Results of the largest multi-institutional randomized clinical trial in swallowing were presented at the 2006 ASHA convention in November. Participating in the study were 126 institutions, including 47 acute care hospitals and 79 sub-acute facilities, in addition to patients living at home who had a consistent caregiver. The 10-year study randomized a total of 742 patients.
The study comprised two sequential clinical trials, referred to as Part I and Part II. Part I included patients with Parkinson's disease (PD), with or without dementia, and patients with dementia only. Speech-language pathologists referred patients to the study because they were suspected of aspirating thin liquids. The patients received six qualifying swallows of thin liquid barium—three 3-ml and three gulps from a cup—while examined videofluorographically. Patients who aspirated on at least one of the six qualifying swallows were eligible to participate. After consent was obtained, those patients were given thin liquid swallows, using three randomized interventions to eliminate their aspiration: chin-down posture, nectar-thickened liquid barium (viscosity: 300 centipoise [cps]), and honey-thickened liquid barium (3,000 cps).
Part I of the protocol examined the immediate effectiveness of the three interventions. Results indicated that 25% of the patients exhibited elimination of their aspiration with all three interventions, while 49% exhibited aspiration on all of the interventions. Honey-thickened liquid was most often successful in eliminating aspiration in the short term, followed by nectar and then chin-down posture. The difference in frequency ranged from a 53% aspiration rate for honey to a 68% aspiration rate for chin-down. Data collected from PD patients without dementia revealed preference for chin-down posture, followed by nectar and then honey.
Results of Part I of this clinical trial encourage clinicians to introduce treatment strategies into their physiologic evaluations of swallow to define what is most appropriate for a particular patient. Introducing all three of these interventions in patients who aspirate thin liquids helps determine which intervention best eliminates aspiration immediately and which successful intervention the patient prefers.
Part II of the study involved patients who aspirated despite all three interventions and those who did not aspirate on any of three interventions. Ethically, patients who successfully eliminated aspiration on only one or two of the three interventions could not be included in Part II, as it involved randomizing the patients to one of the interventions for three months to determine pneumonia rates during that time. Of the 515 patients included in Part II, 52 developed pneumonia in the three months of follow-up. The difference in the frequency of pneumonia was not significant between chin-down or thickened liquid treatment groups (11.6% vs. 9.8%). However, when the rate of pneumonia related to each of the two levels of thickened liquid was compared, honey-thickened liquids resulted in a more than a two-fold pneumonia risk after adjustment for risk factors (p=.06), defined as dementia diagnosis or aspiration on all three Part I interventions.
The honey-thickened liquid used in this study was extremely thick (3,000 cps). This result must be examined in relation to other viscosities labeled as honey-thickened liquids. The concept that "thicker is better" for liquid aspiration management in this specific population is not necessarily true.
Impact of Findings
Results of Part I and II of this study emphasize the importance of examining the effectiveness of dysphagia treatment interventions in general, and liquid aspiration specifically, as critical to defining optimal intervention for patients with oropharyngeal dysphagia. These data also emphasize that many patients with cognitive deficits can derive benefit from dysphagia intervention. The order of effectiveness of the three interventions was the same for all three populations. However, the fact that not all patients received the same benefit from the various interventions emphasizes the need for assessing treatment effects during the patient's instrumental evaluation of swallow.
Results of this study also emphasize the importance of randomized multi-institutional clinical trials in defining the evaluation protocol for patients with dysphagia. Findings also emphasize that clinical assumptions such as "the thicker the better" require study to define more accurately the optimal foods and fluids for various patient groups. The patient with dysphagia deserves the best care, which includes a careful physiologic assessment of swallow disorders and the effectiveness of various rehabilitation strategies. Additional research is needed to follow this trial, including definition of a honey-thickened liquid viscosity that does not result in increased pneumonia in the populations studied.
It is likely that the increased pneumonia in the patients who remained on honey-thickened liquids for three months resulted from the reduced ability of the chest's ciliary action to clear thicker material from the lungs. However, this theory requires additional attention, in that a percentage of research participants responded best to the honey-thick condition in Part I and therefore were treated with that intervention and unable to participate in Part II. Pneumonia results on those specific, and perhaps best, candidates are important for future analysis.
A more complete description of this multi-institutional randomized clinical trial will be published in a late spring issue of The ASHA Leader.