT, Hill JD et al. Extracorporeal membrane oxygenation in severe acute respiratory failure. A randomized prospective study. JAMA 1979;242:2193-96.

3. Morris AH, Wallace CJ, Menlove RL et al. Randomized clinical trial of pressure-controlled inverse ratio ventilation and extracorporeal CO2 removal for adult respiratory distress syndrome. Am J Respir Crit Care Med 1994;149:295-305.

4. Mols G, Loop T, Geiger K et al. Extracorporeal membrane oxygenation: a ten-year experience. Am J Surg 2000;180:144-54.

5. Beiderlinden M, Eikermann M, Boes T et al. Treatment of severe acute respiratory distress syndrome: role of extracorporeal gas exchange. Intensive Care Med 2006;32:1627-31.

6. Hemmila MR, Rowe SA, Boules TN et al. Extracorporeal life support for severe acute respiratory distress syndrome in adults. Ann Surg 2004;240: 595-607.

7. Peek GJ, Mugford M, Tiruvoipati R et al. Efficacy and economic assessment of conventional ventilator support versus extracorporeal membrane oxygenation for severe adult respiratory failure (CESAR): a multicentre randomized controlled trial. Lancet 2009;374:1351-63.

8. Noah M, Peek G, Finney S et al. Referral to an extracorporeal membrane oxygenation center and mortality among patients with severe 2009 influenza A (H1N1). JAMA 2011;306:1659-68.

9. The Acute Respiratory Distress Syndrome Network. Ventilation with lower tidal volumes as compared with traditional tidal volumes for acute lung injury and the acute respiratory distress syndrome. N Engl J Med. 2000 May 4;342(18):1301-8.

10.Davies A, Jones D, Bailey M et al. Extracorporeal membrane oxygenation for 2009 influenza A (H1N1) acute respiratory distress syndrome. JAMA 2009;302:1888-95.

11.Extracorporeal Life Support Organization (ELSO). ECMO Registry Report of the ELSO: International Summary. ELSO, Ann Arbor, Michigan; January 2011.

12.Forrest P, Ratchford J, Burns B, et al. Retrieval of critically ill adults using extracorporeal membrane oxygenation: an Australian experience. Intensive Care Med. 2011 May;37(5):824-30. Epub 2011 Feb 26.

13.Chen YS, Lin JW, Yu HY et al. Cardiopulmonary resuscitation with assisted extracorporeal life-support versus conventional cardiopulmonary resuscitation in adults with in-hospital cardiac arrest: an observational study and propensity analysis. Lancet 2008;16;372:554-61.

Case 10: New and novel strategies for managing propranolol overdose

Introduction

Beta blockers (BB) have been a commonly prescribed cardiovascular medication for over 30 years. BB toxicity can have significant cardiovascular complications that may be refractory to standard medications used for circulatory support. BBs are often overrepresented when it comes to cardiovascular medication deaths.

Clinical problem and relevant management

A 20-year old female presented to her local ED three hours post ingestion of 20 units of alcohol and approximately 1.6g of propanolol (40x tablets of 40mg). There was no verbalised suicidal ideation, suicide note, or previous suicide attempts. On initial assessment she had a heart rate of 46 bpm, BP 92/60, SpO2 99% on FiO2 0.6, respiratory rate of 22 and was drowsy with a GCS 11/15. Her ECG showed first degree heart block with a PR interval of 210ms, QRS duration 128ms, QTc interval 420ms. The patient was intubated, received three 600mcg doses of atropine, 50 mmol of sodium bicarbonate, titrated fluid resuscitation and 50g of activated charcoal. Despite these measures the patient remained hypotensive and bradycardic. An epinephrine infusion was started to maintain mean arterial pressure of 65mmHg and further 50mmol of Sodium bicarbonate. Due to the extreme doses of epinephrine (approaching 0.4mcg/kg/min) and the ECG abnormalities, the patient was discussed with the regional poisons unit which advised further plasma alkalinisation and high dose insulin therapy. Epinephrine requirements quickly reduced once the pH reached 7.4 and an insulin infusion was established at a steady rate of 20u/hr.

The patient was extubated on day three of her ICU admission following an appropriate neurological wake up and was discharged from hospital on day five with appropriate psychiatric follow up.

Discussion

Propranolol is a non-cardioselective beta blocker with sodium-channel blocking effects. Its pertinent pharmacodynamics and pharmacokinetics are:

Absorption: Rapidly absorbed orally. Peak blood levels occur at 1-3 hours following oral administration.

Mechanism of action - Competitive antagonism of beta receptors leads to decreased intracellular cAMP mediated by G-protein transmembrane receptors. This ultimately results in blunting of the chronotropic, inotropic and metabolic effects of catecholamines. Unlike other beta blockers, propanolol and sotalol also block sodium-channels in the myocardium resulting in prolongation of the cardiac action potential with widening of the QRS complex

and potentially ventricular arrhythmias. Blockade of sodium channels in the CNS produces neurotoxic effects.

Distribution: Highly lipophilic agent (93% protein bound) with a high volume of distribution. Rapidly distributed to all tissues.

Metabolism: Undergoes extensive hepatic metabolism, with hydroxylation of the aromatic nucleus and degradation of the isoprenaline side-chain. Over 20 metabolites identified. The 4-hydroxy metabolite has active beta-blocking properties.

Elimination: 95-100% of an ingested dose is excreted in the urine as metabolites and their conjugates.

Half-life: Plasma half-life is around 3-6 hours. The pharmacological effects last much longer than this. Elimination half-life is 12 hours and may be prolonged following overdose

Figure 1. Propanolol

Figure 2. Effect of Na channel blockers on cardiac action potential

Typical features of BB toxicity include:

Hypotension, bradycardia and progressive cardiogenic shock.

Bradyarrhythmias – Sinus bradycardia, junctional or ventricular bradycardia, prolonged PR interval, atrioventricular block (1st-3rd degree).

Bronchospasm Hyperkalaemia Hypo/hyperglycaemia

In addition to other BBs, propanolol with its sodium channel blockade can cause:

Cardiotoxicity – broad QRS (> 100 ms in lead II), right axis deviation of the terminal QRS, ventricular arrhythmias and cardic arrest in high doses. QRS > 160 ms is predictive of ventricular arrhythmias.

Neurotoxicity – coma, seizures and delirium. QRS > 100ms predictive of seizures.

Management

Adult patients that have ingested greater than 1g of propanolol are at significant risk of sudden cardiac dysrhythmias and death, and should be managed in a critical care area. The initial approach for evaluating the critically poisoned patient focuses on thorough assessment, appropriate stabilisation and supportive care. Such individuals should be managed utilising the standard ABCDE approach. Consideration should be given to early intubation and ventilation in patients with neurotoxicity, as this will allow acid-base balance optimisation and airway protection if activated charcoal (1g/kg) gut decontamination is considered. Serial 12-lead ECGs should be performed to assess for beta and sodium-channel blockade. Seizures should be promptly treated with intravenous benzodiazepines with the addition of sodium bicarbonate if QRS > 100ms. Following the initial standardised approach to the poisoned patient, it is vital to address any signs β-adrenergic receptor and sodium channel toxicity.

Propranolol has profound effects on cardiac performance and can result in increased LVEDP, left ventricular dP/dt max, stroke volume and cardiac output [1-3]. First line management of hypotension is titrated crystalloid resuscitation. If this does not achieve an adequate cardiac output, blood pressure and end organ perfusion, there should be a low threshold for the use of inotropes and chronotropes. Repeated doses of atropine (20 mcg/kg) can be used as a temporising measure for a bradycardia related low cardiac output state. Persistent bradycardia despite atropine associated with hypotension is better treated with a titrated infusion of epinephrine, norepinephrine or isoprenaline. Glucagon was traditionally thought to be an “antidote” to BB toxicity as it directly stimulated adenylyl cyclase resulting in increased intracellular cAMP [4,5]. Its clinical application is limited by the prohibitively high doses required and the possible toxicity of the solvent phenol that is utilised in glucagon production [4].

If any kind vasoactive infusion is required consideration should be given to early initiation of high-dose insulin euglycaemic therapy (HIET). HIET should be cautiously initiated and monitored intensively with an preliminary dose of 0.5 U/kg over 30

minutes, followed by a 0.5–1 U/kg/Hr continuous infusion [2,4,5]. To maintain euglycaemia and normokalaemia, an infusion of between 15-30g/Hr of glucose and potassium supplementation will be required [2]. Insulin promotes glycogenesis and cellular glucose uptake, while inhibiting gluconeogenesis, β-oxidation and lipolysis [4]. Insulin also is known to exhibit inotropic effects in myocardium that has been depressed by various insults [6]. Reikeras et al has extensively studied the haemodynamic and metabolic effects of insulin therapy in BB overdose [1,5]. These animal models demonstrated that HIET improves cardiac function with an improved stroke volume and cardiac output [1,2,5].

The inotropic effects of insulin seem to be dose dependent and unrelated to adrenergic mechanisms [2]. Improvement of cardiac function in HIET is not associated with an increase in myocardial oxygen consumption [1,4]. Reikeras et al demonstrated that HIET did not alter myocardial blood flow and oxygen consumption despite the significant improvement in cardiac contractility [1]. While more clinical trials are needed, animal studies and human case reports have shown that in BB toxicity HIET is associated with improved mortality, and maybe superior to glucagon or epinephrine infusions alone [7].

It is hypothesised that intravenous lipid emulsion therapy exerts its beneficial effects by expanding the plasma lipid sink, with consequent reduction in free drug level [8]. Propanolol is a highly lipophilic and protein bound drug that in overdose may be amenable to lipid emulsion therapy. Currently there is only a limited evidence base to suggest that intravenous lipid emulsion may be helpful in potentially lethal propranolol overdoses [9-11]. Lipid emulsion therapy should probably only be considered when there is extreme haemodynamic instability or cardiac arrest refractory to other measures [8].

Figure 3. Suggested dosing for Lipid emulsion therapy [8].

Sodium channel blockade management is vital to a successful outcome in propanolol toxicity and involves:

Sodium bicarbonate is used to treat acidaemia, while improving the prolonged cardiac action potential and QRS prolongation [4].

If QRS>100ms, haemodynamically unstable or in cardiac arrest.

o Sodium bicarbonate 1-2 mmol/kg) bolus every few minutes while monitoring the effect on ECG and haemodynamic response.

o Consider further sodium bicarbonate bolus or infusion once stable to maintain pH 7.5- 7.55 based on hourly blood gas and QRS <100ms.

o Boluses of sodium bicarbonate are likely to be more effective than infusions because they will lead to rapid shifts in the concentration of free drug.

Intubate as soon as possible and hyperventilate to maintain a pH of 7.50 – 7.55 in combination with sodium bicarbonate. This helps to minimise pH related cardiotoxicity.

Lignocaine (1.5mg/kg) IV can be utilised if ventricular arrhythmias persist following sodium bicarbonate and hyperventilation have achieved a pH > 7.5.

In established cardiotoxicity, the dose of sodium bicarbonate can be repeated every few minutes until the BP improves and QRS complexes begin to narrow.

Avoid Vaughan William classification drugs from classes Ia (procainamide), Ic (flecainide) and III (amiodarone) as they could worsen conduction defects and cardiovascular instability [4].

In addition to the above measures haemoperfusion and VA ECMO have also been described, but there are currently only a few specialist centres with limited experience [4].

Learning points

The care of a critically ill poisoned patient should involve appropriate investigation, monitoring and initial resuscitation, accompanied by optimisation of absorption, decontamination, elimination and potential use of antidotes.

In patients who remain haemodynamically unstable despite vasoactive infusions, consideration should be given to early initiation of HIET.

Propanolol and sotalol have significant sodium channel blocking effects which need meticulous acid-base balance management in overdose.

References

1. Reikeras O, Gunnes P, Sorlie D, et al. Haemodynamic effects of low and high doses of insulin during beta-receptor blockade in dogs. Clin Physiol 1985, 5:455-467.

2. Lheureux PER, Zahir S, Gris M, et al. Bench-to-bedside review: Hyperinsulinaemia/euglycaemia therapy in the management of overdose of calcium-channel blockers. Crit Care. 2006;10(3):212. Epub 2006 May 22

3. Kerns W 2nd, Schroeder D, Williams C, et al. Insulin improves survival in a canine model of acute beta-blocker toxicity. Ann Emerg Med 1997, 29:748-757.

4. Kerns W. Management of b-Adrenergic Blocker and Calcium Channel Antagonist Toxicity. Emerg Med Clin N Am 2007: 25:309–331.

5. Reikeras O, Gunnes P, Sorlie D, et al. Metabolic effects of low and high doses insulin during beta-receptor blockade in dogs. Clin Physiol 1985, 5:469-478.

6. Krukenkamp I, Sorlie D, Silverman N, et al. Direct effect of high-dose insulin on the depressed heart after betablockade or ischemia. Thorac Cardiovasc Surg 1986, 34:305-309.

7. Kerns W 2nd, Schroeder D, Williams C, et al. Insulin improves survival in a canine model of acute beta-blocker toxicity. Ann Emerg Med 1997, 29:748-757.

8. Cave G, Harvey M. Intravenous Lipid Emulsion as Antidote Beyond Local Anesthetic Toxicity: A Systematic Review. Acad Emerg Med. 2009 Sep;16(9):815-24.

9. Bania T, Chu J, Wesolowski M. The hemodynamic effect of Intralipid on propranolol toxicity. Acad Emerg Med. 2006; 13 S1):109.

10.Harvey M, Cave G. Intralipid infusion ameliorates propranolol induced hypotension in rabbits. J Med Toxicol. 2008; 4:71–6.

11.Harvey M, Cave G. Lipid emulsion may augment early blood pressure recovery in a rabbit model of atenolol toxicity. J Med Toxicol. 2009; 5:50–1.