14 Aspiration in Patients With Acute Stroke Stephanie K. Daniels, MS, Kevin Brailey, PhD, Daniel H. Priestly, AID, Lisa R. Herrington, BS, Leon A. Weisberg, MD, Anne L. Foundas, MD ABSTRACT. Daniels SK, Brailey K, Priestly DH, Her- rington LR, Weisberg LA, Foundas AL. Aspiration in patients with acute stroke. Arch Phys Med Rehabil 1998;79:14-9. Objectives: To determine the frequency and clinical pre- dictors of aspiration within 5 days of acute stroke. Design: Case series. Setting: Tertiary care center. Patients: Consecutive stroke patients (n = 55) with new neurologic deficit evaluated within 5 days of acute stroke. Main Outcome Measures: Comparison of features identified on clinical swallowing and oromotor examinations and occur- rence of aspiration (silent or overt) evident on videofluoroscopic swallow study (VSS). Results: Aspiration occurred in 21 of 55 patients (38%). Whereas 7 of 21 patients (33%) aspirated overtly, 14 (67%) aspirated silently on VSS. Chi-square analyses revealed that dysphonia, dysarthria, abnormal gag reflex, abnormal volitional cough, cough after swallow, and voice change after swallow were significantly related to aspiration and were predictors of the subset of patients with silent aspiration. Logistic regression revealed that abnormal volitional cough and cough with swal- low, in conjunction, predicted aspiration with 78% accuracy. Conclusions: Silent aspiration appears to be a significant problem in acute stroke patients because silent aspiration oc- curred in two thirds of the patients who aspirated. The prediction of patients at risk for aspiration was significantly improved by the presence of concurrent findings of abnormal volitional cough and cough with swallow on clinical examination. This is a US government work. There are no restrictions on its use. A SPIRATION, the subglottic penetration of liquid or food, is a frequent sequela of dysphagia and stroke. As identified by videofluoroscopic swallow study (VSS), aspiration occurs in 40% to 70% of stroke patients] '2 Silent aspiration, defined as the subglottic penetration of a bolus without elicitation of a cough reflex, occurs in approximately 40% of dysphagic pa- tients who aspirate) '4 Dysphagic patients who aspirate are at increased risk of developing aspiration pneumonia. Specifically, the development of pneumonia is seven times greater in stroke patients who aspirate versus those who do not, and six times greater in patients who aspirate silently as compared with those who cough upon aspirationfl '6 Johnson and colleagues7 found that approximately 50% of stroke patients with dysphagia severe enough to warrant videofluoroscopic evaluation developed aspi- ration pneumonia. In addition, aspiration pneumonia, which is Fromthe Audiology/Speech PathologyService(Ms. Daniels, Ms. Herrington), Psychology Service (Dr. Brailey),RadiologyService(Dr. Priestly),Neurology Service (Dr. Foundas), VeteransAffairsMedicalCenter;and the Department of Psychiatryand Neurology, Tulane University Schoolof Medicine (Drs. Weisberg and Foundas), New Orleans,LA. Submittedfor publication March 3, 1997. AcceptedJune 20, 1997. No commercialparty having a direct financial interest in the results of the research supportingthis article has or will confer a benefituponthe authors or upon any organization with whichthe authorsare associated. Reprintrequeststo Stephanie K. Daniels, MS, SpeechPathology Service(126), VA MedicalCenter, 1601 PerdidoStreet, New Orleans,LA 70146. This is a US government work. There ~c no restrictionson its use. 0003-9993/98/7901-441850.00/0 commonly associated with dysphagia, is the fourth most fre- quent cause of death in the elderly, a Given these data, it is imperative that acute stroke patients at risk for developing aspi- ration pneumonia be identified early in their clinical course to prevent increased morbidity and mortality. Furthermore, it is important to identify whether specific clinical features accu- rately predict aspiration. Chronic stroke patients with persistent dysphagia and risk of developing aspiration have been studied using clinical and videofluoroscopic examinations. In a study of stroke patients 1 to 24 months after stroke, Homer and colleagues,9 using VSS, found that aspiration occurred in half of their patients. Aspira- tion occurred more often in patients with bilateral cranial nerve signs (71%) as compared with patients with unilateral cortical signs (29%). Dysphonia was the most common clinical feature in patients with aspiration. In a subsequent study, Homer and Massey ~ studied stroke patients within 28 months of the acute stroke and found that 11 of 21 (52%) patients aspirated during VSS. Furthermore, 8 of these 11 patients were silent aspirators. On clinical examination, complaints of dysphagia, weak cough, and dysphonia distinguished the aspirating patients from nonas- pirating patients. In addition, Homer' s group m'~ retrospectively studied patients with bilateral strokes and found that the coexis- tence of abnormal gag and voluntary cough were more highly predictive of aspiration than either of these clinical findings in isolation. Linden and Siebens2 clinically and fluoroscopically examined 15 patients with unspecified central nervous system damage 1 to 46 months after onset. Either decreased gag reflex or a wet, hoarse vocal quality were evident in 90% of the patients who aspirated, and both occurred in 9 of the 11 aspirat- ing patients. Furthermore, aspiration did not elicit a cough refex in 82% of the patients. Linden and colleagues ~2 studied 249 patients primarily with neurologic etiologies such as stroke, although time from onset to evaluation was not indicated. Nine clinical indicators were significantly associated with aspiration as identified by videofluoroscopy including: recumbent posture, dysphonia, wet phonation, decreased/absent laryngeal excur- sion, wet spontaneous cough, decreased ability to swallow se- cretions, decreased palatal gag, harsh phonation, and breathy phonation. Additional studies have used water swallow tests, measuring time and/or clinical indicators as predictors of dys- phagia and aspiration, but as with studies by Homer and Linden, these investigations were not limited to acute stroke patients. 13-~6 Other investigators have studied dysphagia in acute stroke patients. Gordon and colleagues 17identified dysphagia in 41 of 91 (45%) consecutive patients, half of whom were evaluated within 48 hours after onset. Elicitation of a cough during a water swallow test was used to identify and determine duration of dysphagia in this study. Using a water swallow test, Barer 18 also studied 357 stroke patients within 48 hours of onset and found that 30% of the patients had dysphagia acutely, but only 6% had persistent dysphagia 1 month after stroke. Odderson and coworkers j9 evaluated stroke patients within 1 day of admis- sion and found that 39% failed the initial swallowing screen, and 19% required enteral nutrition before discharge from acute care. Although these three studies 17-19 evaluated acute stroke patients, videofluoroscopy was not used to confirm the occur- rence of dysphagia and aspiration. In contrast, Kidd's group2° Arch Phys Med RehabU Vol 79, January 1998