Prevalence and Management of Oral Intake Restrictions in Critically Ill Patients: Insights from a Multicenter Point Prevalence Study

This was a multicenter, prospective, cross-sectional, 2-day point prevalence study conducted in Japanese ICUs, approved by the Okayama University Ethics Committee (K2310-031). The study was spearheaded by the Japanese Society for Early Mobilization and designed in accordance with the STROBE guidelines for cross-sectional studies. In line with the Ethical Guidelines for Medical and Health Research Involving Human Subjects in Japan [15], the requirement for ethical approval at each participating institution was waived, as no participant identifiable data were collected in this study. This ethical protocol was shared with all clinicians, with ICU participation depending on their agreement after reviewing regional ethics policies.

First, the study recruited participating institutions from September 8 to October 31, 2023, by sending an invitation letter to members of the Japanese Society for Early Mobilization, along with other local networks This letter provided a brief overview of the study, including a link to the website or a paper-based questionnaire that detailed the study’s objectives and ethical considerations. Upon consenting to participate, representatives from each ICU—comprising intensivists, physician, critical care nurses, speech and language therapists (SLTs), physical therapist, and registered dietitian nutritionist—were asked to submit basic information about their hospital/ICU’s demographics and background. After this information was provided, each ICU was registered as a participating site. The study then conducted 2-point prevalence surveys on Wednesday, November 1, and Friday, December 1, 2023, both at 10 a.m. Participating ICUs were asked to complete the survey and to write the same representative’s name on each one. The day before each survey date, the institution’s representatives received a URL for an online questionnaire or a paper-based questionnaire for the ICU care survey, enabling them to respond via their preferred method. To ensure the quality of the survey and consistent answers, participants were encouraged to collaborate with other healthcare professionals when completing the questionnaire.

A panel of 9 intensivists collaborated to create and review a structured questionnaire. This questionnaire was developed based on the findings from previous studies and our clinical interests [11.16]. The questionnaire sought basic information on hospital/ICU demographics and detailed data on swallowing evaluation and rehabilitation practices prior to the 2-day prevalence study. Additional File 1 presents the initial survey (24 questions, estimated completion time: 20 min) covering hospital and ICU type, number of hospital and ICU beds, number of SLTs, presence of a protocol for swallowing rehabilitation, and components of swallowing screening and rehabilitation. A dedicated SLT is defined as an SLT who exclusively works in the ICU and is not assigned to other wards. The 2-point prevalence study was designed to capture patient characteristics. Additional File 2 provides a guide for representatives conducting a 2-day prevalence study (79 questions, estimated completion time: 60 min). In addition to baseline characteristics of the patients, detailed information from ICU admission to the survey date, including data on the methods and personnel involved in swallowing screening and assessing swallowing function, and details of swallowing rehabilitation, encompassing both compensatory and behavioral swallowing rehabilitation strategies. Compensatory strategies include postural adjustments to make swallowing safer, as well as diet modifications, such as adjusting the texture of food (e.g., pureeing solids or thickening liquids), to make swallowing easier and safer. Behavioral swallowing rehabilitation includes techniques such as ice massage, oral motor exercise, thermal-tactile stimulation, and salivary gland massage. Ice massage is widely used in Japan as a pre-feeding technique to induce dry swallowing, stimulate swallowing apraxia to initiate the swallow action, and as part of daily swallowing training, which involves applying an ice stick applied to the throat, base of the anterior faucial arches, base of the tongue, and posterior pharyngeal wall [17]. Swallowing function at the time of survey was collected using the Functional Oral Intake Scale (FOIS), which is a tool commonly used to measure oropharyngeal dysphagia. Scores vary from 1, indicating no oral intake, to 7, which signifies a complete oral diet without any restrictions; higher scores reflect improved swallowing ability [4]. To distinguish dysphagia from oral intake restrictions due to other reasonable medical reasons, we additionally collected information on the number of patients whose primary reasons for oral intake restrictions were categorized into five groups: respiratory instability, hemodynamic instability, altered level of consciousness, gastrointestinal issues, and others. Furthermore, we conducted an additional survey on the role of dentists in the ICU, as they are considered to play a significant role. This survey was conducted on June 29, 2024 (Additional File 3). To ensure that the responses reflected the situation at the time of the original surveys conducted in 2023, all respondents from the initial survey were asked to provide information based on the circumstances in 2023.

The primary outcome was the prevalence of oral intake restrictions in patients, defined as having a FOIS score of less than 7 among eligible patients for oral intake [18]. Eligible patients for oral intake were defined as those excluding patients who had endotracheal intubation and those with oral intake restrictions due to reasons other than dysphagia, as mentioned in the 5 categories above. Secondary outcomes included the implementation rates of dysphagia-related care among eligible patients for oral intake, such as swallowing screening, swallowing rehabilitation and diet modification, and swallowing protocol among each participating ICU. Further analyses were conducted on specific patient subgroups, namely those post-extubation and post-stroke [4, 5].

We used descriptive statistics to outline the baseline characteristics of both the participating ICUs and patients, focusing on the overall cohort, those eligible for oral intake, as well as post-extubation and post-stroke patients in a subgroup analysis. The prevalence of oral intake restrictions and practice patterns for dysphagia were estimated based on the proportion of patients eligible for oral intake in the two-day survey. Continuous variables were described using the median and interquartile ranges (IQR), whereas categorical variables were presented as numbers and percentages. Statistical analyses were performed using Stata version 18 (StataCorp LP, College Station, TX).

A total of 21 ICUs across Japan participated in the study, with the predominant ICU type being mixed medical-surgical (95.3%, 20/21). The majority of survey were provided by intensivists, who accounting for 61.9% (13/21) of the contributions. Table 1 shows characteristics of the participating institutions.

No ICUs had dedicated SLTs, and majority (85.7%, 18/21) did not have specific protocol for swallowing rehabilitation. Dysphagia screening was predominantly conducted using the water swallowing test (76.1%, 16/21), with nurses as the main screeners (85.7%, 18/21). The most frequently utilized rehabilitation techniques were ice massage (57.1%, 12/21) and oral exercises (52.3%, 11/21), primarily administered by nurses (42.8%, 9/21) and SLTs (47.6%, 10/21). Although dentists provide dental services in 61.9% (13/21) of participating ICUs, only 1 ICU (4.7%) is involved in the assessment or treatment of dysphagia. Table 2 provides further details on the characteristics of dysphagia management and swallowing rehabilitation practices in the ICUs, as obtained from the initial survey.

Table 3 summarizes the findings from a 2-day point prevalence survey. Among the 326 participants, 124 (38.0%) were intubated, 56 (17.1%) were post-extubation, and 25 (7.6%) had a tracheostomy on the survey date at 10 a.m. A total of 32 patients (9.8%) were suffering from a stroke. Of the 326 participants, 139 were not eligible for oral intake due to endotracheal intubation (89.2%, 124/139), gastrointestinal issues (6.4%, 9/139), respiratory instability (0.7%, 1/139), hemodynamic instability (0.7%, 1/139), altered level of consciousness (0.7%, 1/139), and others factors (2.1%, 3/139).

Of the 187 patients eligible for oral intake after excluding these 139 patients, 69.0% (129/187) exhibited dysphagia, leading to oral intake restrictions, as defined by a FOIS score of 6 or less. The FOIS score distribution was as follows: 18.7% with FOIS = 1, 19.8% with FOIS = 2 or 3, 6.4% with FOIS = 4, and 24.1% with FOIS = 5 or 6 (Fig. 1A).

Of the 187 who were eligible for oral intake, 52.4% (98/187) underwent swallowing screening, predominantly using the water swallowing test or its modified version, administered by nurses (87.8%, (86/98)). Among those screened, 36.7% (36/98) were suspected of having dysphagia. Among those who were not screened, 5.6% (5/89) were suspected of having dysphagia based on clinical symptoms such as coughing or choking during meals, and the presence of aspiration pneumonia. The implementation rate of videofluoroscopic swallowing studies (VFSS) or fiberoptic endoscopic evaluation of swallowing (FEES) for swallowing assessment was only 1.0% (2/187). Compensatory and behavioral swallowing rehabilitation were provided to 21.9% (41/187) and 10.6% (20/187) of patients, respectively, from ICU admission to the survey date. Among the patients with suspected dysphagia, compensatory and behavioral rehabilitation were applied to 50.0% (18/36) and 36.1% (13/36), respectively. These interventions were primarily administered by nurses and SLTs. SLTs were involved in swallowing function assessment and swallowing rehabilitation for less than 10% of all patients.

Table 4 presents the demographics and characteristics of post-extubation and post-stroke patients. Of the 17 post-extubation patients with an FOIS of 1, 12 were potentially eligible for oral intake after excluding those with oral intake restrictions due to reasons other than dysphagia (n=5). In post-stroke patients, all those with an FOIS of 1 (n=16) were eligible for oral intake. The prevalence of oral intake restriction among eligible patients for oral intake was 78.4% (40/51) in post-extubation patients (Figure 1B) and 87.5% (28/32) in post-stroke patients (Figure 1C), respectively. Swallowing screening was conducted for 78.4% (40/51) of post-extubation patients and 21.8% (7/32) of post-stroke patients. Compensatory and behavioral rehabilitation were provided to 27.4% (14/51) of post-extubation patients and 9.3% (3/32) of post-stroke patients, with these interventions primarily administered by SLTs.