VOL: 97, ISSUE: 43, PAGE NO: 30
Anne Rowat, PhD, BSc, RGN, is action research training fellow, Department of Clinical Neurosciences, University of Edinburgh, Western General Hospital, EdinburghStroke remains the third most common cause of death and the leading cause of major disability in the developed world. Recent evidence from randomised controlled trials (RCTs) suggests that some pharmacological treatments, such as aspirin (Chinese Acute Stroke Trial Collaborative Group, 1997; International Stroke Trial Collaborative Group, 1997) and thrombolytic ('clot-busting') drugs (Wardlaw et al, 1999), may be beneficial for particular subgroups of stroke patients. However, the benefits of pharmacological treatments are rather modest when compared to 'organised care' on a stroke unit (Stroke Unit Trialists' Collaboration, 1998).
Stroke remains the third most common cause of death and the leading cause of major disability in the developed world. Recent evidence from randomised controlled trials (RCTs) suggests that some pharmacological treatments, such as aspirin (Chinese Acute Stroke Trial Collaborative Group, 1997; International Stroke Trial Collaborative Group, 1997) and thrombolytic ('clot-busting') drugs (Wardlaw et al, 1999), may be beneficial for particular subgroups of stroke patients. However, the benefits of pharmacological treatments are rather modest when compared to 'organised care' on a stroke unit (Stroke Unit Trialists' Collaboration, 1998).
Systematic reviews of RCTs that have studied the organised stroke unit care indicate that patients are more likely to survive, return home and regain independence than those who receive conventional care on general medical wards (Stroke Unit Trialists' Collaboration, 1998). Stroke unit care differs from conventional care in that it is coordinated by a multidisciplinary team - doctors, nurses, and therapists - with a special interest in stroke (Dennis and Langhorne, 1994).
However, it is difficult to distinguish which components of care make the greatest contribution to improving outcome. There is an urgent need to know this to ascertain which components are most cost-effective, and which require improvement.
Nurses are essential members of the multidisciplinary stroke team and often spend the highest proportion of time with patients. So it seems reasonable to assume that nursing interventions play a vital role in improving stroke patients' outcomes. However, current nursing strategies are rarely evidence-based, and there is the concern that sometimes we may do more harm than good.
Particular causes for concern are hypoxia and hypotension resulting from routine positioning of immobile patients. Areas of the brain which are ischaemic but viable (the ischaemic penumbra), surrounding completely damaged tissue (the ischaemic core), may be particularly vulnerable to damage from mild hypoxia or hypotension. This could hinder neuronal recovery of the partially damaged tissue (Langhorne, 1999).
Blood flow in the ischaemic penumbra is critically reduced. Neurones in this area function abnormally. They will survive for a limited period (Fisher and Garcia, 1996), but are vulnerable to any further reduction in cerebral blood flow or oxygenation. Therefore, the penumbral region may be prevented from becoming necrotic if positions that may cause adverse changes in blood pressure (BP) or oxygenation are avoided.
Currently, nurses use positioning to improve patient comfort, prevent skin breakdown, facilitate breathing, and prevent contractures and deep vein thrombosis (Lasater-Erhard, 1995). Its effect on the BP and oxygen saturation of stroke patients has been investigated (Palmer, 1983; Robinson and Potter, 1995; Elizabeth et al, 1993; Chatterton et al, 2000). However, these studies were too small to determine precisely the frequency or magnitude of any posture-related change in BP and oxygenation. The following study was therefore undertaken.
The aim of the study was to investigate the influence of positioning on BP and arterial oxygen saturation (SaO2) in acute stroke patients with hemiparesis. The sample consisted of all patients with an acute ischaemic or haemorrhagic stroke - but not subarachnoid haemorrhage - admitted to medical and surgical wards at a large teaching hospital.
A total of 129 patients with mild to severe hemiparesis (66 with right and 63 with left hemiparesis) were studied within the first seven days of stroke. Twelve young, healthy control subjects were also recruited to investigate the influence of arm position on SaO2 and heart rate when lying on the right side and on the left side.
Non-invasive measurements of SaO2, BP and heart rate were taken at the bedside using a Kontron Kolormon 7250, with subjects in five randomly-assigned positions:
- Sitting in a chair;
- Propped up in bed;
- Non-paretic side down;
- Paretic side down.
Each position was maintained for 10 minutes, although not all patients could tolerate all five positions.
In patients who were able to get out of bed (generally those who had suffered a mild stroke), oxygenation was higher when sitting in a chair compared to any positions in bed. Acute stroke patients, regardless of the side of hemiparesis, and young healthy subjects had the poorest oxygenation (SaO2) when lying on their left side.
The difference in mean SaO2 when lying on the left compared to the right side was statistically significant in patients with a right, but not left, hemiparesis. About 10% of patients experienced clinically significant hypoxaemia (SaO2 ≤90%) when in specific positions for two minutes or more. Significantly more patients who experienced postural-related hypoxaemia than those who did not also had co-existing respiratory problems. (The main results of the influence of positioning on oxygen saturation in the acute phase of stroke have been published in Rowat et al, 2001).
There were no statistically significant changes in BP with positioning. However, 11% of patients experienced falls in systolic BP of ≥20mmHg, and 14% had increases in systolic BP of ≥20mmHg on rising from horizontal positions. Patients who experienced falls of ≥20mmHg were significantly more likely to be taking medication that caused postural hypotension than those who did not experience such a fall in BP.
Implications for nursing practice
In order to avoid adverse changes in SaO2 and BP, consideration should be given to the positioning of acute stroke patients as follows:
- Allowing conscious patients with adequate balance to sit in a chair may improve oxygenation.
- Patients with sitting balance benefit from sitting in a chair rather than being propped up in bed, because the propped-up position often becomes the 'slumped' position, and this may restrict lung expansion and reduce oxygenation.
- Patients with a chest infection or other respiratory problems are at risk of developing hypoxia, particularly if they have a right hemiparesis and are placed on their left side. It would be prudent to avoid placing patients on their left side or to monitor their SaO2 to detect and treat episodes of desaturation (SaO2 of 90% or below).
- Some patients, particularly those taking medication such as antihypertensives, antidepressants and sedation, are at risk of falls in BP when moving from supine to sitting or standing positions. If they are mobilised early, it might be prudent to measure supine and erect BP to detect and treat those who may experience postural BP dysregulation (change in systolic BP of ≥20mmHg).
Modifying current procedures for positioning may help to maintain normal SaO2 and BP levels, which could improve outcome after stroke. Intuitively, brief exposure to mild levels of hypoxaemia and hypotension may be particularly harmful to patients with acute stroke who have a potentially salvageable area of brain around the lesion, whereas those without acute brain damage can probably tolerate longer periods of more severe hypoxaemia and hypotension.
Other studies have shown that patients with head injuries frequently experience episodes of hypoxaemia and hypotension, which correlate with poor functional outcome (Jones et al, 1994). These adverse events were only detected by continuous monitoring, and might have been missed by standard clinical monitoring. Thus, in stroke also, more intensive physiological monitoring, rather than the four-hourly intermittent monitoring used in most wards, might enable earlier intervention. This would prevent or reduce serious complications, such as brain injury, and improve outcome after stroke. However intuitively logical, if not obvious, this hypothesis may be, its validity remains to be shown.
It is important to be aware that even if mild hypoxaemia and hypotension during positioning of acute stroke patients are related to a poor outcome, it does not necessarily imply that measures to reduce or prevent these episodes will improve outcome. These questions could only be answered definitively within the context of an RCT of adequate size (Signorini et al, 1999). However, such a trial would have to be extraordinarily large, including several thousand patients, since the incidence of hypoxaemia and hypotension during the positioning of acute stroke patients is relatively rare.
It would also be wrong to assume that introducing continuous monitoring would be beneficial and would pose no risk. Accurate non-invasive monitoring would take up much of nurses' time, and without additional nurses being made available, this might reduce their ability to do other parts of their job. To purchase, validate and maintain the monitoring equipment, and to train nursing staff in its use, would be extremely expensive.
Furthermore, the monitors restrict movement and require patients to keep still. The lack of movement may promote complications such as deep vein thrombosis or chest infection, or may reduce the speed of motor recovery if physiotherapy and mobilisation are delayed. Finally, nurses' and patients' frustration, as well as day-to-day practicalities such as hygiene and using the toilet, may result in the device being switched off. A large RCT looking at the feasibility, efficacy and cost-effectiveness of continuous non-invasive monitoring would be useful.
Even if such a trial indicates only a small benefit, it would still have a significant public health impact. After all, coordinated stroke unit care improves outcome after stroke more than current pharmacological interventions. Strategies to avoid adverse physiological events introduced in a coordinated widespread manner could have a substantial health impact.
In today's managed care environment, there are no reasons why nurses cannot be at the forefront of such a coherent approach. This should not only improve the quality and cost-effectiveness of stroke patient care, but may also increase the number of nurses interested in stroke research.