Current issues of ACP Journal Club are published in Annals of Internal Medicine


Swallowing problems after stroke were common and associated with continued swallowing problems, chest infection, and aspiration

ACP J Club. 1999 Sept-Oct;131:51. doi:10.7326/ACPJC-1999-131-2-051

Source Citation

Mann G, Hankey GJ, Cameron D. Swallowing function after stroke. Prognosis and prognostic factors at 6 months. Stroke. 1999 Apr;30:744-8. [PubMed ID: 99205325]



In patients with a first stroke, is swallowing dysfunction measured within a median of 3 to 10 days after symptom onset associated with death, recurrent stroke, or chest infection at 6 months?


Inception cohort followed for 6 months.


A stroke unit of a university hospital in Perth, Western Australia, Australia.


128 patients (70% ≥ 65 y, 64% men) with a confirmed first stroke (World Health Organization criteria). Patients were conscious, medically stable, and had no history of swallowing impairment or a medical condition that could alter swallowing function. Follow-up was 91%.

Assessment of prognostic factors

Clinical syndrome, pathology and etiology of stroke, disability and handicap, age, and sex. Clinical swallowing abnormalities (dysphagia) included delayed oral transit, incomplete oral clearance, weak or absent cough, aspiration, and "wet" voice. Fluoroscopic abnormalities included delayed oral transit, delayed or absent swallow reflex, incomplete oral or pharyngeal clearance, penetration of the false vocal cords, and aspiration.

Main outcome measures

Death, recurrent stroke, chest infection, persistent swallowing abnormalities, need for a different diet, and fluoroscopic aspiration at 6 months.

Main results

At 6 months, 56 patients had a persistent clinical swallowing abnormality. 5 patients who died had a baseline fluoroscopic swallowing abnormality. Multiple logistic regression showed that chest infection (26 patients) was associated with delayed or absent fluoroscopic swallowing reflex (odds ratio [OR] 12, 95% CI 3.3 to 50). Swallowing abnormalities (56 patients) were associated with fluoroscopic penetration (OR 17, CI 4.8 to 59), fluoroscopic delayed oral transit (OR 6, CI 1.7 to 24), and any clinical swallowing impairment (OR 4, CI 1.3 to 15). The need for a different diet (15 patients) was associated with delayed oral transit (OR 32, CI 4.1 to 261). The combined end point of swallowing abnormality, chest infection, or fluoroscopic aspiration (59 patients) was associated with laryngeal penetration (OR 14, CI 4 to 51), delayed oral transit (OR 14, CI 4 to 50), age > 70 years (OR 5, CI 1.4 to 21), and male sex (OR 5, CI 1.5 to 18).


Swallowing abnormalities were common after stroke and were associated with chest infections, videofluoroscopic aspiration, and persistent swallowing abnormalities at 6 months.

Sources of funding: Medical Research Foundation of Western Australia; Sandoz Gerontological Society; Royal Perth Hospital Medical Research Foundation.

For correspondence: Dr. G.J. Hankey, Royal Perth Hospital, Wellington Street, Perth, Western Australia, Australia 6001. FAX 61-89-224-3323.


The study by Mann and colleagues on swallowing after stroke with 6-month follow-up shows that delayed or absent swallowing reflex early after stroke detected by videofluoroscopy is common and predicts chest infection. This finding is important because medical complications after stroke, such as aspiration pneumonia, can increase the cost of care and contribute to further disability, morbidity, and death.

Delay in the pharyngeal swallowing reflex after stroke occurs often and is associated with aspiration caused by spillage of oral bolus into the laryngeal vestibule and through the true vocal cords (1). Use of a thickening agent increases liquid viscosity, which can prevent penetration of swallowed bolus into the larynx. If thickened liquids and other strategies fail to prevent aspiration, an alternate method of feeding, such as gastrostomy, is suggested.

Given that fluoroscopic aspiration is found in about 45% of patients with acute stroke, all such patients should be considered at risk for aspiration pneumonia. Bedside swallowing evaluation by a qualified speech pathologist before feeding is recommended for acute stroke care protocols (2). If swallowing dysfunction is suspected, videofluoroscopy assessment is useful to verify aspiration and to test whether modified food consistency or various swallowing techniques can prevent aspiration.

Currently, the natural history of silent aspiration is unknown. Silent aspiration is neither a sensitive nor specific predictor of chest infection (2). Weak cough, oral-motor problems, poor cognition, and general debilitation may also contribute to risk for pneumonia. In addition, the effect of diet modification on prevention of pneumonia has not been investigated, and most studies, including this one, do not control for dietary management of dysphagia. Until such trials are done, prevention of any laryngeal aspiration will be the focus for averting aspiration pneumonia.

Richard L. Harvey, MD
Northwestern University Medical SchoolChicago, Illinois, USA


1. Johnson ER, McKenzie SW, Rosenquist CJ, Lieberman JS, Sievers AE. Dysphagia following stroke: quantitative evaluation of pharyngeal transit times. Arch Phys Med Rehabil. 1992;73:419-23.

2. DePippo KL, Holas MA, Reding MJ. Validation of the 3-oz water swallow test for aspiration following stroke. Arch Neurol. 1992;49:1259-61.