Journal of the Intensive Care Society

The UK has one of the highest rates of deliberate drug overdose in Europe but little is known about patients admitted to intensive care. We reviewed the epidemiology and outcome of patients with self-poisoning admitted to the intensive care unit (ICU) in a tertiary teaching hospital in Scotland and related this to previously established international patterns. All patients admitted to the ICU in the period 2005–2009 were reviewed to identify those with a diagnosis of drug overdose or self-poisoning. The records of these patients (n=481, 3.8%) were analysed to obtain details about epidemiology, investigation, organ support and outcome. Median age was 37 years and male: female ratio was 1.44:1. The substances most commonly involved were: alcohol (41%), tricyclic antidepressants (28%), benzodiazepines (21%), recreational drugs (23%), opioids (14%), and paracetamol (19%). The majority of patients (69%) required ventilatory support, 6% received inotropic support and 6% received renal replacement therapy. A CT brain scan was performed in 176 patients (37%): acute changes were found in six patients but these did not influence the acute management of any patient. Median length of stay in intensive care was 0.7 days and only 12% of patients remained in intensive care longer than 48 hours. Twenty patients (4%) died in hospital. The demography of patients admitted to intensive care after self-poisoning and the substances involved are changing, with increasing prevalence of alcohol, recreational drugs and opioids. This has significant implications for society and health care now and in the future.

Induced hypothermia has established indications in cardiac arrest in adults and in hypoxic-ischaemic encephalopathy in infants. Despite substantial research effort its application in the setting of trauma remains controversial. In head and spinal trauma mild cooling may help to limit secondary injury. In penetrating trauma, profound cooling at the time of cardiac arrest may offer an extended window to control haemorrhage before irreversible ischaemic brain damage occurs. Both of these potential indications are the subject of clinical trials. This review seeks to set in context these studies and previous work in this field.

Organisation of critical care services affects patient outcomes, as does the quality of care preceding intensive care unit (ICU) admission. Opportunities for improvement in both these spheres were identified in a district hospital high dependency unit (HDU). Changes were made to the medical and nursing leadership and staffing in HDU including enhanced ICU clinician and nursing responsibility for patient care, admission and discharge, development of a common critical care nursing pool, dedicated daytime supervised trainee medical staff and the option for ward staff to refer patients for an HDU evaluation. Data evaluating the number of patients admitted to ICU, requiring invasive ventilatory support and requiring renal replacement therapy were collected in real time on the existing Scottish Intensive Care Society database and retrospectively analysed using statistical process control (SPC) chart methodology. Organisational changes in HDU care were associated with SPC evidence of statistically significant reductions in patients receiving invasive ventilation, number of patient ventilation days, level 3 care days and renal replacement therapy days. Changing the organisation of HDU care in our setting was associated with marked changes in the pattern of intensive care use. It reduced the number of people receiving invasive ventilation and reduced number of ventilation, level 3 and renal replacement therapy days.

Acute kidney injury in the intensive care unit (ICU) requiring renal replacement therapy (RRT) is common and mortality is high. The dose delivered is important and is usually inadequate. Evidence for dose is quoted as clearance, but RRT is usually prescribed as pump flow rates. Accurately delivering an evidence-based dose to a patient is difficult because of inefficiencies of RRT, the complexity of its mathematics and poor understanding. Inadequate dose can result from inadequate prescribing, which should be by ideal body weight and possibly by indication. Inadequate delivery of a prescription can occur because the delivered dose depends not only on the dialysate and ultrafiltrate pump flow rates, but also blood flow rate, predilution inefficiency, fluid removal rate and downtime. To investigate the feasibility of using a web-based calculator to make prescribing by clearance easy and to predict and compensate for these factors, a web page with a RRT calculator using JavaScript was used. Data were collected from 19 treatments before the introduction of the calculator and 20 after. Results showed that dose delivery was significantly improved (p<0.001). There was an improvement in prescribing an evidence-based dose which did not reach statistical significance (p=0.056) but the standard deviation was significantly smaller, indicating more rational prescribing. The calculator significantly improved prescribing and delivery of RRT in our ICU.

ICU physicians frequently need an immediate echocardiogram to rule out major pathology. Discussions about a formal certification process in Europe are in progress and remain largely theoretical. To aid these discussions we describe our practical experience of an internal certification programme that could serve as one possible template.

Fluid management of critically ill patients and those undergoing major surgery can be extremely difficult. It is important to predict which patients will respond to volume expansion in order to avoid undesired hypovolaemia and fluid overload. Traditionally, central venous pressure and pulmonary artery occlusion pressure have been used to guide fluid management. However, assimilation of available evidence from the past twenty years demonstrates that neither central venous pressure nor pulmonary artery occlusion pressure appears to be a useful predictor of haemodynamic response to fluid challenge. Recent evidence suggests that stroke volume variation, pulse pressure variation and the variation of the amplitude of the pulse oximeter plethysmographic waveform are far more accurate in predicting volume responsiveness in the intensive care unit. It is important for intensivists and perioperative physicians to understand the strengths and limitations of each method in order to scrutinise and interpret such data appropriately.

Quoted rates of complications following tracheostomy insertion vary greatly, with little information available on complications occurring in patients with tracheostomies in situ or following their removal. A series of 200 consecutive tracheostomies in critical care patients were reviewed for complications occurring at insertion, in patients while cannulated, and following decannulation. A questionnaire was completed at insertion and patients then received weekly follow-up where any complications were recorded. High rates of follow-up were achieved. Insertion complications included: major bleeding (5%), tracheal wall injury (0.5%), pneumothorax (0.5%). Complications in cannulated patients included: prolonged bleeding (5%), pneumothorax (2%), accidental decannulation (4%) and tube blockage (6%). Of the blockages and displacements, 40% resulted in severe hypoxia and in two patients, cardiac arrest. Post-decannulation one patient (0.8%) required immediate recannulation. Initial problems relating to swallowing difficulties post-decannulation were seen to resolve with time. This work demonstrates that tracheostomy can be safely performed in critical care patients although serious adverse events may occur due to tube blockage or displacement.

This review looks at the indications for the use of aircraft in medical transfers, the planning required and the potential problems specific to aeromedical transfers. Additionally some relevant aspects of flight and meteorological factors are discussed. The focus of this review is on the transport of civilians. The authors acknowledge the huge number of complex transfers that are presently being undertaken by the military. Although much of the review is relevant to this population group, it is not our intention to discuss these transfers specifically.

Intensivists and anaesthetists working in district general hospitals have an important role to play in the management of critically ill children. The ability to maintain the skills needed for paediatric resuscitation and stabilisation will partly depend upon clinical exposure to paediatric cases. However, the centralisation of paediatric services and the expansion of critical care services are likely to have reduced the paediatric caseload for many clinicians. We undertook a review of paediatric cases referred to the Department of Critical Care at the Queen Alexandra Hospital, Portsmouth to determine the caseload for both the service and individual consultants. The caseload for individual consultants was relatively low (an average of eight cases per consultant over 12 months) and highlights the challenge facing clinicians in district general hospitals who need to maintain their expertise in managing critically ill children.

Patients presenting to critical care with severe sepsis are often elderly and have multiple co-morbidities. We constructed four clinical scenarios describing patients compromised to varying degrees by advanced age or co-morbidity and presenting with severe sepsis. Clinicians attending the Northwest Critical Care Symposium in April 2011 were asked to specify the interventions they would offer each patient, after consideration of the clinical information, with regards to components of the sepsis six-hour resuscitation care bundle. In all four scenarios a significant proportion of clinicians would have withheld more invasive measures in the Surviving Sepsis resuscitation bundle such as central venous access, the use of vasopressors and inotropes and admission to a high dependency setting. The majority of clinicians would, however, have offered other standard, less invasive treatment measures to these patients. Clinical judgment relating to age and co-morbidity influences the implementation of the sepsis management bundle and should be taken into account when assessing compliance with the bundle.

The 2009 swine flu pandemic was an exacting time for intensive care medicine. The epidemiology, and the individual cases, presented ethical, legal, and clinical problems which were undertaken against a background of global public scrutiny. When now viewed, two years later, with the experience of a further pandemic many of the cases are still heart rending and many of the clinical questions remain unanswered. We have the opportunity to debate the management of the consequences of a global pandemic affecting so many young and previously healthy people and certain cases may serve to aid this debate. A young pregnant mother, diagnosed with the 2009/H1N1 virus, presented in respiratory failure requiring rapid escalation to intubation and multiple organ support. She was in the second trimester of pregnancy with an anencephalic fetus and elective termination was proposed to aid resolution of her critical illness. During the initial conversations with the obstetric team, prior to intubation, she suggested that she wanted more time to think about termination and elected to continue with the pregnancy until her acute illness had resolved. Comprehensive debate ensued, both among the intensive care clinicians and with the obstetricians about the benefits and risks of termination. At the time there was limited experience in this patient group and there was an absence of clinical precedent with which to guide treatment. This case explores the clinical medicine and the ethical and legal circumstances around termination in the absence of consent in an obtunded critically ill patient.

In this case report, two patients are presented who developed myoclonic status epilepticus after severe hypoxic brain injury and were diagnosed as having Lance-Adams syndrome (post-hypoxic myoclonus). The diagnosis of Lance-Adams syndrome and the controversies and difficulties that surround its diagnosis and treatment and other aspects of prognostication in cardiac arrest are reviewed.

In critical care, situations may occur where optimal treatment is unknown. Advances in patient management in other specialties can impact on referrals for our expertise in physiological and pharmacological support. A 63-year-old man required sedation and monitoring in critical care to suppress endogenous catecholamine release for treatment of an electrical storm (defined as three or more episodes of ventricular tachycardia or fibrillation within 24 hours). This case report illustrates the importance of multi-disciplinary input when novel techniques are employed.

Compared with the lowered shoulder position during subclavian line placement, the neutral shoulder position was associated with less frequent aberrant placement.

Level of evidence: 1⁻ (RCT with a high risk of bias)

In a mixed medical-surgical intensive care unit (ICU), plasma C-reactive protein (CRP) levels on the day of discharge were not a predictor of readmissions to ICU or death within a two-week time frame.

Level of evidence: 2B (Cohort observational study without randomisation)

In this underpowered, randomised controlled trial, early induction of hypothermia did not improve Glasgow Outcome Score six months following traumatic brain injury.

Level of evidence: 1⁻ (RCT with a high risk of bias)

This large randomised controlled trial (RCT) found that late initiation of parenteral nutrition was associated with faster recovery and fewer complications but no difference in mortality when compared to early parenteral nutrition.

Level of evidence: 1⁺ (RCT with low risk of bias)