Effectiveness of the “CEME” oral health intervention program for preventing stroke-associated pneumonia in patients with post-stroke dysphagia: a randomized controlled trial

Abstract

Background

Pneumonia is common among post-stroke dysphagia (PSD) patients, especially in the week following a stroke; pneumonia may prolong hospital stays or lead to mortality. We examined whether a tailored “CEME” (oral cleaning, oral motor exercises, Chinese acupoint and salivary gland massage, and safe-swallowing education) oral health intervention program could improve swallowing function and oral health and reduce the incidence of stroke-associated pneumonia (SAP) in PSD patients.

Methods

We conducted an assessor-blinded randomized controlled trial in a tertiary hospital in China. Eighty-four PSD patients were recruited and randomly assigned to either the control group (n = 42) or the experimental group (n = 42). The control group received conventional oral care, whereas those in the experimental group participated in the “CEME” oral health intervention program, which included oral cleaning, oral motor exercises, Chinese acupoint and salivary gland massage, and safe-swallowing education. Outcome measures were conducted on day 7 of the implementation of the intervention, including stroke-associated pneumonia incidence, swallowing function (measured via the Water Swallowing Test (WST) and the Functional Oral Intake Scale [FOIS]), overall oral health (measured via the Oral Health Assessment Tool [OHAT]), salivary secretion status (measured via the modified Schirmer test [MST]), oral hygiene status (measured via the Plaque Index [PI]), and stroke severity (assessed via the National Institute of Health Stroke Scale [NIHSS]).

Results

The incidence of SAP in the experimental group was 10.26% (4/39), lower than the corresponding figure in the control group of 32.43% (12/37) (P < 0.05); furthermore, the WST grades, OHAT scores, and Plaque Index exhibited by the experimental group were better than the control group (P < 0.05). No significant differences were observed in FOIS grades, MST or NIHSS scores (P > 0.05).

Conclusion

The tailored CEME intervention effectively reduced SAP incidence in PSD patients while concomitantly improving oral health and water swallowing function.

Trial registration

ChiCTR2300074510, registered on August 08, 2023.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12903-025-07190-w.

Stroke-associated pneumonia (SAP), defined as pneumonia occurring in non-mechanically ventilated patients within seven days of stroke onset, affects approximately 5–56% of stroke patients [1, 2]. SAP exacerbates post-stroke disability, reduces quality of life, increases social and economic burden, and elevates mortality risk [3–5]. Compared with stroke patients without dysphagia, those with dysphagia have a markedly higher risk of developing pneumonia and demonstrate increased mortality rates [6, 7]. The primary mechanism of SAP involves the breach of the patient’s defense systems by oropharyngeal secretions containing pathogenic bacteria, aspirated food residues, or microorganisms delivered to the distal bronchus and alveoli via inhaled aerosols or gel particles [2, 8]. Once aspirated, these pathogens multiply in the lungs and cause invasive damage.

Patients with post-stroke dysphagia (PSD) often present with oral health impairments, including impaired oral sensation and motor function, salivary gland dysfunction, retention of food debris, and poor oral hygiene. These problems may act independently or interactively [9, 10]. Neural regulation deficits after stroke can reduce sensation in the face, lips, and tongue, while impairing oral coordination [11]. Consequently, bolus formation and propulsion are compromised, leading to food retention, poor oral hygiene, and choking [12, 13]. Salivary gland dysfunction, characterized by reduced secretion and decreased salivary protein content, further weakens the antibacterial and pH-regulating functions of saliva [14]. Diminished salivary flow also reduces lubrication, hindering bolus formation and transport, and exacerbates oral residue and microbiome imbalance. Under these conditions, the normal oral flora of PSD patients shifts, and oropharyngeal secretions containing pathogenic bacteria and food debris are more likely to be aspirated, increasing pneumonia risk. Therefore, maintaining oral health in PSD patients requires not only attention to oral hygiene, but also enhancing oromotor function, sensorimotor coordination, and improving salivary gland activity. Accordingly, stroke management guidelines emphasize that multidimensional interventions targeting oral health—particularly in dysphagic patients—are essential to reduce pneumonia incidence [15, 16]. Despite this, research on oral care management in stroke patients remains limited. Current studies mainly evaluate the effects of antimicrobial solutions [17, 18] or specialized oral care devices [19, 20], with relatively little focus on multidimensional strategies that integrate oral health, dysphagia management, and pneumonia prevention.

Our previous study [21] investigated SAP incidence in 211 PSD patients and revealed a high occurrence of SAP, accompanied by poor oral health scores, inadequate salivary secretion, and suboptimal oral hygiene. Based on these findings, we developed a predictive model for SAP in PSD patients, incorporating factors such as National Institute of Health Stroke Scale (NIHSS) score, Water Swallowing Test (WST) grade, Oral Health Assessment Tool (OHAT) score, oral hygiene, and salivary secretion. These results suggest that targeted oral care should address multiple domains, including swallowing rehabilitation, oral hygiene improvement, and salivary function promotion.

Following the UK Medical Research Council (MRC) Framework for Design and Evaluation of Complex Interventions [22–24], and integrating our SAP prediction model, we designed a multi-component, multi-pathway program named CEME. This acronym derives from its four core elements: Cleaning (oral cleaning), Exercises (oral motor exercises), Massage (Chinese acupoint and salivary gland massage), and Education (safe-swallowing education). The objective of this clinical trial was to assess whether the CEME program could improve oral health, enhance swallowing function, and reduce SAP incidence in PSD patients.

Methods

This study was registered with the Chinese Clinical Trial Registry prior to its initiation.

Design

An assessor-blinded randomized controlled trial was conducted. This study was approved by the research ethics committees of the participating hospitals. All participants provided written informed consent.

Participants

A convenience sample of inpatients in the Department of Neurology at the Second Affiliated Hospital of the Army Medical University was surveyed between August and December 2023. Eligible subjects were required to be older than 18 years of age, to be within 24 h of stroke onset, to achieve a water swallow test grade of 3 or higher, and to provide informed consent. We excluded participants who had a history of dysphagia, who were undergoing mechanical ventilation, who had severe facial injuries or who were participating in other clinical trials that would affect the results of this study. Intervention discontinuation criteria: planned completion of intervention per protocol; new indication for mechanical ventilation secondary to disease progression; consciousness deterioration preventing safe intervention delivery; participant withdrawal of consent.

Ethical considerations and trial registration

Ethical approval was obtained from the Medical Ethics Committee of the Second Affiliated Hospital of Army Medical University (PLA) (Approval No. 2023-New Technology No. 074 − 01). All procedures were conducted in accordance with relevant guidelines and regulations. Informed consent was secured from all participants. The study was registered with the China Clinical Trial Registry (Registration No.: ChiCTR2300074510; Registration Date: August 8, 2023; URL: https://www.chictr.org.cn/bin/project/edit? pid=203490).

Sample size

PASS 2021 software was used to estimate the sample size required for this research. The results of a pretest and literature review indicated that the incidence rate of SAP was 36% in the control group and 10% in the experimental group [2, 25]. Furthermore, the test efficacy was 80% (α = 0.05). The total sample size for the two groups was calculated to be 76 cases (including 38 cases in the control group and 38 cases in the experimental group). To account for potential withdrawals and losses to follow-up, we increased the sample size by 10%, resulting in a total sample size of 84 cases (42 in the control group and 42 in the experimental group).

Randomization and blinding

The researchers systematically screened stroke patients admitted to the hospital between August and December 2023. Subjects who met the inclusion criteria were consecutively recruited and randomly assigned to either the experimental or control group with the assistance of SPSS 23.0 software (with a fixed value of 2,000,000). Allocation concealment was achieved by using opaque closed envelopes. Individuals who were blinded to the group assignments assessed the endpoint outcomes.

Intervention

The “CEME” oral health intervention program was implemented for the experimental group. This program included four aspects: oral cleaning, oral motor exercise, Chinese acupoint and salivary gland massage, and safe-swallowing education (Table 1). The control group received routine oral care (Table 2), which was administered in accordance with the Canadian Oral Care Best Practice Guidelines, Oral Health: Supporting Adults Who Require Assistance Second Edition [16], and the Oral Health Guidelines for Chinese Residents [15]. The intervention commenced within 24 h of patient enrollment and continued for 7 days.

Table 1.

The “CEME” oral health intervention program

Contents	Form	Time/Frequency	Intervention Methods
Oral Cleaning	Nursing staff provide bedside guidance or implementation.	Oral care at least 3 times a day (including at least 2 brushings); more often if necessary, depending on the oral condition.	C1. Conduct an oral assessment using an oral health assessment tool.
			C2. Use an electric toothbrush for brushing.
			C3. Brush teeth with a pasteurized toothbrush. Hold the toothbrush at a 45° angle to the root of the tooth, with bristles extended toward the root. Brush two or three teeth at a time for at least three seconds. Brush the outer, inner and occlusal surfaces of the teeth in turn. Subsequently, brush your tongue. The brushing process should last at least two to three minutes.
			C4. Check and clean the mouth of food debris after each transoral meal; specifically, this can be accomplished by rinsing the mouth with warm water, brushing the teeth, or using a gauze-wrapped finger to reach into the mouth to remove food residue.
			C5. If the patient is unable to manage their saliva, the nurse should use a suction device to prevent aspiration. Use low-foaming fluoride toothpaste and avoid long-term use of antibacterial mouthwashes.
Oral Motor Exercises	Nursing staff guidance at bedside.	After oral cleansing, Repeat each exercise set 3 times, at least 3 times a day. Intensive exercises for damaged areas of the mouth.	E1. Set 1, lip and mouth movements: Open mouth - Pursed lips - Pouting lips - Grinning.
			E2. Set 2, cheek exercise: Lips closed - Cheeks puffed out - hold for 30 s - Exhale and relax - Lips closed - Cheeks puffed out - Air shifted in left and right cheeks.
			E3. Set 3, tongue movement: Extend the tongue to touch the left corner of the mouth - Retract the tongue tip against the palate - Extend the tongue to touch the peak of the lips - Retract the tongue tip against the palate - Extend the tongue to touch the right corner of the mouth - Retract the tongue against the palate - Move the tip of the tongue clockwise around the teeth - Move the tip of the tongue counterclockwise around the teeth (For patients with paralyzed tongue muscles, assist with passive movement).
			E4. Set 4, chewing and swallowing exercise: Empty chewing - Deep inhalation - Breath holding (2 s) - Empty swallowing 2 times - Coughing.
Acupoint and Salivary Gland Massage	Nursing staff implementation at bedside.	After oral cleansing, Stimulate each area for 10 s, three times a day.	M1. Stimulate the buccal area, soft palate, palatal arch, root of the tongue with light pressure using the vibration of the brush head of the electric toothbrush.
		Stimulate each point for 30 s, three times a day.	M2. Stimulate saliva secretion by applying light pressure with an electric toothbrush to the sublingual points, including the Jinjin and Yuye points (located on the left and right sides of the tongue tie).
		Perform 30 sets of massage for each area, twice a day.	M3. ChengJiang point: press the index finger or thumb into the middle depression of the lower lip groove on the face.
			M4. Parotid gland: Use the four fingers (other than the thumb) to press firmly against the patient’s cheeks near the earlobes on both sides, and perform circular pressing actions in both clockwise and counterclockwise directions.
			M5. Submaxillary gland: Press from under the ear along the edge of the mandible using the pad of the thumb.
			M6. Sublingual gland: press the thumb or index finger of both hands against the lower jaw and press upward.
Safe-swallowing Education	Nursing staff provide bedside one-to-one health education, video presentations, or live demonstrations	Health education is provided within 24 h of admission, on Mondays, and daily.	E1. Choose foods that are uniform in density, not easily dispersed, easily deformed, and easy to chew.
			E2. Choose appropriate dining utensils and encourage patients to eat independently.
			E3. Check and clear oropharyngeal secretions before eating.
			E4. If the patient cannot sit up, raise the head of the bed at least 30°. Additionally, place a soft pillow under the shoulder on the hemiplegic side. Have the patient bend their head forward or tilt it toward the healthy side. Do not let the patient lie flat.
			E5. During swallowing, place food near the base of the tongue or into the healthy cheek.
			E6. Take one bite at a time; if there is still food left in your mouth, you should swallow more than once.
			E7. Food and medicines for tube feeders should be administered by or under the direction of a nurse.
			E8. Avoid talking while swallowing to prevent aspiration and choking.
			E9. Remain seated for 30 min after eating and avoid lying down immediately.
			E10. Check and clean food particles from the mouth immediately after eating.

Table 2.

Oral health intervention program for the control group

Form	Time/Frequency	Intervention Methods
Nursing staff provide bedside guidance or implementation.	Oral care should be performed at least twice daily; more frequently if necessary, depending on the oral condition.	(1) The nurse assesses the oral condition of the patient daily (2) The nurse guides and assists the patient in brushing their teeth at least twice daily for oral care; the frequency can be increased if necessary, based on the patient’s oral condition. (3) Brush teeth for at least 2–3 min each time. (4) Clean the mouth using a toothbrush or cotton swabs. (5) Use low-foaming fluoride toothpaste for patients who are unable to independently clear sputum and saliva from the oral cavity. Avoid long-term use of chlorhexidine and other antibacterial mouthwash components; if necessary, follow the doctor’s recommendations. (6) Nurses should have a negative pressure suction device ready for use to prevent accidental suction. (7) Provide oral health education, swallowing safety guidance, and support to patients and their families.

Measures

Baseline data

Baseline assessments were performed within 24 h of admission. Collected data included demographic variables (age, sex, smoking status, denture use, and toothbrushing habits), clinical characteristics (level of consciousness, comorbidities, and stroke subtype), nutritional status [Nutritional Risk Screening 2002 (NRS-2002)], and functional status (Barthel Index). Data were obtained from the hospital electronic medical record system or through direct patient interviews.

Outcome measures

The primary outcome was the incidence of stroke-associated pneumonia (SAP). SAP diagnosis followed the Chinese Expert Consensus on Diagnosis and Treatment of Stroke-Associated Pneumonia (2019 Update) [2].

The diagnostic criteria for SAP encompass three aspects: clinical symptoms, physical signs, and imaging findings (detailed in Supplementary 2). An investigator blinded to group allocation confirmed SAP occurrence by reviewing medical records at the end of the intervention (day 7 after stroke onset).

The secondary outcomes included swallowing function, oral health status, oral hygiene, salivary secretion, and stroke severity. These were assessed by a blinded examiner at baseline (within 24 h of admission) and at the end of the intervention (day 7 after stroke onset). The following tools were used:

Water Swallowing Test (WST): Evaluates swallowing function by recording the time required to consume 30 ml of warm water, number of swallows, and presence of hoarseness or coughing/choking. Grades range from 1 (normal) to 5 (severe dysfunction). Grades 3 ~ 5 indicate abnormal swallowing, with higher grades reflecting poorer function [26]. To minimize aspiration risk, participants first attempted to swallow 3 ml of water sublingually. Failure at this step resulted in discontinuation of further testing and classification as WST grade 5.

Functional Oral Intake Scale (FOIS): A seven-level scale assessing current oral intake, ranging from “nothing by mouth” to “total oral diet with no restrictions.” Lower scores indicate poorer intake capacity [27, 28].

Oral Health Assessment Tool (OHAT): Recommended in international stroke guidelines [16], OHAT evaluates eight items (lips, tongue, gums, saliva, natural teeth, dentures, oral cleanliness, and dental pain). Each item is scored from 0 (healthy) to 2 (unhealthy), yielding a total score of 0–16, with higher scores reflecting poorer oral health [29].

Modified Schirmer Test (MST): Assesses salivary flow by placing Schirmer strips under the tongue for 3 min. A wetting length of ≤ 25 mm indicates reduced salivary secretion [30].

Plaque Index (PI): Determined using O’Leary Plaque Control Record method [31]. The PI is calculated as: (number of plaque−positive surfaces/total surfaces examined) × 100%. Each tooth was divided into four surfaces. A PI ≥ 20% indicated unsatisfactory oral hygiene.

Stroke Severity: Assessed using the National Institutes of Health Stroke Scale (NIHSS).

Adverse Events: All adverse events during the intervention were documented, including aspiration during oral care, oxygen desaturation, and choking episodes.

Statistical analysis

Statistical analysis was performed using IBM SPSS Statistics and R Statistical Software. The appropriate descriptive statistics were calculated for each variable. Normality was assessed using the Kolmogorov-Smirnov test. Continuous variables with normal distribution are presented as means with standard deviation (SD) and compared between groups using independent samples t-tests. Non-normally distributed continuous variables are expressed as medians with interquartile ranges (IQRs) and analyzed with Mann-Whitney U tests. Categorical data are presented as frequencies (n) with percentages (%) and evaluated using chi-square or Fisher’s exact tests as appropriate. Outcome data were compared between the CEME and control groups, with proportional differences reported for categorical variables along with an effect size. All statistical tests were two-sided and conducted at a significance level of 0.05.

Results

A total of 1,078 patients who had experienced stroke were screened between August and December 2023. Among these patients, a total of 84 met the criteria for inclusion in the trial and were randomly assigned to either the experimental or control group. During the study, 6 patients (3 in the control group and 3 in the experimental group) were lost to follow-up, and 2 patients were withdrawn due to mechanical ventilation and the inability to consciously cooperate.This process resulted in a final sample size of 76 patients (Fig. 1). A comparison of the baseline characteristics between these two groups of patients revealed no statistically significant differences (Table 3).

Fig. 1.

Trial flow diagram

Table 3.

Comparison of baseline data between the experimental group and the control group

Variables	Experimental Group(n = 39)	Control Group (n = 37)	χ2/t/Z	P value
Age #	65.67 ± 10.32	66.24 ± 11.94	0.930 a	0.338
Gender,%			1.653 b	0.199
Male	23(58.97)	27(72.97)
Female	16(41.03)	10(27.03)
Education level, %			2.833 c	0.412
Primary and below	15(38.46)	18(48.65)
Junior high school	14(35.90)	13(35.14)
Some college or university	10(25.64)	5(13.51)
University degree or higher	0(0.00)	1(2.70)
smoking status, %			3.318 b	0.069
Yes	13(33.33)	20(54.05)
No	26(66.67)	17(45.95)
Denture use, %			0.030 b	0.861
Yes	8(20.51)	7(18.92)
No	31(79.49)	30(81.08)
toothbrushing habits, %			0.004 b	0.951
≥ 2 times/day	37(94.87)	34(91.89)
< 2 times/day	2(5.13)	3(8.11)
Stroke type, %			0.323 b	0.570
Ischemic stroke	38(97.44)	34(91.89)
Hemorrhagic stroke	1(2.56)	3(8.11)
Level of consciousness, %			0.000 b	1.000
Clear-headed state	34(87.18)	32(86.49)
Somnolence state	5(12.82)	5(13.51)
NIHSS score*	5(3, 10)	5(4, 10)	−0.445 d	0.656
Comorbidities, %
Hypertension	27(69.23)	22(59.46)	0.791 b	0.374
Diabetes	10(25.64)	7(18.92)	0.494 b	0.782
Chronic respiratory disease	1(2.56)	1(2.70)	0.000 b	1.000
Heart disease	9(23.08)	4(10.81)	0.730 b	0.393
Barthel score, %			3.458 c	0.323
100 points	0(0.00)	2(5.41)
61 ~ 99 points	13(33.33)	7(18.92)
41 ~ 60 points	8(20.51)	9(24.32)
≤ 40 points	18(46.15)	19(51.35)
NRS-20021, %			0.043 b	0.836
< 3 points	22(56.41)	20(54.05)
≥ 3 points	17(43.59)	17(45.95)
WST grades *	3(3, 4)	3(3, 4)	−0.282 d	0.778
FOIS grades*	5(5, 6)	5(4, 6)	−0.139 d	0.889
MST, %			0.005 b	0.945
> 25 mm/3 min	25(64.10)	24(64.86)
≤ 25 mm/3 min	14(35.90)	13(35.14)
Plaque index, %			3.498 b	0.061
< 20%	9(23.08)	16(43.24)
≥ 20%	30(76.92)	21(56.76)
OHAT scores *	4(3, 4)	3(2, 5)	−0.436 d	0.663

1: NRS-2002 is used for rapid screening of nutritional risk in hospitalised patients. A total score of ≥ 3 indicates nutritional risk [30].

NIHSS National Institute of Health Stroke Scale, NRS-2002 Nutritional Risk Screening 2002, WST Water Swallowing Test, FOIS Functional Oral Intake Scale, MST Modified Schirmer Test, OHAT Oral Health Assessment Tool

a: t value; b: χ² value; c: Fisher's Exact Test; d: Z value

#Continuous variables with normal distribution are presented as Mean ± standard deviation and compared between groups using independent samples t-tests

*The data of course of duration was non-normal, using median (interquartile range, IQR) to describe and Mann–Whitney U test

Within seven days of stroke onset, the incidence of SAP was 32.43% (12/37) in the control group and 10.26% (4/39) in the experimental group, moreover, the difference between these two groups was significant. Compared with the control group, the experimental group exhibited superior improvements in oral hygiene, OHAT scores, and WST test results (Table 4). No intervention-related adverse events occurred in either group.

Table 4.

Comparison of outcome indicators between the experimental and control groups

Variables	Experimental Group (n = 39)	Control Group (n = 37)	χ2/Z	Effect size	P value
SAP, %			5.818 a	4.121	0.018
Yes	4(10.26)	12(32.43)
No	35(89.74)	25(67.57)
WST grades*	3(2, 3)	3(3, 4)	−2.382 b	−0.292	0.017
FOIS grades*	6(5, 6)	5(4, 6)	8.625 b	0.182	0.169
MST, %			0.446 a	1.431	0.504
> 25 mm/3 min	31(79.49)	27(72.97)
≤ 25 mm/3 min	8(20.51)	10(27.03)
Plaque index, %			4.206 a	3.211	0.040
< 20%	34(87.18)	25(67.57)
≥ 20%	5(12.82)	12(32.43)
OHAT scores*	2(1, 2)	2(2, 4)	−3.750 b	−0.482	< 0.001
NIHSS scores*	5(3, 8)	5(3, 8)	−0.444 b	−0.062	0.657

WST Water Swallowing Test, FOIS Functional Oral Intake Scale, OHAT Oral Health Assessment Tool, NIHSS National Institute of Health Stroke Scale

a: χ² value; b: Z value;1: OR; 2: Cliff’s δ

*The data of course of duration was non-normal, using median (interquartile range, IQR) to describe and Mann–Whitney U test

Discussion

This randomized controlled trial evaluated the CEME oral health intervention program specifically developed for stroke patients with dysphagia. Grounded in the MRC Framework for Design and Evaluation Complex Interventions, the program integrates the Stroke–Dysphagia–Oral Health–SAP pathophysiological pathway with clinical prediction models. It comprises four components: oral cleaning, oral motor exercises, Chinese acupoint and salivary gland massage, and safe-swallowing education. A complex intervention, by definition, consists of multiple components that may act independently as well as synergistically. Each contributes to overall efficacy, though identifying the specific active ingredients is often challenging [32, 33]. For post-stroke dysphagia, targeted oral care must simultaneously address swallowing recovery, oral hygiene, and saliva stimulation, with strong interactions among these domains. The CEME program thus fully reflects the characteristics of complex interventions—multi-component, multi-pathway, and context-dependent. Results demonstrated that CEME significantly improved oral health and water swallowing function, and importantly, reduced the incidence of SAP.

Maintaining good oral hygiene is essential for oral health and the prevention of pulmonary infections. After intervention, the oral hygiene status of both groups of patients improved from baseline, indicating that even routine oral cleaning techniques can improve patients’ oral hygiene status. However, the experimental group exhibited a significantly higher oral hygiene compliance rate (87.18%, 34/39) than did the control group (67.57%, 25/37). Toothbrushing remains the fundamental method of plaque removal [34], and numerous studies have confirmed that the high-frequency vibration generated by electric toothbrushes is more effective at removing plaques than the vibration produced by ordinary toothbrushes, thereby improving oral health and reducing the risk of pneumonia [35–37]. Consistent with these reports, plaque removal in our experimental group using an electric toothbrush was more effective than in the control group. Furthermore, because many stroke patients suffer from hemiplegia and impaired self-care ability, electric toothbrushes offer both enhanced plaque removal and improved feasibility in this population. These findings support the preferential adoption of electric toothbrushes for frail or functionally impaired groups.

Saliva plays a crucial role in maintaining oral homeostasis and regulating the microbial community, while its lubricating function facilitates bolus formation and transport during swallowing [38, 39]. Salivary secretion is dually regulated by the autonomic nervous system (parasympathetic and sympathetic pathways) [40]. The reflex arc involves: (1) neural impulses generated by oral gustatory/mechanoreceptors, transmitted via the facial or glossopharyngeal nerves to the medullary salivary nuclei; (2) efferent signals returning through the same pathways to stimulate salivary glands. Additionally, traditional Chinese Medicine (TCM) posits that massaging acupoints (e.g., Chengjiang, Yuye, and Jinjin) activates meridians to promote saliva production [41]. Clinical evidence demonstrates that: Type 2 diabetes mellitus patients receiving thrice-daily salivary gland massage for 3 months showed significant improvement in xerostomia and oral hygiene [42]. Sjogren’s syndrome patients exhibited increased salivary flow rates following acupoint massage combined with atomization therapy [43]. In contrast, our study found no statistically significant difference in salivary secretion between intervention and control groups among post-stroke dysphagia patients receiving combined acupoint and salivary gland massage. This null finding may be attributed to: Insufficient intervention duration. Prior evidence supporting massage efficacy derived from interventions three month or more [44] whereas our 7-day protocol may be inadequate for neural reorganization. Impaired neural pathways in stroke patients, may reduce mechanical sensitivity, weakening the salivary reflex arc. Salivary output depends on frequency and intensity of neuro-mediated stimuli, with insufficient stimulation potentially diminishing the masticatory-parotid reflex [14, 45]. Confounding clinical factors, such as stroke-associated conditions (e.g., diabetes), medication side effects [46], and dietary modifications (e.g., nasogastric feeding, soft diets) may counteract intervention effects [5, 47]. Limitations in salivary assessment: Our study measured salivary flow rate only at a single time point, which may not capture dynamic changes or diurnal variations in secretion. Existing studies emphasize that saliva collection methods (e.g., resting vs. stimulated) and timing significantly influence outcomes [40, 48]. Future study should extend the intervention period and integrate neuromuscular electrical stimulation, gustatory agents, or acupuncture to systematically evaluate longitudinal changes in salivary flow rates and composition (e.g., mucins, lysozyme) for optimizing rehabilitation strategies targeting salivary dysfunction in stroke survivors.

For patients with post-stroke dysphagia, restoring oral function is vital for maintaining oral health and preventing pneumonia. Key objectives of dysphagia rehabilitation include enhancing orofacial muscle strength and improving the swallowing reflex [26, 49]. Our results showed that lip and tongue exercises, swallowing maneuvers, and vibratory stimulation using an electric toothbrush effectively enhanced swallowing ability. However, improvements in functional oral intake were less pronounced. Vibratory stimulation of sensory receptors in the oral cavity, buccal mucosa, tongue, or facial regions can increase the volume of sensory input transmitted to the brainstem swallowing center, thereby strengthening the patient’s swallowing reflex [50]. Furthermore, oral motor exercises can improve parameters such as orofacial muscle strength, tongue elevation, tongue protrusion, cheek compression, lip compression, and alternating motion rate in dysphagia patients [51]. Chen et al. [52] reported that a 4-week regimen significantly enhanced oromotor function in dysphagia patients, and a subsequent trial demonstrated improved oral health after 3 weeks of intervention [53]. However, no significant effects were seen on FOIS scores or nasogastric tube removal, likely due to insufficient intervention duration. Prior research suggests that ≥ 6 months may be required to achieve meaningful improvements in oral intake ability [54]. Future studies should therefore incorporate intermediate physiological measures (e.g., tongue strength, elevation amplitude) as secondary outcomes to capture earlier treatment effects.

This study has several limitations. The short-term intervention (7 days) precluded assessment of long-term outcomes, including sustained FOIS improvements, stroke severity modulation, cost-effectiveness metrics (e.g., hospitalization duration/expenditure), and longitudinal prognostic trajectories. Recruitment from a single institution limits the generalizability of the findings. Future multi-center RCTs are warranted to enhance external validity across diverse healthcare settings. Additionally, while the use of a low-foaming fluoride toothpaste was recommended in accordance with current guidelines, the lack of standardization regarding specific brands or fluoride concentrations may introduce a potential source of variation. However, this pragmatic approach may improve practicality and compliance in actual clinical practice. The complex intervention design incorporates components that may operate independently or interact synergistically. Although each element contributes to therapeutic efficacy, identifying specific active constituents remains challenging. Future studies should adopt a step-wedge methodology with sequential component introduction to reduce operational burden and isolate individual therapeutic contributions.

Conclusions

Functioning as a tailored oral health intervention for post-stroke dysphagia, the CEME Oral Health Program significantly reduces SAP incidence, with parallel improvements in oral health and water swallowing ability.

Abbreviations

FOIS

Functional Oral Intake Scale

MRC

Medical Research Council

MST

Modified Schirmer test

OHAT

Oral Health Assessment Tool

PI

Plaque Index

SAP

Stroke-associated Pneumonia

Authors’ contributions

Qian Li designed the study, performed the data analysis and interpretation, and drafted the manuscript.Yan Tian and Juan Xiang collected the data. Ting Xu managed and verified the data. EnJie Tang provided statistical advice and recommendations. Yan Ling Li provided critical feedback and helped shape the interpretation of the results. Jing He conceived and supervised the study, reviewed and edited the manuscript, and served as the corresponding author. All authors read and approved the final manuscript.

Funding

This work was funded by the Nursing Cultivation Project of the Second Affiliated Hospital of the Army Military Medical University (2024HLPY-05).

Data availability

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

Declarations

Ethics approval and consent to participate

This study was approved by the Human Research Ethics Committee of the Second Affiliated Hospital of Army Medical University (PLA) (Approval No. 2023-New Technology No. 074 − 01). It was conducted in accordance with the principles of the Declaration of Helsinki and relevant guidelines, and informed consent was obtained from all participants.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.