Patients who require long-term mechanical ventilation with tracheostomy are at risk for swallowing disorders for various reasons. Individuals on mechanical ventilation frequently exhibit difficulty coordinating breathing and swallowing, decreased laryngeal elevation, decreased pharyngeal and laryngeal sensation, and decreased subglottal pressure [1,2,3,4,5,6,7,8]. Additionally, individuals who are dependent upon a ventilator are likely to aspirate [2,4,5,9,10,11,12,13], and the aspiration is usually silent [14]. Both oral and pharyngeal stage swallowing abnormalities have been documented in this group, including oral transit and bolus formation deficits, delayed swallow initiation, residue throughout the pharynx and pyriform sinuses [4,10,13], and decreased duration of vocal fold closure during swallowing [8]. Swallowing disorders create serious health risks and can complicate respiratory disease and nutritional status.

Speech–language pathologists (SLPs) use a combination of evaluation techniques to determine the presence and characteristics of dysphagia. The clinical or bedside swallowing evaluation (CBSE) provides important information about oral motor abilities and cognitive skills and also allows the SLP to observe visible and audible signs of aspiration [15,16]. However, the clinical evaluation without the support of an instrumental measure like a Modified Barium Swallow Study (MBSS) will not provide an accurate assessment of pharyngeal skills and may underestimate the severity of dysphagia and aspiration which is often asymptomatic [13,14,16,17,18,19,20,21]. The CBSE is especially unreliable for identifying aspiration in patients requiring mechanical ventilation or tracheostomy [7,21].

Instrumentation like the MBSS is also necessary to objectively determine strategies or maneuvers to increase swallowing safety and efficiency and to make dietary recommendations [4,10,20,22,23,24]. Dietary recommendations based on only a clinical evaluation often differ from recommendations after a MBSS and may either be overly restrictive or place the individual at risk for aspiration and choking [21,23,24]. Definitive diagnosis and appropriate swallowing guidelines are especially critical for patients with compromised respiratory status since they are more likely to have complications if they aspirate [5,14].

Although researchers are attempting to create more accurate and reliable clinical evaluations for dysphagia assessment, existing measures serve only as a screening to determine those patients who require instrumental assessment. Several studies have identified clinical indicators of dysphagia, and these in combination have been used to indicate a need for further diagnostic testing [17,22,25]. The use of multiple clinical features can better identify individuals who need further instrumental study, but this practice is not reliable enough since underestimating dysphagia can have serious consequences for the medical status of the patient. Specifically, patients with ventilator dependency who are being considered to begin oral feeding should routinely be referred for the MBSS because of the frequent occurrence of silent aspiration in this group.

The use of mechanical ventilation in elderly patients is increasing [26,27], and age may play a factor in mortality and morbidity for those with respiratory conditions that require mechanical ventilation [26,28,29,30,31,32,33]. Mortality rates after respiratory failure are higher with increasing age although other factors such as premorbid health status also impact prognosis. Aging creates physiologic changes in pulmonary function such as gas exchange and elastic recoil, respiratory muscle strength, and the functional reserve of the other systems that support respiration and swallowing [34,35,36,37].

The purpose of this study was to determine the swallowing characteristics and incidence of dysphagia in elderly patients on mechanical ventilation admitted to a long-term, acute-care hospital. The co-occurrence of aspiration with swallowing disorders has been documented in patients with respiratory complications, but specific characteristics of different patient groups have not been explored extensively. Most of the patients in this study used a Passy–Muir speaking valve (PMV) as part of respiratory and swallowing rehabilitation, and although evidence is conflicting, studies have indicated that the PMV improves swallowing function and decreases aspiration [3,12,38,39]. However, since this was a retrospective chart review, it was not possible to determine the direct impact of the PMV with this group because of the nature in which the original dysphagia evaluations were conducted. Specific areas of interest in the study included any differences in recommendations from the clinical swallowing evaluation versus the MBSS, characteristics and incidence of dysphagia, and the effect of demographic variables.

Methods

Sample

Individuals in the study were adults admitted to this violates confidentiality, a 60-bed, long-term, acute-care hospital with an intensive care unit in a midwestern US city. The population of this hospital was predominantly elderly and medically complex individuals with multiple diagnoses, and many required long-term ventilator support. A major focus of the hospital was ventilator weaning and physical rehabilitation in patients transferred from area hospitals.

Procedure

Data were obtained from a retrospective review of patient medical records from 1994 when the hospital opened until April 2002. Each medical chart in the facility was reviewed by the researchers. All patients on mechanical ventilation with tracheostomy who were being considered as candidates to begin oral feeding and had undergone a MBSS were included in the study. Patients who were orally intubated, who had only a tracheostomy without mechanical ventilation, or who were comatose were excluded. The average length of stay in the facility across diagnoses was 28 days. The underlying causes for the respiratory conditions varied, and the respiratory diagnoses of the subjects included respiratory failure, ventilator dependency, and airway obstruction. A protocol was designed and used by the researchers to assure that uniform information was obtained (see Appendix). Information collected from each medical record included ventilator status, dysphagia symptoms from both the CBSE and the MBSS, recommendations from the CBSE and the MBSS, demographics, and method of nutritional intake.

Protocol for the Clinical Bedside Swallowing Evaluation

Individuals were administered two presentations of 1/2 teaspoon of thin liquids, puree, and soft solid food consistencies (i.e., graham cracker) in that order unless the prior consistency had resulted in obvious signs of aspiration or choking. Patients who wore a PMV (91.4%) had the valve on and the cuff deflated for trial swallows. Patients who could not tolerate cuff deflation or the PMV received trial swallows with the cuff inflated. The use of food coloring or suctioning after swallows was not routinely utilized since this was regarded only as a gross screening measure for aspiration. The average length of time between the CBSE and MBSS was 9.9 days. Any delay in the MBSS after the CBSE was due to individuals being inappropriate candidates for the MBSS at the time of the CBSE because of a low level of consciousness, poor response to the trial swallows, or a medical complication prior to the study being conducted.

Definition of Terms

Aspiration was defined as material passing below the level of the true vocal folds. Silent aspiration was defined as aspiration that occurred without an audible reaction or cough from the individual. Penetration was defined as the presence of material in the laryngeal vestibule or airway entrance that may or may not have been cleared but that did not pass below the level of the true vocal folds during the observed period. A uniform CBSE protocol and MBSS protocol was utilized by the SLPs who performed the original assessments. The policy for SLPs in this facility was to recommend the MBSS for ventilator-dependent patients as soon as individuals could tolerate the procedure if there was any suspected pharyngeal component from the clinical examination. Therefore, the majority of patients on mechanical ventilation underwent a MBSS before beginning oral intake.

Results

Data Analysis

Statistical comparisons were made with the measures of chi squared, t-test, and one-way ANOVA using SPSS version 11.0 statistical software (SPSS Inc., Chicago, IL). Descriptive statistics were also calculated for means and frequency counts.

Demographics

There were 58 patients who met the requirements for inclusion in the study over the specified time period. A description of each subject and diet recommendations are provided in Table 1. Summary information about age, gender, PMV use, and NPO status is listed in Table 2. The primary respiratory condition which created the need for mechanical ventilation was respiratory failure for 86% of subjects (50/58) and ventilator dependence for 14% (8/58). History information about the underlying disease creating the respiratory condition was not available. Individuals who had been previously prescribed a PMV by their physicians used the valve during the dysphagia evaluations. Gender distribution was roughly commensurate at 58.6% female and 41.4% male. The mean age of the group was 69.7 years of age with a range of 36–86 years. Discharge dispositions at the end of the hospital stays were home (19%), expired (36.2%), nursing home (37.9%), and hospital transfer (6.9%). This information gives an indication of the severity of illness in these individuals.

Table 1 Descriptions of subjects, diagnoses, and diet recommendations
Table 2 Summary demographics of subjects and frequency of NPO

CBSE and MBSS Results

Prior to the MBSS, 84.5% of the participants were NPO or prohibited from any oral intake. A paired-samples t-test revealed that the differences between diet recommendations from the CBSE and the MBSS were statistically significant for both food (t = −7.021, df = 57, p = 0.001) and liquid (t = −7.592, df = 57, p = 0.001). After the MBSS, only 24.1% of individuals remained NPO, and 69.6% were allowed thin or nectar-consistency liquids indicating that the results of the MBSS allowed for a less restrictive diet than was recommended from the CBSE (see Table 2). However, this result may have been impacted by the fact that the SLPs at this facility regularly used MBSS with patients on mechanical ventilation prior to beginning an oral diet due to the high incidence of clinically undetected pharyngeal dysfunction. SLPs in other hospitals may be more likely to recommend an oral diet from only the CBSE without instrumental assessment. The most common dietary consistency initiated after the MBSS was puree (34.5%) followed closely by mechanical soft (32.8%). The most common liquid recommendations for patients after the MBSS were thin (44.6%), nectar (25%), honey (7.1%), and no liquids (23.3%). Compensatory strategies that alleviated aspiration were identified on the MBSS for 63.8% of patients.

Dysphagia Symptoms

Patients exhibited multiple symptoms of both oral and pharyngeal stage dysphagia during the MBSS, including impaired anterior–posterior bolus transport, delayed swallow, laryngeal penetration, and vallecular and pyriform sinus residue (see Table 3). Aspiration was present in 41.4% of patients and occurred with equal frequency before and during the swallow (12.1%) and at times during both the oral and pharyngeal stages of the swallow (10.3%). Aspiration was silent 83.3% of the time when it occurred.

Table 3 Pharyngeal abnormalities observed on MBBS

Chi-squared analysis was used to determine associations between the presence of aspiration and gender, laryngeal penetration, cognitive deficits, vallecular residue, pyriform residue, delayed pharyngeal swallow, and decreased anterior–posterior bolus transport. The presence of residual material in the pyriform sinuses (p = 0.039), laryngeal penetration (p = 0.001), and decreased anterior–posterior bolus transport (p = 0.027) were significantly associated with aspiration. A one-way ANOVA demonstrated no significant association between age and aspiration although it was expected that aspiration would be associated with increasing age.

Discussion

Many patients who are ventilator-dependent are elderly and have swallowing disorders that result in aspiration. In this study, as in other studies, patients on mechanical ventilation aspirated frequently. With mechanical ventilation and tracheostomy, adverse changes in the swallowing mechanism occur as a result of disuse muscle atrophy, desensitization, and chronic laryngeal damage, and these affects accumulate over time. Reinstituting oral intake as soon as it is safe may prevent further loss of feeding and swallowing skills. Since aspiration in patients with ventilator dependence is typically silent, the MBSS is necessary when these individuals are being considered as candidates to begin oral feeding.

Diet recommendations made from the CBSE in this study were significantly different from those made after the MBSS was conducted and, almost always, the individual was allowed to begin oral intake or upgrade the diet. These data demonstrate that although elderly patients on mechanical ventilation are at high risk for dysphagia, they can often safely have some oral intake and should not be assumed to be profound aspirators because of their respiratory status. It is important to provide oral intake as soon as patient safety is ascertained with the MBSS in order to avoid the negative affects of being NPO, including discomfort, muscle atrophy, sensory deprivation, tube feeding risks, and nutritional deficiencies. A return to oral intake is associated with increased quality of life, social interaction, independence, and as an important step toward recovery [40,41,42,43,44,45,46,47].

It should be noted that 91.4% of individuals utilized the PMV while eating and swallowing, although it was not possible to determine its impact on swallowing. It is interesting that the individuals in this study on mechanical ventilation aspirated with equal frequency before and during the swallow and also tended to aspirate during the oral and pharyngeal stages of the swallow. Oral stage dysfunction may occur from a combination of factors including generalized weakness and frailty, dry mouth from the ventilator from being NPO, decreased oral motor skills from disuse, and oral motor deficits from underlying neurological conditions. The multiple instances of aspiration that occurred may have been due to difficulty coordinating swallowing with the cycling of the ventilator.

Additionally, patients on mechanical ventilation frequently require anti-anxiety medications and these may cause sedation and decrease cognitive awareness needed for safe swallowing. Laryngeal penetration and pyriform sinus residue may occur from reduced sensitivity to material in the pharynx, poor laryngeal elevation from the tracheostomy tube tethering, and a weak swallow from decreased subglottal pressure. Once food or liquid reached the level of the airway entrance, these patients were unable to clear the material.

Conclusion

This research provides information about the characteristics and prevalence of dysphagia in elderly patients on mechanical ventilation and, as with other patient groups, demonstrates the importance of the MBSS. These results provide additional justification for the routine use of the MBSS to improve patient safety and to facilitate a timely return to oral feeding. The clinical examination is valuable but is not sufficient for determining the safety of oral feeding or appropriate dietary recommendations. Elderly adults with ventilator dependence vary in their strengths and weaknesses, and many can tolerate some level and amount of oral intake. These individuals may have multiple symptoms of dysphagia that are not visible or audible, and SLPs and medical personnel should not make assumptions about the swallowing status of individuals who are ventilator-dependent.

Instrumental assessment like the MBSS is critical for making any type of dietary recommendation and to assure an adequate diagnosis of oropharyngeal and laryngeal function. The MBSS can be performed with patients on mechanical ventilation using a portable ventilator in the radiology suite or a portable fluoroscopy unit at the bedside. An individual’s quality of life may be greatly improved by determining if there are any consistencies that can be eaten safely as opposed to unnecessarily restricting oral intake.

Eating is a positive functional outcome of dysphagia intervention, may prevent further loss of skills and muscle atrophy, and is less expensive than tube feeding. As technology extends lifespan, the number of patients with ventilator dependence will increase. SLPs will be frequently required to evaluate and make recommendations regarding swallowing safety for these patients and must be aware of the complicated issues involved. Long-term acute care with mechanical ventilation and nonoral feeding is very costly with uncertain outcomes and no reliable method of prediction for mortality rates and quality of life. Healthcare providers are seeking the tools to best assess and intervene with these chronically ill patients to balance safety, quality of life, and costs of care. The ability to swallow safely is a very important part of overall health and well-being, and appropriate dysphagia intervention can increase both psychosocial and physical outcomes.

Figure 1
figure 1

.