Screening Test for Silent Aspiration at the Bedside
Many screening tests for dysphagia can be given at bedside. However, they cannot accurately screen for silent aspiration (SA). We studied the usefulness of a cough test to screen for SA and combined it with the modified water swallowing test (MWST) to make an accurate screening system. Patients suspected of dysphagia (N = 204) were administered a cough test and underwent videofluorography (VF) or videoendoscopy (VE). Sensitivity of the cough test for detection of SA was 0.87 with specificity of 0.89. Of these 204 patients, 107 were also administered the MWST. Fifty-five were evaluated as normal by the screening system, 49 of whom were evaluated as normal by VF or VE. Sixteen were evaluated as “SA suspected” by the screening system; seven of them were normal, and seven were evaluated as having SA by VF or VE. Nineteen were evaluated as aspirating with cough, 14 of whom had aspiration with cough as shown by VF or VE. Seventeen were evaluated as having SA, 15 of whom had SA shown by VF or VE. The cough test was useful in screening for SA. Moreover, a screening system that included MWST and a cough test could accurately distinguish between the healthy who were safe in swallowing and SA patients who were unsafe.
KeywordsDeglutition Deglutition disorders Screening test Silent aspiration Cough test
Videofluorography (VF) and videoendoscopy (VE) are considered the most accurate methods for the examination for dysphagia. However, many institutions lack the equipment required for these diagnostic tests. Thus, alternative screening studies have been developed that easily can be performed at the patient’s bedside. The criteria for many current screening tests include the presence of a cough [1, 2, 3, 4]. Therefore, it is possible that diagnoses are missed in patients who show aspiration without cough (silent aspiration, SA) . Furthermore, clinical bedside assessment has been shown to miss up to 40% of dysphagia patients with SA .
One study reported that SA was present in more than 70% of elderly patients with community-acquired pneumonia . SA can cause aspiration pneumonia in the elderly . Patients with laryngeal penetration, tracheobronchial aspiration, or silent tracheobronchial aspiration were approximately 4 times, 10 times, and 13 times, respectively, more likely to develop pneumonia than those with normal swallowing . There is no clear evidence that SA relates to sensory loss. SA may relate to gradual desensitization if aspiration is chronic . SA has been reported to occur in over 40% of patients referred for dysphagia evaluation in a rehabilitation hospital and in as many as 77% of ventilator-dependent patients . Therefore, a screening test for SA is clinically valuable.
We studied the usefulness of a cough test to screen for SA. The cough test uses citric acid to detect SA. Because the cough test looks at a patient’s cough rather than aspiration, we then combined the cough test methods with other usual screening tests for aspiration and investigated whether the combined screening system could improve the accuracy of dysphagia diagnosis.
Participants in this study were 204 patients (131 men and 73 women) who were suspected of having dysphagia and underwent VF or VE. They included both inpatients and outpatients who had several clinical symptoms of dysphagia such as coughing with eating, difficulty with eating, a fever or pneumonia by aspiration, and past history of suffocation. None had a history of asthma. The average age was 69.90 ± 11.70 years (mean ± SD) and ranged from 18 to 100 years. The following preexisting conditions were found: 39% had cardiovascular disease (CVD), 24% head or neck cancer, 17% neuromuscular disease, 15% respiratory disease, and 5% other diseases. This study received approval from the ethics committee of the faculty of dentistry of the Tokyo Medical and Dental University. Informed consent was obtained from all participants or their legal guardians. The study conformed to the Tokyo Medical and Dental University policies concerning research on human subjects.
In the cough test, patients inhaled a mist of citric acid-physiologic saline orally for 1 min with an ultrasonic nebulizer. The inspector observed the number of times the patient coughed during the 1 min of nebulizing. More than five coughs was considered negative (normal), while less than four coughs was regarded as positive [12, 13, 14]. The concentration of citric acid was 1.0 w/v% . Subjects were directed to breathe through the mouth mask because inhaling through the mouth allows for a higher concentration of inhaled particles than does inhaling through the nose . If patients had difficulty following these instructions, a nose clip was applied to encourage the patient to inhale orally. The ultrasonic nebulizer used was an OMRON NE-U17 with a particle size of 1–8 μm and output rate greater than 3 ml/min. The cough test was conducted on the day before, the same day, or the day after the VF or VE examination.
The cough test results were combined with the modified water swallowing test (MWST) for diagnosis . Cold water (3 ml) was placed on the floor of the mouth using a 5-ml syringe. Placement on the floor of the mouth prevented premature spillage of test water into the pharynx. The patient was then instructed to swallow. If the patient was unable to swallow, a score of 1 was given. If the patient was able to swallow but experienced dyspnea (difficulty breathing) after swallowing, a score of 2 was given. If the patient was able to swallow and experienced coughing or wet-hoarseness after swallowing, a score of 3 was given. Otherwise, the patient was asked to perform two dry (saliva) swallows. If the patient was able to swallow the water but unable to perform either of the two dry swallows, a score of 4 was recorded. If the patient was able to complete the water and both dry swallows, a score of 5 was recorded. The entire procedure was repeated twice more and the final score was defined as the lowest score on any trial.
MWST was not conducted on patients clinically regarded as severe saliva aspirators. As a result, the number of the patients receiving both screening tests was 107 (83 men and 24 women),with an average age of 69.23 ± 13.09 years.
All patients received VF (Medix-900DR, Hitachi Medical Corp) or VE (Olympus ENF-P4) assessment of swallowing to detect aspiration. VF was conducted followed by modified barium swallow (MBS)  and VE was followed by fiberoptic endoscopic evaluation of swallowing (FEES) . The consistencies of the test foods used were thin liquid, thick liquid, cookies, and cornflakes. All the foods used in the VF included barium sulfate. Thin and thick liquids were colored with green food dye for ease of visualization in the VE.
Using the results of the VF or VE examination as the standards, the sensitivity, specificity, efficiency, positive predictive value (PPV), and negative predictive value (NPV) for SA detection were calculated. Then the results of the screening system combining the cough test and the MWST were compared with the results of VF or VE.
Results of the Cough Test
Results of cough test
Aspiration with cough
SA by little aspiration
Screening for SA Using the Cough Test
Screening of SA by cough test
Screening of SA by cough test
Screening of SA by cough test in patients with aspiration
Screening of SA by cough test in patients with aspiration
Results of the Cough Test and MWST
The Method of the Cough Test
The medical efficacy of a nebulizer is influenced by particle size and the volume of particles attained in one inhalation . Two main parameters are generally used to evaluate the performance of nebulizers: the droplet size distribution of the aerosol and the drug output rate . The particle size most effective at adhering to peripheral tracheal branches and tracheal branches is 1–8 μm and to the larynx 10–20 μm [20, 22]. The ultrasonic nebulizer makes microscopic (0.5–5.0 μm) particles of uniform size consistently, while the size of particles produced by a jet nebulizer is larger and influenced by the jet speed and water current blown off and thus varies widely .
Studies have shown that breathing through the mouth with a tube deposits more particles in the lower respiratory tract compared with nasal breathing and mouth breathing without a tube . Iwata  studied the best way to deliver particles to the airway and determined that the best way was through the mouth with an ultrasonic nebulizer, the second was through the nose with an ultrasonic nebulizer, the next was through the mouth using a jet nebulizer, and the worst was through the nose using a jet nebulizer. Furthermore, a similar study showed that an ultrasonic nebulizer deposited more particles to the middle and lower pharynx compared to a jet nebulizer. Moreover, when breathing deeply, the ultrasonic nebulizer carried particles more deeply compared to eupnea normal, unlabored breathing .
In summary, the previous literature indicates that to stimulate the airway one should use an ultrasonic nebulizer with deep breathing through the mouth with a tube. Therefore, our cough test method was thought to be appropriate.
Usefulness of the Cough Test for Screening for Dysphagia
There are many studies of swallowing assessments that can be performed at bedside. Because of the high number of patients with dysphagia, the bedside examination for detecting aspiration is commonly used without VF and VE evaluation.
The water swallowing test has several variations and is widely used. The volume of water varies from 3 cc to 85 cc [1, 2, 3, 4, 17, 26]. McCullough et al.  states that the test bolus of 5 cc may be superior to 3 oz (85 cc) because the 3-oz water test had high sensitivity (86%) but low specificity (50%). Most of these water-swallowing tests look for a patient’s cough, which can be a weakness of these tests when attempting to diagnose silent aspirators.
Pulse oximetry is a noninvasive way of performing a bedside swallowing test [27, 28, 29], but several studies have shown no relationship between aspiration and desaturation [30, 31], which occurs easily during a postural change, cough, or swallow, in addition to during aspiration . Thus, the usefulness of SpO2 for diagnosing aspiration is now controversial.
Similarly, there have been many screening tests for dysphagia [33, 34, 35, 36, 37]. In general, the sensitivity of these tests was high but the specificity was low. In addition, many of these tests have weaknesses in practical use; for example, a test may require a highly trained examiner or have limited applicability to the patients. Furthermore, the usefulness of these tests to screen for aspiration has been reported, but the ability to detect SA is not mentioned.
Splaingard et al.  evaluated 107 patients using a bedside clinical examination and VF and monitored swallowing of various volumes and consistencies of food. The bedside evaluation identified only 42% of the patients who aspirated on VF. Twenty percent of the entire sample who did not show any symptoms of aspiration on the bedside evaluation had aspiration during VF. They concluded that these screening tests were easy to perform without special instruments and possessed good reproducibility, but they were not able to detect saliva aspiration and SA.
There are many studies on the relationship between cough and aspiration or pneumonia [39, 40, 41, 42, 43], some of which have examined the cough test with respect to aspiration and pneumonia. Addington et al. [44, 45, 46, 47] studied the reliability of the cough test in detecting the risk of pneumonia in acute stroke patients. They concluded that the cough test was sufficient to assess the cough reflex and risk of aspiration pneumonia in stroke patients. Sekizawa et al.  used the cough test to investigate whether weakness of the cough reflex was related to aspiration pneumonia. Patients with aspiration pneumonia did not cough at the highest concentration. Hammand et al.  analyzed the coughs of aspiration patients and healthy subjects. Objective analysis of the cough was sufficient to detect aspiration without stress. Patients with an absent or weak cough reflex were at a higher risk of aspiration. They also assessed the cough reflex and aspiration of patients with acute CVA, and the cough test showed a sensitivity of 68% and specificity of 82%. Nakazawa et al.  determined that the thresholds of the cough test and swallow reflex induction test of patients who developed pneumonia were higher than those of healthy subjects. Emergence of pneumonia was related to weaknesses in the airway protective reflex and swallowing ability. However, SA detection was not mentioned. Horner et al.  investigated the gag reflex, cough reflex, and swallowing reflex of acute stroke patients and found that more than 60% of SA patients showed a weakened cough response.
In summary, the relationship of the cough test and aspiration or pneumonia has been reported in the past; however, these studies did not specifically attempt to detect SA. Furthermore, many of these studies demonstrated that a weak or absent cough response resulted in a higher risk of aspiration pneumonia.
The cough test is inexpensive and easy to perform, puts little stress on the patient, and enables a fast diagnosis of swallowing difficulty. It can be performed on patients who have difficulty following instructions, who are at high risk, and who have low immunity. From the results of our study, the cough test has good statistical reliability in the detection of SA. Therefore, the cough test is a useful SA screening tool.
Usefulness of the Combined Screening System
The bedside examination has limitations in accurately predicting or detecting the occurrence of aspiration. By combining tests that assess different important points clinically, the accuracy will increase . Several studies have examined a combination of screening tests. Smith et al.  combined oxygen saturation and standard bedside swallowing assessments on 53 patients with acute stroke confirmed by CT. Sensitivity and specificity of bedside swallowing assessment were only 80% and 68%, respectively, and those of O2 desaturation were only 87% and 39%. However, the combination of bedside assessment and O2 desaturation were 73% and 76%. Lim et al.  combined the oxygen saturation test and the 50-ml water-swallowing test. Sensitivity and specificity of only oxygen desaturation of 2% or more alone were 76.9% and 83.3%, and those of the water-swallowing test alone were 84.6% and 75.0%. On the other hand, the combination of these tests had 100% sensitivity and 70.8% specificity. Similarly, Tohara et al.  studied the accuracy of three nonvideofluorographic (non-VFG) tests for assessing risk of aspiration: MWST and the food test (4 g of pudding) and the X-ray test (static radiographs of the pharynx are taken before and after swallowing liquid barium). When MWST was combined with the food test, sensitivity was 90% and specificity was 56%. When all three non-VFG tests were combined, sensitivity was 90% and specificity was 71%.
We combined the MWST and the cough test because we wished to assess the swallowing reflex with the MWST and assess the airway protective reflex with the cough test. We chose to use the MWST because it is safe, requires little volume, and its appropriateness was reported statistically. In our combination screening system of MWST and the cough test, 89.1% of the predicted normal group were actually normal, 73.7% of the aspiration with cough group were actually aspirators with cough, and 88.2% of the SA group were actually silent aspirators. The confidence of the combined screening system is thought to be clinically high. Furthermore, the “SA suspected” group was actually half normal and half SA. This demonstrates that if a patient is considered normal by the MWST but abnormal by the cough test, it is possible that the patient is SA up to 50% of the time. The results of the combined screening system are demonstrated in Figure 1.
There are many bedside screening tests available to detect dysphagia. However, SA cannot be screened with those tests. Therefore, we studied the usefulness of a cough test in screening for SA that resulted in a high sensitivity and specificity. Moreover, we combined MWST and the cough test to make a combined screening system that can assess the airway protective reflex in addition to swallowing reflex. This system performs well in classifying the condition of patients who are suspected to have dysphagia.
This research was supported in part by a Research Grant for Sciences (16390550) from the Ministry of Education, Japan.
- 9.Lana P, Marc SL, Yu-Xiao Y, Stephen ER, David AK, Igor L, Warren BG. Videofluoroscopic studies of swallowing dysfunction and the relative risk of pneumonia. AJR Am J Roentgenol 2003;180:1613–1616Google Scholar
- 12.Sekizawa K, Ujiie Y, Sasaki H. Measurement of cough threshold. Respir Circul 1992;40:157–159, [in Japanese]Google Scholar
- 14.Arai H, Sato T, Nakajoh K, The standard in decision of appropriateness of ingestion by cough test. A study report: Nutrition intake of aged people with dysphagia. The study project of health and welfare of aged people. 1997;60–64 [in Japanese]Google Scholar
- 15.Tanaka T, Baba T, Saito E, Okui Y, Tohara H, Mizuno M. An investigation of screening test of silent aspiration by cough test-the decision of the concentration of citric acid. Jpn J Dysphagia Rehabil 2000;4:163, [in Japanese]Google Scholar
- 16.Iwata S. Current and attachments of aerosols in the use for allergic laryngitis. Otorhinolaryngology 1995;38:169–178, [in Japanese]Google Scholar
- 18.Logemann JA. Evalutation, treatment of swallowing disorders 2nd ed. Austin TX: Pro-Ed, 1998Google Scholar
- 19.Langmore SE. Endoscopic evaluation and treatment of swallowing disorders. New York: Thieme Medical Publishers, 2001Google Scholar
- 20.Kataoka M, Omori R. Jet nebulizer. Jpn J Child Nurs 2003;26:654–656, [in Japanese]Google Scholar
- 22.Maruoka M. Ultrasonic nebulizer. Jpn J Child Nurs 2003;26:657–658 [in Japanese]Google Scholar
- 23.Sinbo T. Ultrasonic nebulizer and jet nebulizer Q&A. Jpn J Nurs 2004;24:78–84 [in Japanese]Google Scholar
- 25.Ogoshi T, Iwata A. Distribution of aerosol-particle in human respiratory tract. J Med Soc Toho Univ 1987;34:207–214, [in Japanese]Google Scholar
- 35.Zenner PM, Losinski DS, Mills RH. Using cervical auscultation in the clinical dysphagia examination in long term care. Dysphagia 1998;2:127–135Google Scholar
- 37.Oguchi K, Saitoh E, Mizuno M, Baba M, Okui M, Suzuki M. The repetitive saliva swallowing test (RSST) as a screening test of functional dysphagia (1) normal values of RSST. Jpn J Rehabil Med 2000;37:383–388, [in Japanese]Google Scholar
- 40.Stanners AJ. Clinical predictors of aspiration soon after stroke. Age Aging 1993;2(Suppl 2):A47Google Scholar
- 47.Addington WR, Stephens RE, Ockey RR, Kann D, Rodriguez M. A new aspiration screening test to assess the need for a modified barium swallow study. Arch Phys Med Rehabil 1995;76:1040, [abstract]Google Scholar
- 48.Hammand CA, Goldstein LB, Zajac DJ, MD LG, Davenport PW, Bolser DC. Assessment of aspiration risk in stroke patients with quantification of voluntary cough. Neurology 2001;56:502–506Google Scholar