A Pilot Study of a Modified Swallowing Screening Tool for Critically Ill Patients in the Intensive Care Unit

Abstract

The lack of early assessment tools for swallowing function in patients in the intensive care unit (ICU) may lead to delays in oral intake. This study assessed the effectiveness of a new bedside swallowing screening tool in detecting dysphagia in patients in the ICU or isolation settings, where isolation settings refer to conditions such as COVID-19, where patient mobility is limited. We assessed swallowing function in 13 patients with severe acute respiratory distress syndrome. To ensure patient safety, a bedside preliminary investigation was performed to assess the patient’s alertness level, and ultrasound findings of the vocal cords were obtained. Patients were nasogastric(NG) tube feeding or fasted if they did not meet a certain readiness level. Additionally, patients who passed the preliminary investigation underwent methylene blue dye or citric acid swallowing tests if they had or had not undergone tracheostomy, respectively. The Gugging Swallowing Screen (GUSS) test was used to determine whether an oral diet was appropriate. Of the 13 patients, 1 failed to meet the criteria during the preliminary examination, and 12 underwent the examination and initiated oral intake. Significant differences were found in the GUSS scores between the tracheostomy and non-tracheostomy groups but not in the examination validity. NG tube was implemented if the examination criteria were not met at each stage. Patients who passed the newly developed dysphagia screening tool exhibited no aspiration symptoms post-oral feeding initiation, enabling their discharge. In conclusion, the proposed screening test can be performed safely and easily, allowing for early dysphagia detection, reduced aspiration risk, and safe oral feeding in patients in the ICU or isolation settings.

Introduction

Dysphagia poses a substantial challenge on both patients in the intensive care unit (ICU) and the healthcare system. This condition may be caused by various factors, such as severe sepsis, tracheostomy, advanced age, and ICU-acquired weakness [1]. Most importantly, it is usually identified after the removal of an endotracheal tube used for mechanical ventilation care in the ICU [2]. The worldwide prevalence of post-extubation dysphagia (PED) ranges from 3 to 62% (mean prevalence of 42%) in acute and sub-acute healthcare facilities [3]. However, this broad range may be attributed to heterogeneity in the study design, patient population, diagnostic methods, timing of assessment, and outcome definition [4].

Dysphagia significantly burdens the healthcare system, contributing to complications, including increased aspiration risk, pneumonia, delayed oral intake, malnutrition, and reduced health-related quality of life. Moreover, this leads to prolonged ICU/hospital stays and increased morbidity and mortality rates. Therefore, addressing dysphagia is crucial for effective healthcare management [1]. Furthermore, given the high risk of dysphagia in most patients in the ICU due to various factors, assessing this condition is essential in the ICU setting.

Although some hospitals use different types of swallowing screening tools, such as the Gugging Swallow Screen (GUSS), these assessments are usually limited to specific patient populations, such as patients with stroke [5]. The newly developed GUSS-ICU method has demonstrated relatively high sensitivity and specificity in PED evaluation. However, established guidelines for dysphagia evaluation in the ICU are lacking [6].

A videofluoroscopic swallowing study (VFSS) or a fiber-optic endoscopic evaluation of swallowing (FEES) is the gold standard for assessing swallowing in patients with dysphagia [5]. FEES has excellent accessibility for examination in an ICU setting since it can be installed immediately next to the patient’s bed. Additionally, it is more convenient than VFSS, which requires transporting the patient to the radiology department. However, the implementation of FEES in the ICU poses practical challenges. First, while acquiring the FEES equipment is feasible in major medical institutions, obtaining this equipment is financially challenging for smaller hospitals, making it practically difficult to be employed for routine screening. Second, the need for meticulous scope sterilization for each patient further limits the routine use of FEES in the ICU. Third, procuring the FEES equipment poses challenges compounded by the arduous process of achieving proficiency among FEES examiners, a specialized skill set not possessed by all healthcare providers, making it difficult to scale up its use without adequate staff training. Therefore, the demand for efficient screening tools remains high for various reasons, and the need for widely applicable screening tools has become increasingly evident.

During the COVID-19 pandemic (2019), we developed a modified dysphagia screening tool built on existing dysphagia evaluation methods for patients with severe COVID-19 admitted to the ICU when access to FEES testing was restricted to hospitals. The tool is relatively intuitive and simple and carries a low risk of generating aerosols compared to FEES, making it applicable in ICU settings. Therefore, this pilot study aimed to assess the effectiveness of this newly developed screening tool before formal validation (See Fig. 1).

Fig. 1.

A modified dysphagia screening protocol for critically ill patients. USS, Gugging Swallowing Screening; NG, Nasogastric; US, Ultrasonography

Methods

Study Population

This study was approved by the Institutional Review Board (approval number: 2307-015-129). A prospective study assessing 13 patients with polymerase chain reaction (PCR)-verified COVID-19 was conducted from January 2 to October 25, 2021, in the ICU of a tertiary referral hospital in Korea. After extubation for acute respiratory distress syndrome (ARDS), patients suspected of dysphagia were assessed for behavioral characteristics, including alertness, cooperation, and motivation, as well as signs of motor speech or voice abnormalities, focusing on patients who were assessed 24 h post-extubation. Due to the challenging circumstances posed by the COVID-19 pandemic, we employed a newly developed screening tool to evaluate patients for dysphagia because the patients’ confirmed infection impeded accessibility.

Preliminary Investigation

Through preliminary assessment, we verified the appropriateness of employing the newly developed screening tool on patients by evaluating their capacity to meet the minimum criteria for the use of this tool and their practical ability to swallow during the evaluation. We assessed the patient’s alertness for at least 15 min, ability to cough voluntarily, presence of hoarseness, ability to swallow saliva, and absence of drooling using this investigation. The preliminary test was originally an indirect item from the original GUSS examination; the five points obtained from this test were added to the five points from the GUSS examination performed at the end of the screening tool [7].

Vocal cord Ultrasound

Regarding the replacement of the vocal cord movement test usually conducted using FEES, a vocal cord ultrasound was performed to confirm abnormalities in vocal cord movement. Dr. Myung Hun Jang, who specializes in ultrasound procedures, obtained board certification in rehabilitation medicine five years ago and has since focused on ICU rehabilitation, conducting ultrasound evaluations of vocal cord function for the past five years in accordance with the scope of this research. The vocal cord ultrasound study utilized a combination of front-side transverse-axis ultrasound (FTU) and lateral-side longitudinal-axis ultrasound (LLU) to verify vocal cord abnormalities by identifying four sonographic landmark views as follows: the anterior commissure, true cords, false cords, and arytenoids [8]. The patient’s tracheostomy tube cuff was completely deflated prior to the commencement of the examination. A GE portable Ultrasound System (LOGIQ e. GE Wauwatosa, WI, USA) with a 10- or 12-MHz linear array transducer was used for the ultrasound imaging examination.

In step 1 (FTU), patients lay supine with a neck pillow, slightly extending the neck for full exposure. The ultrasound probe was placed on the skin perpendicular to the neck, scanning downward from the hyoid bone. After revealing the vocal cords, the probe was tilted using the thyroid cartilage to improve visibility. We had the patient produce vocalizations to assess the movement of the vocal cords. Finally, the probe was adjusted for better viewing if complications, such as calcifications or a prominent larynx, hindered clarity (See Fig. 2).

Fig. 2.

(A) Ultrasound image of normal vocal cords on an ultrasound vocal cord movement test. (B) Ultrasound image of hypomobile vocal cords on an ultrasound vocal cord movement test

For step 2 (LLU), the probe positioned perpendicular to the neck’s long axis examined both sides, including the thyroid, arytenoid, and cricoid cartilages. Vocal cord movement was assessed by detecting arytenoid cartilage movements. Similarly, we had the patient produce vocalizations to assess the movement of the vocal cords. While comparing motions, noting that the arytenoid cartilage behind the thyroid cartilage on both sides cannot be observed simultaneously due to anatomical constraints is important.

Nasogastric (NG) tube feeding or fasting was considered if complete vocal cord palsy was confirmed during the examination. Subsequently, the patients were categorized into the tracheostomy and non-tracheostomy groups, and a different testing method was performed on each group.

Non-tracheostomized Patients

A citric acid swallowing test was performed on non-tracheostomized patients to screen for the presence of silent aspiration. A citric acid solution (approximately 0.8 mol/L) was prepared by mixing 1 g of citric acid powder with 6 mL of normal saline. Patients were directed to ingest approximately 3 mL of the solution to determine whether coughing occurred due to silent aspiration [9]. To ensure accurate results, the patients were instructed to promptly swallow the solution without retaining it in their mouths because bicarbonate becomes diluted when mixed with saliva. A cough reflex occurring within 10 s of swallowing was considered a positive test result [10]. However, the next test step on the screening tool was not pursued if patients failed this stage.

Tracheostomized Patients

The patient’s tracheostomy tube cuff was deflated, and upper airway secretions were removed before checking the airflow through the upper airway to ensure test accuracy. Subsequently, the patient was administered 10 mL of methylene blue-dyed water; the presence or absence of aspiration was determined by visually detecting the dye exiting through the tracheostomy or by suctioning with a catheter immediately after drinking and 3 min later to assess for direct or delayed aspiration [11].

GUSS test

Patients were evaluated for their ability to resume an oral diet after passing the citric acid swallowing or methylene blue test. The GUSS, which is a direct swallowing test, was administered, involving the evaluation of semisolid, liquid, and solid items to determine the presence of swallowing difficulties, coughing, drooling, and voice changes [6]. Specifically, the GUSS stands out by employing diverse consistencies, starting with pudding and strategically minimizing aspiration risk during screening, and by its unique capability for simultaneous aspiration risk assessment and personalized dietary recommendations [12]. Patients who scored ≥ 14 points on the GUSS, including the initial 5 points obtained during the preliminary screening tool, were considered suitable for an oral diet, and dietary adjustments were made based on the patient’s score. In contrast, patients who scored < 19 points were considered to have dysphagia symptoms.

Results

A total of 13 patients diagnosed with severe COVID-19, comprising 9 men and 4 women with a mean age of 61.4 ± 13.9 (range 26–86) years, were included in the study. Among these, eight patients underwent intubation because of severe ARDS. All 13 patients underwent preliminary investigation. One patient did not pass the preliminary investigation and was immediately started on NG tube feeding without further evaluation. One patient did not undergo the vocal cord ultrasound test due to low cooperation, and the remaining 12 underwent subsequent screening. According to the vocal cord ultrasound test, all 12 patients had no complete bilateral vocal cord palsy, while 4 had unilateral vocal cord palsy. The patients were not restricted from further evaluation, as complete vocal cord palsy was not observed. Among the 12 patients, 1 did not pass the initial methylene blue dye test, and another did not pass the initial citric acid swallowing test. According to the results of the GUSS test, of the 10 patients, 9 and 1 initiated oral feeding with semisolids and pureed food, respectively. Patients who underwent tracheostomy had longer hospital stays and intubation periods and lower Medical Research Council sum scores than those who did not undergo tracheostomy. In the GUSS for diet type selection, non-tracheostomized patients scored higher on average and consumed more types of food than tracheostomized patients. All patients who passed the evaluation started oral feeding without developing any symptoms of aspiration and were discharged from the hospital without dysphagia-induced complications. One patient was transferred to a rehabilitation hospital, and the remaining 12 were discharged (See Table 1).

Table 1.

Patient demographic information and the results of the modified dysphagia screening tool

Variable		Total	Tracheostomy	Non-tracheostomy
Sex		13	5	8
	Men	9	3	6
	Women	4	2	2
Age (years)		61.4 ± 13.9	53.2 ± 14.9	66.5 ± 11.3
BMI (kg/m2)		24.7 ± 3.7	24.0 ± 1.8	25.1 ± 4.6
APACHE II		14.4 ± 2.8	14.4 ± 0.9	14.4 ± 3.6
Days in hospital LOS		34.2 ± 26.4	54 ± 33.3	21.9 ± 10.3
Days in ICU LOS		21.8 ± 18.7	38.4 ± 20.7	11.4 ± 6.0
Days of intubation		7.6 ± 6.0	12.2 ± 4.0	4.75 ± 5.4
MRC sum-score		41.8 ± 12.6	38.8 ± 12.5	43.8 ± 13.2
US vocal test
	Normal	8	2	6
	Unilateral palsy	4	2	2
	Bilateral palsy	0	0	0
	Not tested	1	1	0
Citric acid Swallowing test
	Pass			7
	Fail			1
	Not tested			0
Methylene blue dye test
	Pass		3
	Fail		1
	Not tested		1
GUSS score			11.8 ± 9.3	17.6 ± 2.5
GUSS severity
	No dysphagia	4	2	2
	Slight dysphagia	6	1	5
	Mild dysphagia	2	1	1
	Severe dysphagia	0	0	0
Diet after dysphagia evaluation
	NG tube	3	2	1
	Pureed	1	0	1
	Semisolid	9	3	6
	Solid	0	0	0
FOIS before dysphagia evaluation			1	1
FOIS after dysphagia evaluation			2.8 ± 1.6	4 ± 0.5
FOIS at discharge			3.6 ± 1.5	5 ± 0.5
Discharge place
	Home	12	4	8
	Hospital	1	1	0

APACHE, Acute Physiology and Chronic Health Evaluation; GUSS, Gugging Swallowing Screen; LOS, Length of Stay; MRC, Medical Research Council; US, Ultrasonography; NG, Naso Gastric; FOIS, Functional Oral Intake Scale

Discussion

Central to the management of dysphagia is the relatively precise identification of potential risks through a straightforward evaluation while concurrently implementing timely intervention. Our results showed that the modified dysphagia screening tool is feasible and effective for detecting swallowing impairments in the ICU setting. In future studies, there will be a need to validate its potential by comparing it with the accuracy of FEES testing. Consequently, we developed this novel dysphagia screening tool by combining existing dysphagia tools that were designed similarly to FEES in terms of accuracy and methods.

Impaired cognition and reduced consciousness are significant considerations before a full-scale dysphagia evaluation, as decreased orientation and inability to follow commands have been shown to increase the odds of liquid aspiration by 31% and 57%, respectively [13]. Therefore, in our study, the preliminary testing was mainly focused on whether the patient could maintain alertness for 15 min before starting the examinations to detect possible aspiration. Vocal cord movement disorders, such as vocal cord paralysis or arytenoid cartilage dislocation, are common complications in the ICU and can have severe implications for patient outcomes [14]. Particularly, bilateral vocal cord palsy is associated with a high risk of aspiration [15]. The best advantage of FEES is that it verifies vocal cord movement through endoscopy while the patient is vocalizing and swallowing [16]. To take advantage of this feature, we applied a combination of the FTU and LLU ultrasound to the modified dysphagia screening tool. This combination successfully visualizes the vocal cord abnormalities, and the diagnostic results are similar to the diagnostic rate of FEES [17]. Evaluating the level of alertness and assessing the movement of the vocal cords via ultrasound before conducting a comprehensive dysphagia evaluation are efficient safety measures. Therefore, implementing these measures can help minimize the need for unwarranted diagnostic testing and reduce the likelihood of aspiration during subsequent assessments.

Approximately 24% of mechanically ventilated patients in the ICU undergo tracheostomy [18]. Disruption of the upper airway by tracheostomy may lead to physiological and biomechanical changes in swallowing, thereby increasing the risk of dysphagia [19]. Therefore, distinct approaches to dysphagia evaluation are necessary for patients with and without tracheostomies. The methylene blue dye test for tracheostomized patients is a highly objective test [20] that can detect direct and delayed aspiration. In contrast, the citric acid swallowing test is used to validate silent aspiration in non-tracheostomized patients. Because citric acid is a tussigenic stimulus, it induces coughing by acidifying the upper airway, which triggers the discharge of action potentials in the jugular C-fibers and nodose A-δ fibers [21]. Therefore, employing distinct and appropriate assessments for patients with and without tracheostomies can enhance the objectivity and accuracy of the overall dysphagia evaluation.

Finally, we confirmed the suitability of diet selection using the GUSS, which has been validated in previous studies using FEES [7]. Although the GUSS is a valid and reliable test for identifying the risk of oropharyngeal dysphagia in healthy older adults without secondary dysphagia [22], its adoption alone may not be a valid dysphagia screening tool for patients in the ICU because most patients acquire dysphagia as a secondary condition, such as PED. Therefore, the GUSS was used to determine an appropriate diet.

Because all patients in our study were COVID-19 positive, aerosol testing for dysphagia was restricted. Although FEES is an accurate dysphagia evaluation method, it is unsuitable for patients with suspected viral transmission in isolation settings due to concerns regarding aerosolization. Therefore, a modified dysphagia evaluation method that incorporates pre-existing dysphagia assessments and involves fewer aerosol-generating procedures was necessary. Additionally, FEES is considered a high-risk procedure for airborne transmission despite its proven benefit in the evaluation of dysphagia in the ICU [23]. Direct stimulation of the nasopharynx, as well as the insertion and removal of a scope from the nasal cavity, has been shown to increase the risk of viral spread. Consequently, we could not conduct aerosol-based testing for dysphagia because all participants in our study were COVID-19 positive. As FEES may not be appropriate for patients with suspected viral transmission in isolation settings owing to concerns about aerosolization, we developed a modified dysphagia evaluation method. This new method incorporates preexisting assessments and involves fewer aerosol-generating procedures to ensure safety in isolation settings.

This study aimed to develop a modified swallowing examination for patients within the ICU or an isolation setting. Our results demonstrate that this modified, less invasive examination is feasible and effective for detecting swallowing impairments in these patient populations. This pilot study can be considered an advanced form of bedside swallowing screening compared with the previously validated GUSS-ICU, as it incorporates the use of methylene blue and vocal cord ultrasound examination, allowing for a more comprehensive swallowing evaluation that includes tracheostomized patients.

The modified dysphagia screening tool developed in this study can be applied in various clinical settings, such as rehabilitation units and nursing homes, where dysphagia is a prevalent concern. This newly developed screening tool is not only quick and easy to perform but is also suitable for high-volume clinical settings with limited resources. Finally, the modified dysphagia tool provides a cost-effective solution for assessing the swallowing function without the need for specialized equipment or personnel, thereby presenting a significant advantage to healthcare providers and patients.

Limitations

Challenges remain in vocal cord assessment, including the need for standardized ultrasound protocols and further research to establish the reliability and validity of ultrasonography for dysphagia assessment. Our study was limited by its small sample size and lack of validation through additional evaluations, such as VFSS and FEES. Therefore, large-scale validation studies are required to assess the efficacy of this modified screening tool. According to the former literature, basic abnormalities in vocal cord movement can be observed through approximately 40 practice sessions of ultrasound testing; however, the accuracy of the examination may be limited by the proficiency of the examiner. Despite these limitations, our results suggest that this tool can be used with little risk of transmission and can be performed safely and simply by medical staff in the ICU.

Conclusions

The evaluation of dysphagia is critical for patients in the ICU. VFSS and FEES are considered optimal methods for evaluating dysphagia, but their application can be limited in the ICU. This study implemented a relatively simple, reliable, and sensitive algorithm, resulting in a modified dysphagia screening tool. All patients who underwent the modified evaluation showed no aspiration symptoms after oral feeding, which enabled safe discharge. Therefore, the proposed modification of the dysphagia evaluation algorithm for patients in the ICU is safe and feasible. Subsequently, validation of FEES and VFSS, which are instrumental examinations for dysphagia, further substantiates our modified dysphagia screening tool as a clinically effective evaluation tool. Therefore, it can be considered an appropriate method for efficient dysphagia evaluation in ICU settings.

Author Contributions

Conceptualization: Myung Hun Jang, Yong Beom Shin, Myung Jun Shin. Data curation: Myung Hun Jang, Byung Hoon Lee, Jae Sik Seo. Formal analysis: Myung Hun Jang, Kwangha Lee. Methodology: Yong Beom Shin, Myung Jun Shin. Project administration: Yong Beom Shin, Myung Jun Shin. Supervision: Yong Beom Shin, Myung Jun Shin, Kwangha Lee. Validation: Myung Hun Jang, Jae Sik Seo. Roles/writing: Myung Hun Jang, Byung Hoon Lee, Jae Sik Seo.

Funding

This research was supported by the Bio&Medical Technology Development Program of the National Research Foundation (NRF) funded by the Korean government (MSIT) (No. RS-2023-00223764)

Data availability

The data that support the findings of this study are available from the corresponding author, MH Jang, upon reasonable request.

Declarations

Ethical approval

IRB No: 2307-015-129.

Competing interests

The authors declare they have no financial interests.