12.1.1. Clinical introduction

Aspiration pneumonia may be a devastating complication of stroke and is associated with increased mortality and poor outcomes. 40% of patients with stroke acutely have swallowing difficulties, some of which may not be detectable clinically. Those assessed as having dysphagia are recommended a variety of options for oral intake, from no oral intake, through modification of fluid (small volumes, thickened fluids) and food (puree or soft consistency) to normal intake. It is common sense that patients who aspirate oral contents should not be fed orally, but little is known about the safety of water by mouth. Normal saliva production is 1–2 litres per day, so even patients denied access to oral fluids are aspirating large fluid volumes. Patients are often much more distressed by withdrawal of fluids than of food, and despite adequate fluid replacement by other means (intravenous or subcutaneous infusions or tube feeding) feel thirsty and have a dry mouth. Oral hygiene is very much more difficult to maintain in patients denied oral fluids, and aspiration of infected saliva may contribute to the development of pneumonia.

The clinical question to be addressed is how best to reduce the likelihood of patients with acute stroke developing aspiration pneumonia.

12.1.2. Clinical methodological introduction

One RCT was identified that compared acute stroke patients within 3 weeks of stroke onset with previously identified thin liquid aspiration (verified by videofluoroscopy) (N=20). Groups were randomised between those who were given thickened liquids plus additional water compared with those given thickened liquids only.¹⁸⁷ Level 1+

The patient population had a mean age of 77 years. In the group given thickened liquids with additional water, five of the patients had consistent aspiration of thin liquids (>50% of swallows), four had inconsistent aspiration of thin liquids (10 to 15% of swallows) and one had infrequent aspiration of thin liquids (<10% of swallows). Seven of the patients had an absent protective cough reflex at the level of the vocal folds. Level 1+

In the control group (thickened liquids only), two patients had inconsistent aspiration and six infrequent aspiration. Five of the patients did not have a protective cough reflex at the level of the vocal folds. Level 1+

12.1.3. Health economic methodological introduction

No papers were identified.

12.1.4. Clinical evidence statements

12.1.5. From evidence to recommendations

There is insufficient evidence addressing the withdrawal or modification of oral intake in the prevention of aspiration pneumonia after stroke. One study showed that the modification of oral intake through thickened fluids was not as unsafe as the group had expected. No studies were identified which addressed the issue of giving water to patients who are assessed as having dysphagia for fluids. The withdrawal or modification of oral intake when trying to prevent aspiration pneumonia is a contentious issue amongst healthcare professionals and the insufficient evidence highlights the need by the group for a research recommendation in these areas. There was a consensus that an important reason for giving patients water would be to enable them to maintain adequate oral hygiene and to reduce patient distress. The consensus of the group was that there was insufficient evidence on which a recommendation could be made.

12.1.6. RECOMMENDATION

R56.

In people with dysphagia, food and fluids should be given in a form that can be swallowed without aspiration following specialist assessment of swallowing.