key: cord-0909116-dkq53flg
authors: Webler, Kathleen; Carpenter, Julia; Hamilton, Valerie; Rafferty, Miriam; Cherney, Leora R.
title: Dysphagia Characteristics of Patients Post SARS-Co-V2 during Inpatient Rehabilitation()
date: 2021-10-29
journal: Arch Phys Med Rehabil
DOI: 10.1016/j.apmr.2021.10.007
sha: 5f793aa21220ca59117e174846080cd7ddc43ffb
doc_id: 909116
cord_uid: dkq53flg

OBJECTIVE: To investigate dysphagia in patients recovering from SARS-CoV-2 admitted to acute inpatient rehabilitation by summarizing clinical swallow evaluation and videofluoroscopic swallow study findings. DESIGN: Retrospective cohort study SETTING: Urban inpatient rehabilitation hospital PARTICIPANTS: First forty inpatient admissions with SARS-CoV-2 who participated in a videofluoroscopic swallow study INTERVENTIONS: None MAIN OUTCOME MEASURES: Patient characteristics upon admission (duration of intubation, tracheostomy status, comorbidities, VFSS completion at previous level of care); admission International Dysphagia Diet level (IDDSI); Mann Assessment of Swallowing Ability (MASA), Functional Oral Intake Scale (FOIS), dysphagia severity rating; Penetration Aspiration Scale (PAS) rated during VFSS; and IDDSI level recommended following completion of VFSS. RESULTS: 20% of patients had been evaluated by videofluoroscopy in acute care. 19/37 (51.%) individuals were upgraded to IDDSI level 7 regular diet with level 0 thin liquids and achieved a FOIS of 7 after the completion of the VFSS. Five individuals (13%) received a diet downgrade or remained on the same diet recommendations from their admission. Total numerical score (TNS) of less than 170 on the Mann Assessment of Swallow Ability (MASA) predicted presence of aspiration in 54% (7/13) of patients. 56% (15/27) of the sample had a TNS less than 170 but did not demonstrate any instances of aspiration. The odds of patients having a PAS greater than or equal to 3 increased by approximately 15% (OR=1.15, 95% CI: 1.03, 1.27, p=0.013). Thus, with each additional day of intubation during acute care stay, there was an 15% greater likelihood of having airway invasion. CONCLUSIONS: Instrumental swallow evaluations are imperative to diagnose and treat dysphagia in the post COVID population. Because of the heterogeneity of this population, high incidence of prolonged intubation, and limitations of the clinical swallowing evaluation, instrumental assessments need to be performed on a more consistent basis as infection prevention protocols evolve.

Patients recovering from SARS-CoV-2 reportedly present with dysphagia or difficulty swallowing. [1] With the onset of the COVID-19 pandemic, the nature of dysphagia care shifted significantly, with multiple guidelines published to address the safe evaluation and treatment of patients with dysphagia in the acute care [2] [3] [4] [5] . However, little has been published providing guidance for the evaluation and treatment of this patient population after they leave acute care and enter inpatient rehabilitation.

The pandemic has prompted reflection on infection prevention protocols as well as a designation by some groups that dysphagia assessment and treatment produce aerosolized droplets, resulting in recommendations to use non-invasive and non-instrumental means to evaluate dysphagia [2, 3, 5] . However, instrumental examination via videofluoroscopy (VFSS) or via fiberoptic endoscopic evaluation of swallowing (FEES) is considered best practice in evaluating swallowing physiology, determining the presence or absence of aspiration, directing behavioral interventions to improve function, and guiding diet consistency recommendations [6] . With more restricted use of instrumental assessments, a higher reliance was placed on the clinical swallow examination, with caveats limiting completion of a full oral mechanism examination, assessment of cough strength, and gag reflex [7] .

To date, specific information related to the etiology and nature of dysphagia has not yet been published for the inpatient rehabilitation setting following acute care hospitalizations for COVID-19 [3] .

Proposed mechanisms of dysphagia post COVID-19 include peripheral and central nervous system disruption, intubation, debility, and pulmonary dysfunction [1] Confounding the impact of SARS-Co-V2 on swallowing function was an acute care management trend toward longer periods of endotracheal intubation [8] . The Centers for Disease Control identify tracheostomy placement as a procedure that produces aerosolized droplets with greater risk for infection spread, and many infection prevention protocols initially discouraged tracheostomy placement with this population [8] [9] [10] As a result, increased durations of intubation were observed [11] For example, during the first month of the pandemic, Hur and colleagues report that 64% of patients in acute care were intubated for greater than 14 days [12] .

Prior to the SAR-CoV-2 outbreak, a relationship between prolonged intubation and dysphagia had been found in critically-ill patients requiring mechanical ventilation [13] as well as those with acute respiratory distress syndrome [1] . Dysphagia was present in greater than 50% of patients who were intubated longer than 48 hours [14] . Moderate to severe dysphagia was seen in patients who were intubated for longer than 7 days [15] . Prolonged intubation may lead to reduced laryngeal sensation which can result in silent aspiration [16] .

Though previous literature establishes a relationship between intubation and dysphagia, it is not clear how the swallowing mechanisms and physiology are impacted by SARS-CoV-2. Given the limited information to date on dysphagia during inpatient rehabilitation in patients post COVID-19, our aims were to 1) describe clinical and videofluoroscopic swallow study characteristics at admission to inpatient rehabilitation and 2) explore the relationships between clinical swallow evaluation results, videofluoroscopic swallow study results and prior intubation in patients admitted to an inpatient rehabilitation hospital following acute hospitalization for SARS-CoV-2 .

This retrospective study was conducted at a large urban acute rehabilitation hospital. Our Institutional Review Board approved extraction of patient data from the electronic health record (EHR) and granted a waiver of informed consent for use of de-identified, historical data. Objective measures were used to classify patient performance to avoid recall bias. We reviewed the electronic medical records of the first 40 patients with prior SARS-CoV-2 diagnoses who were referred for a VFSS between the months of April-August 2020. Referrals for VFSS were determined by the evaluating speech-language pathologist (SLP) after completion of a comprehensive speech, language and swallowing evaluation that included a clinical swallow examination with administration of the Mann Assessment of Swallowing Ability [17] and the Functional Oral Intake Scale [18] . Demographic characteristics were extracted from the electronic medical record, including current medical co-morbidities. Co-morbidities were extracted based on admitting physicians ICD-10 codes [19] . The first 40 patients who were referred for a VFSS were considered to be a representative sample of the 121 patients admitted to inpatient rehabilitation with 2019-nCoV acute respiratory disease, pneumonia due to SARS-associated coronavirus and/or pneumonia due to SARS-associated coronavirus diagnosis. An independent reviewer completed a reliability check on data extracted from 20% of the patients included in the sample determining reliability to be 97% with the electronic health record.

The Mann Assessment of Swallowing Ability (MASA) [17] was administered at admission and discharge as a part of the standardized clinical swallow examination [20] . The MASA standardizes administration and scoring components of the clinical swallowing evaluation to determine the presence of dysphagia and severity of impairments. The MASA was initially validated in patients after acute stroke [17] . It provides a score that ranges from 0-200 with a stroke-based risk cutoff score of ≥170 indicating no abnormality [21] . Further, likelihood ratios were used to create four categories to define the risk for aspiration as follows: unlikely (≥170), possible (149-169), probable (141-148), and definite (≤140) [22] . While not validated with the post COVID population, research has shown application of the MASA with populations other than stroke [23, 24] and thus it is administered as a part of the hospital's usual care in order to standardize the clinical swallow evaluation.

The Functional Oral Intake Scale (FOIS) [18] was rated at admission and discharge by the treating SLP. The FOIS is a seven-point scale that documents the impact of dysphagia on oral intake of food and liquid, and includes both feeding tube dependence and consistency modifications. The FOIS has adequate reliability, validity and sensitivity to change in functional oral intake over the course of rehabilitation [18] .

The International Dysphagia Diet Standardization Initiative (IDDSI) framework consists of eight food and liquid consistencies on a scale from 0-7 [25] . Each patient's admission and discharge IDDSI diet as well as recommended IDDSI diet following VFSS were extracted from the electronic health record.

A videofluoroscopic swallow study (VFSS) was completed on each of the patients. If a patient had more than one VFSS for reassessment purposes during the study period, only the first was included in this study. A VFSS evaluates the "functional anatomy and physiology of the swallowing mechanism, swallowing efficiency, and airway protection." [26] . The speech-language pathologist who completed the VFSS determined the presence or absence of aspiration using the Penetration Aspiration Scale (PAS) [27] , identified the need for compensatory strategies, and provided diet recommendations. The PAS is an 8-point scale that rates the degree of airway invasion. Scores of 3 or greater indicate airway invasion, whereas a score of 6 or greater indicates aspiration. For this study, we extracted the highest (worst) PAS score, the IDDSI food or liquids bolus consistency on which it occurred, and the size of the bolus. This extracted score reflects the nature of the presenting dysphagia and is a common method for summarizing PAS scores [28] .

Descriptive statistics were used to characterize the patient demographic information and comorbidities, the MASA and FOIS scores, and the results of the VFSS. Correlation coefficients were used to determine the relationship between the MASA total numeric score and the PAS determined from the VFSS. We used a logistic regression with one single continuous predictor (number of days of intubation) to estimate the odds of patients having a penetration aspiration score greater than or equal to 3 (indicating airway invasion), as compared to having a penetration aspiration score of less than 3. The presence of stroke and the presence of having a tracheostomy were assessed as independent predictors for PAS because strokes and tracheostomies may be associated with PAS impairments to a greater extent than intubation alone. Analyses were performed using SAS version 9.4 [1] .

A total of 40 patients with confirmed diagnosis of SARS-COV-2 at the time of inpatient rehabilitation were included in data collection (age 32 -86; Mage = 65.9 ± 13; 29 identified as males and 11 identified as females). Most (43%) individuals had at least one acute comorbidity, the most common of which were hematologic and neurologic (Table 1) 

At the time of initial evaluation during inpatient rehabilitation, 37/40 (93%) of patients had a FOIS score of 6 or less indicating restriction of oral intake ( Table 2) . Of the 40 patients, three (8%) were achieving nutrition by mouth, with no modifications; 62% of patients achieved all nutrition by mouth, though with modification to consistencies; 10% achieved some oral intake but required supplementation by gastrostomy tube; and 20% could consume nothing by mouth and were dependent on a gastrostomy tube for nutrition and hydration (Table 3) 

We compared the MASA score from the clinical swallowing evaluation with results obtained during the VFSS. We calculated the Total Numerical Score (TNS) and the Likelihood of Aspiration Ratio (LAR) achieved on the MASA. A TNS of less than 170 predicted the presence of aspiration with 50% (6/12) accuracy 56% (15/27) of the patients had a TNS less than 170 but did not demonstrate any instances of aspiration. 50% of patients who aspirated (6/12) achieved a likelihood of aspiration ratio (LAR) of "unlikely to aspirate" on the MASA. 7% (2/27) achieved a "definite" LAR but did not aspirate.

Results of the logistic regression indicated that for each additional day of intubation, the odds of patients having a PAS greater than or equal to 3 increased by approximately 15% (OR=1.15, 95% CI: 1.03, 1.27, p=0.013). Thus, with each additional day of intubation, there was a 15% greater likelihood of having airway invasion. Stroke and tracheostomy placement were not significant predictors of PAS alone or in combination with days intubated.

Our study shows that instrumental swallow evaluation provides important information for the diagnosis and treatment of dysphagia in patients who have had SARS-CoV-2 and are admitted to inpatient rehabilitation. The retrospective analysis also demonstrates a high correlation between acute care intubation and ongoing dysphagia during inpatient rehabilitation for patients with SARS-CoV-2 diagnoses. Specifically, the likelihood of airway invasion, as identified on videofluoroscopy, increases with each day of intubation and persists into the subacute rehabilitation period of recovery. This finding is consistent with prior evidence that prolonged intubation decreased laryngeal sensation and puts patients at risk for silent aspiration or airway invasion [15, 16] .

In addition to increased duration of intubation, other underlying factors likely impacted dysphagia. Many of the patients in this sample experienced multiple acute co-morbidities. Critical illness contributes to deconditioning and debility, impaired cognition, neuromyopathy, and decreased coordination of swallow and breathing [13, 15, 16] . The presence of airway invasion and silent aspiration in our sample may indicate a common pathophysiology of dysphagia in patients following SARS-CoV-2 intubation. The incidence of aspiration after an acute stroke varies, but can be as high as 33% [29] . A high number of patients in this study experienced a neurological co-morbidity, yet our analysis found that the presence of stroke was a non-contributing factor for airway invasion in this population. Further previous literature supports a high incidence rate of aspiration in patients with tracheostomies, citing up to 30-50% [30, 31]. Again, our analysis found that the presence of a tracheostomy alone was a non-contributing factor for airway invasion in this population.

Silent aspiration occurred on 27% of patient sampled. An instrumental swallow study would be imperative for identification and intervention. VFSS was important for identifying strategies that were appropriate for helping patients return to intake of unmodified liquids. Though 11 patients were able to upgrade or remain on level 0 thin liquids, they required modifications for safe liquid intake including taking single cup sips, restriction from using straws, and drinking a small volume, which could only have been determined with the completion of the VFSS.

Further, our analysis underscores the need for conducting instrumental swallowing evaluations in this patient population. In our sample, MASA TNS and LAR were poor predictors of aspiration. 56% of patients achieved a TNS of less than 170 but did not aspirate. 50% of patients who aspirated achieved a LAR of "unlikely" on the MASA. Only 20% of patients in our sample had a VFSS conducted during their acute care hospitalization, yet 93% were receiving modified diets as rated on the FOIS, indicating conservative management in acute care. This management trend may reflect early decision-making during the pandemic to hold or defer VFSS and FEEs given infection prevention concerns with aerosol generating procedures.

Collaboration with the hospital infection preventionist allowed early implementation of measures to allow completion of VFSS with the SARs-CoV-2 population at our rehabilitation hospital.

Our guidelines limited exposure during patient transport, reduced the number of personnel during the examination, and placed a time restriction on room use after the completion of the aerosol generating procedure. Later in the pandemic, guidelines such as these were suggested for the medical community

[32]. Our study supports recent efforts of many hospital systems to revisit their protocols for instrumental dysphagia evaluations with COVID positive patients [2] .

There are several limitations related to the retrospective nature of this study. Compared to prospective dysphagia research, we report fewer objective measures of swallowing physiology, because the VFSS were documented using standard clinical rather than research practices. The timing of the completion of the VFSS was not extracted as part of this data analysis. One limitation of using FOIS is that it can be influenced by non-dysphagia related factors such as poor dentition or consistency preference, which we did not track due to their low frequency of occurrence in this sample. Our patient sample was taken early in the pandemic and may not represent current patients discharged from acute care given evolving practices in acute care management.

Instrumental swallow evaluations are imperative to diagnose and treat dysphagia in the post COVID population. Because of the heterogeneity of this population, the potential for multiple medical co-morbidities, and the lack of sensitive screening tools, instrumental assessments need to be performed on a more consistent basis. Infection prevention protocols need to evolve to prioritize the completion of instrumental swallow studies such as videofluoroscopy for this patient population. This population, with a high incidence of prolonged intubation, is at greater risk for aspiration, and clinical swallow evaluations are limited in their sensitivity in identifying aspiration. Further research is warranted to specify the pathophysiology of dysphagia in this population through more precise measurement. 

Postintubation Dysphagia During COVID-19 Outbreak-Contemporary Review

Aerosol generating procedures, dysphagia assessment and COVID-19: A rapid review

The Long-Term Effects of COVID-19 on Dysphagia Evaluation and Treatment. Archives of physical medicine and rehabilitation

Guidelines of clinical practice for the management of swallowing disorders and recent dysphonia in the context of the COVID-19 pandemic. European annals of otorhinolaryngology, head and neck diseases

Dysphagia Care Across the Continuum: A Multidisciplinary Dysphagia Research Society Taskforce Report of Service-Delivery During the COVID-19 Global Pandemic

Best Practices in Modified Barium Swallow Studies

Speech-Language Pathology Management for Adults With COVID-19 in the Acute Hospital Setting: Initial Recommendations to Guide Clinical Practice

Long-term intubation and high rate of tracheostomy in COVID-19 patients might determine an unprecedented increase of airway stenoses: a call to action from the European Laryngological Society. European archives of oto-rhino-laryngology : official journal of the European Federation of Oto-Rhino-Laryngological Societies (EUFOS) : affiliated with the German Society for Oto-Rhino-Laryngology -Head and Neck Surgery

European archives of oto-rhinolaryngology : official journal of the European Federation of Oto-Rhino-Laryngological Societies (EUFOS) : affiliated with the German Society for Oto-Rhino-Laryngology -Head and Neck Surgery

Tracheostomy in the COVID-19 era: global and multidisciplinary guidance

Early Outcomes From Early Tracheostomy for Patients With COVID-19

Factors Associated With Intubation and Prolonged Intubation in Hospitalized Patients With COVID-19. Otolaryngol Head Neck Surg

Post-extubation dysphagia is associated with longer hospitalization in survivors of critical illness with neurologic impairment

The incidence of dysphagia following endotracheal intubation: a systematic review

Recovery from Dysphagia Symptoms after Oral Endotracheal Intubation in Acute Respiratory Distress Syndrome Survivors. A 5-Year Longitudinal Study

Relationship Between Laryngeal Sensation, Length of Intubation, and Aspiration in Patients with Acute Respiratory Failure

MASA : the Mann assessment of swallowing ability

Initial psychometric assessment of a functional oral intake scale for dysphagia in stroke patients

ICD-10 : international statistical classification of diseases and related health problems : tenth revision

Assessment Across the Life Span: The Clinical Swallow Evaluation

Analysis of a Physician Tool for Evaluating Dysphagia on an Inpatient Stroke Unit: The Modified Mann Assessment of Swallowing Ability

Assessment of Aspiration Risk Using the Mann Assessment of Swallowing Ability in Brain-Injured Patients With Cognitive Impairment

Evaluation of a dysphagia screening system based on the Mann Assessment of Swallowing Ability for use in dependent older adults

Efficacy of Predicting Videofluoroscopic Results in Dysphagic Patients with Severe Cerebral Palsy Using the Mann Assessment of Swallowing Ability. American journal of physical medicine & rehabilitation

The Need for International Terminology and Definitions for Texture-Modified Foods and Thickened Liquids Used in Dysphagia Management: Foundations of a Global Initiative

Breathing and swallowing dynamics across the adult lifespan

A penetration-aspiration scale

Use of the Penetration-Aspiration Scale in Dysphagia Research: A Systematic Review

Level 4: Total oral diet of a single consistency. Level 5: Total oral diet with multiple consistencies, but requiring special preparation or compensations. Level 6: Total oral diet with multiple consistencies without special preparation, but with specific food limitations. Level 7: Total oral diet with no restrictions IDDSI: Level 7 regular diet, level 6 -soft and bite size, level 5 -moist and minced, level 4 -puree, level 4 liquids -extremely thick liquids, level 3 -moderately thick liquids, level 2 -mildly thick liquids, level 0thin liquids PAS: 1 -no penetration or aspiration, 2 -penetration, contrast remains above the vocal folds and subsequently ejected, 3-penetration, contrast remains above the vocal fold and not ejected, 4penetration, contrast contacts vocal folds and subsequently ejected, 5 -penetration, contrast contacts vocal folds and not ejected, 6 -aspiration, contrast below vocal folds and subsequently ejected, 7aspiration not ejected despite effort, 8 -aspiration, no effort made to eject