Researchers are taking a close look at clinical screening, instrumental assessment, and treatment procedures in an effort to help clinicians use the most effective tools efficiently and uniformly to make decisions about patient care. The clinical community is not in consensus on the accuracy and utility of screening procedures, nor is there agreement on the appropriate use of neuromuscular stimulation for dysphagia rehabilitation. Most agree, however, that one of the hallmark instrumental assessment procedures—the videofluoroscopic evaluation of swallowing (commonly called the modified barium swallow)—is clinically useful, but clinicians are not always accurate in interpreting it (McCullough, et al. 2001; Logemann et al., 2000). Clinicians treating dysphagia must continually read and evaluate new research studies on these "hot" topics to ensure they are providing the most current, evidence-based services to their patients.
The 3-Ounce Water Test
Debra M. Suiter and Steven B. Leder recently published their analysis of the 3-ounce water swallow test on more than 3,000 participants with a wide range of ages and diagnoses (Suiter & Leder, 2008). The results of the study are generating discussion among dysphagia experts. Suiter is an assistant professor in speech-language pathology at the University of Memphis, and Leder is professor in the Speech and Swallow Center at Yale University School of Medicine.
DePippo et al. (1992) first described the 3-ounce water test as a method of screening individuals at risk for oropharyngeal dysphagia and aspiration. Individuals are required to drink 3 ounces of water without interruption; those who stop, cough, choke, or show a wet-hoarse vocal quality during the test or for one minute afterward are considered to have failed. Suiter and Leder examined the clinical usefulness of the 3-ounce water swallow test for determining aspiration status and oral feeding recommendations in a large and heterogeneous population sample. Leder performed a fiberoptic exam (FEES®) on each patient. Criteria for passing the FEES were dependent on six boluses of 5 ml each (three puree and three thin liquid). After the exam, Leder administered the 3-ounce water swallow test.
Did the 3-ounce water swallow test in this study identify individuals who aspirated thin liquids? According to Suiter, the answer is an overwhelming yes. Ninety-eight percent of the individuals who passed the water swallow test did not aspirate during the FEES. The 3-ounce water swallow test appears to be a good predictor of an individual's ability to tolerate thin liquids (that is, a high negative predictive value). These negative predictive values remain high when analyzed for type of diagnostic category, ranging from 94.9% for post-cardiothoracic surgery to 100% for post-neurosurgery and brainstem stroke.
However, Leder noted that the converse is not true—failing the 3-ounce water swallow test does not mean it is unsafe for the individual to eat by mouth. He noted that 70.6% of those who failed the test were able to tolerate a diet of some kind based on the FEES and more than half were able to tolerate thin liquids based on the FEES results. If the patient fails the water swallow test, Suiter and Leder recommend proceeding directly to an instrumental assessment rather than a more comprehensive bedside/clinical exam. (Leder asserts that bedside screening procedures miss 20% of the patients, and Suiter agrees that no bedside test can reach the 99% accuracy of the water swallow test.)
The two researchers posed another question: If the patient passes the water swallow test, can a diet recommendation be made without further objective swallow assessment? To answer this question, they cross-tabulated the diet recommendation from the FEES and the water test results. Because only 1.5% of the patients (17 of the 1,151) who passed the water swallow test exhibited trace aspiration on the FEES results, Suiter contends that clinicians should feel confident placing patients on a diet without further assessments.
At the Speech and Swallow Center at the Yale University School of Medicine, Leder generally places a patient on a regular diet and thin liquids if the patient passes the water swallow test and has teeth/dentures. If edentulous, the patient is placed on a pureed diet and thin liquids. Both researchers agree that speech-language pathologists should weigh other factors such as cognitive status and overall medical condition before making diet recommendations.
The question of whether patients placed on a diet based on passing the water swallow test will eat and drink safely over the long term remains to be answered. Leder visited more than 500 of the patients in the study the day after they passed the water swallow. The criterion used to determine if the patients were successful was the amount of thin liquids they drank at meals. The researchers are currently analyzing this data. Longer-term follow-up is not possible because data are being collected in an acute-care hospital.
What about silent aspiration? Is the patient at risk when given 3 ounces of water to drink or placed on a diet with thin liquids without an instrumental study? Leder thinks further analysis of the data will support his hypothesis that "silent aspiration is volume-dependent and that patients only silently aspirate on smaller volumes." Therefore, if the patient is aspirating on 3 ounces, it will not be silent.
Others doubt that hypothesis. Cathy Lazarus, assistant professor in the Department of Otolaryngology at New York University School of Medicine, shared her data at the Dysphagia Research Society meeting in March 2009. She is administering 3 ounces of thin liquid barium during videofluoroscopic swallowing examinations and has tested 40 patients. Ten aspirated on the 3-ounce portion of the exam; of those, seven silently aspirated, but Lazarus noted that no aspiration was observed when those 10 patients used individual cup sips. Further, a chin tuck or head rotation eliminated aspiration in a majority of those patients. Lazarus recommended that these 10 patients have thin liquids with instructions to take one sip at a time, or use a head posture with sequential sips.
Leder admits he was skeptical of the 3-ounce water swallow test prior to the study but says the results have radically changed his practice. If he has placed a patient on a modified diet (e.g., thickened liquids) based on an instrumental study and the patient improves during the hospitalization, Leder does not repeat an instrumental before upgrading the diet recommendations. Instead he repeats the 3-ounce water swallow test. If the patient passes, he upgrades the diet. "If the 3-ounce water swallow test is valid at admission, it's just as valid for repeat tests over the course of the patient's recovery," says Leder.
The researchers recommend the 3-ounce water swallow test be incorporated into screening programs developed for nurses to use. It is not known whether nurses' interpretation of the patient's responses will be similar to those of an SLP (e.g., assessment of wet-hoarse vocal quality). The water swallow test will, however, result in over-referral to SLPs because of the high false-positive rate. Over-referral for instrumental exams is not necessarily unwelcome as it allows for greater objective identification of aspiration and gives the clinician more information on which to base diet and treatment recommendations.
As with any research findings, clinicians should carefully analyze the study's methodology. In this case, questions raised include possible examiner and order bias. The same examiner performed the FEES and the 3-ounce water swallow test and thus was not blinded. The water swallow test was always administered after the FEES which, in addition to order bias, raises the question of patient fatigue affecting performance. Patients were not randomized in any way. In addition, criteria for passing the FEES ere based only on six 5-ml boluses. A 3-ounce liquid challenge was not administered during the instrumental exam, which would have made the two tests more comparable.
Swallowing Impairment Profile
Giving the clinician more objective and standardized information from an instrumental study is the goal of Bonnie Martin-Harris and colleagues at the Medical University of South Carolina (MUSC) and Saint Joseph's Hospital of Atlanta, Ga. The research team created the first standardized tool to measure swallowing impairment based on judgments of structural movement relative to bolus flow from videofluoroscopic images using standardized bolus volumes and consistencies (Varibar®, Bracco Diagnostics, Inc.) and the Modified Barium Swallowing Impairment Profile (MBSImP). The research was funded by the National Institute on Deafness and Other Communication Disorders (NIDCD).
The MBSImP includes 17 physiologic components ranging from lip closure to esophageal clearance. An example of a five-point rating scale on the component "initiation of pharyngeal swallow" ranges from "0" when the pharyngeal swallow initiates (onset superior-anterior hyoid movement) as the bolus head approximates the posterior ramus of the mandible, to a "4" if there is no appreciable initiation of hyoid movement at any bolus location.
Components 1-16 fit a two-factor model: oral factor (swallowing components related to movements of the oral tongue) and pharyngeal factor (swallowing components related to pharyngeal clearance and airway protection). Esophageal clearance (Component 17) stood alone in the analysis. It was interesting that Component 6, "initiation of the pharyngeal swallow," is grouped in the oral factor vs. the pharyngeal factor, a classification researchers say likely relates to the dependence of timely pharyngeal swallowing initiation on the integrity of oral tongue movement, in keeping with historical findings by Logemann and others.
The MBSImP permits the development of a treatment plan directed toward the physiologic impairment exhibited by the patient and devises compensatory strategies that target safe and efficient bolus flow.
The study demonstrated that it was unnecessary to score each swallow for every volume and consistency; rather, an overall impression (Overall Impression Score, or OI) of the impaired components across all textures could be captured. This method significantly decreases the labor intensity of utilizing the MBSImP for clinical purposes. Fifteen minutes is the average time for analysis of a complete, recorded examination by a trained clinician. Scoring requires frame-by-frame analysis.
The research team found that the oral and pharyngeal impairment scores were significantly associated with penetration-aspiration scores from the Penetration-Aspiration Scale (PAS), which is used in conjunction with the MBSImP. However, many patients who exhibited swallowing impairment with nutritional and quality-of-life implications did not have penetration or aspiration. Martin-Harris reminds us that "we teach patients to compensate for penetration and aspiration via posture or maneuver, but we apply evidence-based exercises to the physiologic impairment (tongue base retraction, laryngeal closure, superior-anterior hyoid excursion, etc.). Penetration and aspiration are not necessary or sufficient measures of swallowing impairment."
The standardized boluses of 5-ml thin and 5-ml nectar-thickened liquid reliably captured impairment for the majority of the swallowing components. The ability of these two bolus types alone to capture impairment suggests that they could be used as screening swallows during a videofluoroscopic exam. Impairment found in any component using these bolus types may signal the need to continue with the examination. Similarly, lack of impairment found using these two bolus types may permit discontinuation if clinical yield of administering additional bolus types would not outweigh the risks of additional X-ray exposure.
Martin-Harris is not ready to promote this practice without a longitudinal study comparing the clinical outcome of abbreviated and full-protocol videofluoroscopic swallowing evaluations. She believes the findings speak to the potentially misguided practice of foregoing the thin-liquid swallow because of the perception that the patient will perform better (i.e., less aspiration) on a more viscous bolus. These data suggest that it is possible to miss substantial physiologic information when using such practices.
For clinicians who would like to use the standardized protocol, terminology, and scoring system reported in the study, Web-based training modules are being developed on the administration and scoring of the MBSImP tool. Each component will be represented by videofluoroscopic images across patient populations with corresponding three-dimensional animations. Martin-Harris plans to have prototypes ready for peer review later this year and plans to offer ASHA CE credit for the training.
Martin-Harris plans to develop a registry of de-identified dysphagic patients undergoing videofluoroscopy by clinicians who have achieved the 80% reliability scoring criterion. Clinicians who enter the registry will be provided with a standardized electronic summary report of the evaluation that can be modified to fit their institutional needs. This national registry of VFSS exams and MBSImP scoring would provide the opportunity for involvement in and development of multi-site clinical research.
Electrical Stimulation in Adults
A good review of what we do and don't know about neuromuscular electrical stimulation (NMES) for swallowing in adults was provided by Christy Ludlow in the March 4, 2008, issue of The ASHA Leader. Ludlow is chief of the Laryngeal and Speech Section, National Institute of Neurological Disorders and Stroke, National Institutes of Health, in Bethesda, Md. When asked what is new since that time, Ludlow points to the study by Bulow et al. (2008) that compared traditional therapy to NMES in patients with stroke at three European swallowing centers. The pre- and post-ability to swallow was assessed videographically as well as with analysis of nutritional status and oral motor function. The study found that patients in both groups (traditional treatment and NMES) made progress, but there was no statistical difference between the two groups. Ludlow points out that the study might be underpowered as there were only 25 subjects; she also notes that some researchers are beginning to look at submental placement of electrodes, but until this is evaluated no judgment of effectiveness can be made. Ludlow continues to suspect that the effects seen for patients receiving NMES may be related to the intensity of treatment received rather than to the specific methodology of NMES.
Dysphagia is an area of practice that changes rapidly and in order to provide the best, evidence-based care, clinicians are compelled to stay updated. They should take advantage of multiple opportunities to do so.