Long-term effects of voice training and swallowing function exercises on postoperative functional recovery in laryngeal cancer patients: a retrospective cohort study

Abstract

Objective

To investigate the long-term effects of combined voice training and swallowing function exercises on functional recovery, quality of life, and complications in patients after laryngeal cancer surgery.

Methods

In this retrospective cohort study, 120 patients who underwent laryngeal cancer surgery between June 2019 and June 2023 were divided into a rehabilitation group (n = 60, receiving systematic voice and swallowing training) and a routine care group (n = 60). Voice Handicap Index (VHI), Swallowing Function Scale (SFS), and EQ-5D were used to assess outcomes at 6, 12, and 24 months postoperatively. Complications were also recorded.

Results 

At 12 months postoperatively, the rehabilitation group showed significantly better outcomes than the routine care group: lower VHI score (35.25 ± 5.12 vs. 58.70 ± 6.84, P < 0.001), lower dysphagia incidence (10.00% vs. 30.00%, P = 0.008), higher EQ-5D index (0.82 ± 0.09 vs. 0.68 ± 0.11, P < 0.001), lower total complication rate (11.67% vs. 26.67%, P = 0.032), and higher satisfaction score (87.80 ± 7.93 vs. 72.21 ± 9.22, P < 0.001). Multivariate analysis confirmed comprehensive rehabilitation as an independent protective factor for voice function (β = −9.215, P < 0.001), swallowing function (β = −3.268, P < 0.001), and complication reduction (OR = 0.218, P < 0.001).

Conclusion

Systematic combined voice and swallowing rehabilitation training can effectively promote functional recovery, significantly improve long-term quality of life, and reduce the risk of postoperative complications in laryngeal cancer patients.

Introduction

Laryngeal cancer is one of the common malignant tumors of the head and neck, among which the incidence of laryngeal squamous cell carcinoma accounts for 95% of laryngeal malignancies and 5% of systemic malignancies [1]. As the tumor progresses, its impact on laryngeal functions such as phonation, respiration, and swallowing gradually exacerbates [2]. Currently, surgical treatment remains the primary radical approach for laryngeal cancer, including procedures such as partial laryngectomy and total laryngectomy. While effective in controlling tumor progression, postoperative complications such as voice dysfunction and swallowing abnormalities are common [3]. Studies indicate that approximately 30–50% of patients experience dysphagia, manifesting as choking during eating, aspiration, etc., significantly increasing the risk of malnutrition and aspiration pneumonia [4, 5]. Over 60% of patients have long-term voice abnormalities postoperatively, such as hoarseness and vocal fatigue, leading to communication limitations and social avoidance [6]. These functional impairments not only harm physical health but also profoundly affect the patient’s psychological state and social functioning. Therefore, preserving laryngeal function has become a key and core issue in laryngeal cancer treatment.

With the shift in the medical model towards the “biopsychosocial” model, laryngeal cancer treatment has gradually transitioned from solely pursuing tumor eradication to also considering postoperative functional recovery and quality of life improvement. Previous studies have confirmed that early, systematic voice training can promote compensatory functional recovery through breathing support and vocal efficiency exercises, effectively improving postoperative functional outcomes[7]. Swallowing function training, by strengthening pharyngeal muscles and improving sensory feedback, helps rebuild safe swallowing patterns and improves swallowing function [8]. In recent years, although voice and swallowing training have gradually gained attention in postlaryngeal cancer surgery rehabilitation, their combined application and long-term effects lack systematic research. There is still a relative lack of systematic analysis regarding their clinical efficacy, making it difficult to fully provide evidence-based support for clinical rehabilitation protocol development.

Therefore, this study adopted a retrospective cohort design. After propensity score matching, the long-term differences between combined voice-swallowing rehabilitation and routine care were compared. This study systematically evaluated the impact of this comprehensive protocol on functional recovery, quality of life, and complications in patients after laryngeal cancer surgery, aiming to provide real-world evidence for constructing standardized and efficient postoperative rehabilitation pathways.

Materials and methods

General data

A total of 120 patients who underwent radical surgery for laryngeal cancer in the Department of Head and Neck Surgery of our hospital from June 2019 to June 2023 were retrospectively selected as the study subjects. The inclusion criteria are as follows: (1) pathological diagnosis of laryngeal squamous cell carcinoma; (2) undergoing partial or total laryngectomy for the first time; (3) age 43–73 years; and (4) complete clinical data, follow-up time ≥ 24 months. The exclusion criteria are as follows: (1) pre-existing severe voice or swallowing disorders; (2) accompanied by other malignant tumors or severe heart, lung, or brain diseases; (3) cognitive dysfunction unable to cooperate with assessment and training; and (4) occurrence of severe postoperative complications preventing participation in the rehabilitation plan.

This study was approved by the hospital’s Ethics Committee (ZYY2023037), and informed consent was waived. All patients who underwent radical surgery for laryngeal cancer in our hospital were routinely included in a standardized postoperative functional follow-up protocol, and their assessment data were recorded in the electronic medical record system, thereby ensuring the completeness and reliability of the data collected in the present study. All data processing followed privacy protection principles. To reduce selection bias and confounding bias in the retrospective study, propensity score matching (PSM) was used for 1:1 pair matching analysis. Using age, gender, TNM stage, and surgical method as covariates, the nearest neighbor matching method was applied with a caliper value set at 0.02. After matching, there were 60 cases in each group. The standardized mean differences (SMD) for all covariates were < 0.1, indicating balanced and comparable baseline characteristics between the groups. See Table 1.

Table 1.

Comparison of baseline data between the two groups of patients ([$\overline{\chi }\pm s$]/n,%)

Characteristic	Rehabilitation training group (n = 60)	Routine care group (n = 60)	t/χ2	P
Age (years)	58.73 ± 7.85	60.12 ± 8.24	0.946	0.346
Gender			0.776	0.378
Male	52 (86.67)	55 (91.67)
Female	8 (13.33)	5 (8.33)
BMI	23.52 ± 2.85	23.18 ± 3.02	0.638	0.524
TNM stage			0.358	0.949
Stage I	15 (25.00)	17 (28.33)
Stage II	22 (36.67)	20 (33.33)
Stage III	18 (30.00)	19 (31.67)
Stage IV	5 (8.33)	4 (6.67)
Surgical method			0.315	0.575
Partial laryngectomy	38 (63.33)	35 (58.33)
Total laryngectomy	22 (36.67)	25 (41.67)
Comorbidities	12 (20.00)	10 (16.67)	0.214	0.643
Smoking history	38 (63.33)	40 (66.67)	0.151	0.704
Family support	52 (86.67)	50 (83.33)	0.284	0.597
Postoperative radiotherapy	40 (66.67)	38 (63.33)	0.151	0.704

Grouping basis

The routine care group received standard postoperative care, including vital signs monitoring, wound care, pain management, nutritional support, discharge guidance (such as general dietary advice), and basic verbal hygiene advice (e.g., vocal rest, hydration, and avoidance of throat clearing). Patients who underwent total laryngectomy were informed about the existence of esophageal speech training or TEP guidance; however, no formal instruction or supervised practice sessions were provided. For the rehabilitation training group, in addition to routine postoperative care, a comprehensive intervention comprising graded voice and swallowing rehabilitation, along with nutritional, pain, and psychological support, was initiated within 24–48 h after nasogastric tube removal, with the specific components tailored according to individualized assessment, physician recommendation, or patient preference. Specific content is shown in Table 2.

Table 2.

Review of comprehensive rehabilitation training and intervention content for the rehabilitation training group

Training category	Core training content	Training purpose	Precautions
Voice function rehabilitation training	1. Breathing support training: abdominal breathing exercises to prolong maximum phonation time Training regimen: 15–20 min/session, once daily during hospitalization; after discharge, 20 min/session, 5 sessions/week for months 1–3, 3–4 sessions/week for months 4–6, and 2–3 sessions/week for months 7–12. Advance when phonation ≥ 5 s without vocal fatigue; stop if laryngeal pain or hoarseness > 30 min postexercise occurs	Enhance respiratory support for phonation, prolong sustained phonation time, improve vocal stability	Avoid excessive forceful breathing causing chest tightness/dizziness; precede with 5 min slow relaxation before core exercises
	2. Laryngeal relaxation: gentle massage around the cricoid cartilage in the neck	Relieve laryngeal muscle tension (e.g., cricothyroid, thyroarytenoid), prevent vocal fatigue, reduce vocal cord strain risk	Massage pressure should be pain-free; avoid thyroid area (especially with thyroid disease); follow with slow deep breaths
	3. Vocal training: steady vowel phonation, glide exercises, intensity control, etc	Improve voice quality (reduce hoarseness/breathiness), enhance intensity stability, optimize pitch control	Avoid prolonged high-intensity training; early postoperative patients: initial session ≤ 5 min (avoid excessive vocal cord vibration that may cause edema); patients of chronic voice disorders: initial session ~ 10 min. Stop immediately if laryngeal discomfort occurs
	4. Advanced communication training: intonation/rhythm practice, communication strategies in noisy environments	Improve communication efficiency in complex scenarios (e.g., restaurant, workplace), reduce social barriers due to voice issues	Practice intonation/rhythm in real conversational contexts; noise training intensity should increase gradually from low (< 50 dB)
	5. Alternative voice production: esophageal speech or tracheoesophageal puncture (TEP) valve use guidance for total laryngectomy patients	Help total laryngectomy patients rebuild vocal function, restore basic verbal communication	Conduct under ENT staff guidance; observe for bleeding/infection during initial TEP use; avoid excessive air swallowing causing bloating in esophageal speech training
Swallowing function rehabilitation exercise	1. Basic function exercises: - Orofacial muscle exercises (e.g., cheek puffing, resistance) - Tongue muscle exercises (protrusion, retraction, resistance) - Laryngeal elevation exercises (e.g., Mendelsohn maneuver) - Sensory stimulation (e.g., thermal-tactile stimulation)	Enhance oral chewing/closure, reduce residue Improve tongue control precision, prevent oral residue Increase laryngeal elevation amplitude, protect airway from aspiration Improve swallow reflex sensitivity, shorten reflex delay	Resistance training: increase gradually (e.g., tongue resistance start at 100 g), avoid muscle strain Mendelsohn maneuver: strictly follow therapist demonstration; severe dysphagia patients (e.g., Water Swallow Test grade IV +) should not practice alone Wrap ice stick with gauze for thermal stimulation; avoid high concentration citric acid (e.g., 0.5%) for acid stimulation
	2. Feeding training: adjust food consistency (pureed → soft → solid) following the International Dysphagia Diet Standardisation Initiative (IDDSI) framework, control bolus size/eating pace, use compensatory postures (e.g., chin tuck); liquid consistency progression: moderately thick → mildly thick → thin liquids, with bolus volume increasing stepwise from 3 to 5 mL to 10 mL per swallow	Safely transition food consistencies, reduce aspiration risk, restore oral intake ability	Advancement requires: (i) SSA improvement ≥ 3 points from baseline; (ii) no choking or wet-hoarse voice during 3 consecutive meals at current level; (iii) patient-reported confidence to proceed; all transitions supervised by trained staff Family/medical staff should accompany initial trials of new food consistencies, observe for choking/hoarse voice (aspiration signs); chin tuck: maintain 15°−30° forward head tilt, avoid extending neck
Nutritional & pain management support	1. Nutritional support: develop diet plan based on swallowing assessment; nasogastric tube feeding if swallowing poor	Meet nutritional needs for postoperative tissue repair (e.g., protein, vitamins), reduce malnutrition risk	NG tube feed temperature 38–40 °C (close to body temperature); review swallowing function weekly, transition to oral intake gradually if criteria met
	2. Pain management: use analgesic medications combined with relaxation techniques, cold therapy, etc. to relieve the pain	Alleviate postoperative laryngeal/neck pain, improve rehabilitation compliance	Follow prescribed analgesic dosage; wrap ice pack, avoid surgical incision/skin breakdown
Psychosocial support & health education	1. Health education & family guidance: explain postoperative recovery knowledge and home training precautions to patients/families	Enhance patient/family understanding of rehab process, reduce training deviations	Use simple language, visuals/videos; emphasize home training frequency/contraindications (e.g., stop voice training if laryngeal discomfort)
	2. Psychological support: provide group/individual counseling to relieve anxiety; refer patients with high anxiety scores (SAS > 50) to psychology	Alleviate anxiety/inferiority due to voice/swallowing impairments, enhance rehab confidence	Promote peer support among patients
	3. Social adaptation: guide patients in communication skills practice for workplace/public settings	Reduce social avoidance due to communication barriers, promote social reintegration	Train from familiar (e.g., simulated dialogue with family) to unfamiliar scenarios; for total laryngectomy patients, incorporate alternative voice devices (TEP electrolarynx) in scenario practice

Observation indicators

Voice function was assessed using the Chinese version of the Voice Handicap Index (VHI) (Cronbach’s α = 0.85) [9]. This scale contains 30 items divided into three subdomains: Functional (F), Physical (P), and Emotional (E), with 10 items each. Each item uses a 5-point Likert scale (0 = Never, 4 = Always). The total score ranges from 0 to 120, with higher scores indicating greater perceived voice handicap.

Swallowing function was assessed using the Standardized Swallowing Assessment (SSA) (Cronbach’s α = 0.78) [7]. This scale includes three parts: (1) clinical examination including consciousness, head/trunk control, respiration, lip closure, soft palate movement, laryngeal function, gag reflex, and voluntary cough (score 8–23); (2) patient swallows 5 mL water three times, observing laryngeal movement, repetitive swallowing, stridor during swallow, postswallow laryngeal function, etc. (score 5–11); (3) if no abnormalities above, observe time to swallow 60 ml water, presence of cough, etc. (score 5–12). The minimum total score is 18, and maximum is 46. Higher scores indicate worse swallowing function.

Dysphagia was assessed using the Eating Assessment Tool-10 (EAT-10) (Cronbach’s α = 0.81)[10]. This is a 10-item self-report questionnaire assessing the impact of dysphagia on daily life. Each item uses a 5-point scale (0 = No problem, 4 = Severe problem), total score 40. A total score ≥ 20 indicates risk of dysphagia; higher scores indicate greater risk.

Quality of life was assessed using the Chinese version of the EuroQol five-dimension questionnaire (EQ-5D-3L), comprising a health descriptive system and a visual analog scale (EQ-VAS). The descriptive system covers 5 dimensions: Mobility, Self-Care, Usual Activities, Pain/Discomfort, Anxiety/Depression. Each dimension has 3 levels (1 = No problems, 2 = Some problems, and 3 = Extreme problems). The combination of responses across the 5 dimensions is converted into a single utility index representing overall health state, calculated using the value set model for mainland China, where 1 represents full health. The EQ-VAS is a vertical visual analog scale where 100 represents the best imaginable health state and 0 the worst imaginable health state. Patients mark their overall health on that day on the scale.

Complications included pulmonary infection, pharyngocutaneous fistula, incision bleeding, and malnutrition. All complications were recorded within 1 month postoperatively and defined as follows: Pulmonary infection was defined according to the Chinese Guidelines for the Diagnosis and Treatment of Hospital-Acquired Pneumonia and Ventilator-Associated Pneumonia in Adults (2018) [11]: radiographic evidence (new or progressive pulmonary infiltrates, consolidation, or ground-glass opacities on chest X-ray/CT) plus clinical evidence (meeting at least 2 of the following): temperature > 38.0 °C; new or worsened cough/sputum, or purulent respiratory secretions; WBC count > 10.0 × 10⁹/L or < 4.0 × 10⁹/L, neutrophil percentage > 80%, serum procalcitonin (PCT) ≥ 0.5 ng/mL; and auscultatory findings of moist rales or signs of consolidation. Pharyngocutaneous fistula was diagnosed based on clinical manifestations: saliva or food leakage from the neck incision, local skin redness, tenderness, subcutaneous emphysema, or purulent discharge; aspiration of pus or saliva-containing secretion confirms. Incision bleeding referred to active, continuous bleeding from the incision, or hematoma formation causing neck swelling and airway compression, requiring return to the operating room for exploration and hemostasis. Malnutrition was defined as meeting any of the following: (1) BMI < 18.5 kg/m²; (2) Serum albumin < 35 g/L; and (3) Patient-Generated Subjective Global Assessment (PG-SGA) grade B or C.

Satisfaction was assessed using the Patient Satisfaction Survey Questionnaire (Cronbach’s α = 0.79), including four dimensions: voice function (3 items), swallowing function (4 items), pain management (4 items), and nursing service (5 items) (Cronbach’s α = 0.78). A Likert 5-point scale was used (0 = Very dissatisfied, 4 = Very satisfied), standardized to a score out of 100.

Statistical methods

Data analysis was performed using SPSS 26.0 software. Measurement data are expressed as mean ± standard deviation [$\overline{\chi }\pm s$], and intergroup comparisons were conducted using independent samples t-tests. Count data are expressed as rates (%), and intergroup comparisons were conducted using the χ² test or Fisher’s exact test. Multiple regression models were used, with VHI score, SSA score, and EQ-5D index as dependent variables respectively, grouping as the main independent variable, and adjusting for covariates such as age, gender, TNM stage, surgical method, and postoperative radiotherapy, to analyze the independent impact of rehabilitation training on various functional indicators. A P value < 0.05 was considered statistically significant.

Results

Comparison of VHI scores between the two groups

Preoperatively, there was no significant difference in VHI scores between the two groups (P > 0.05). At all postoperative time points, the VHI scores in the rehabilitation training group were significantly lower than those in the routine care group (P < 0.001), indicating a lesser degree of voice handicap in the rehabilitation training group. See Table 3.

Table 3.

Comparison of postoperative VHI scores between the two groups ([$\overline{\chi }\pm s$], points)

Group	Cases (n)	Preoperative	6 months Postop	12 months Postop	24 months Postop
Rehabilitation training group	60	48.51 ± 6.87	42.15 ± 5.86*	35.25 ± 5.12*#	32.10 ± 4.85*#^
Routine care group	60	49.22 ± 7.10	46.62 ± 7.05*	44.70 ± 6.84*#	41.27 ± 6.15*#^
t		0.579	3.777	8.567	9.069
P		0.557	< 0.001	< 0.001	< 0.001

^*vs. preoperative in the same group, P < 0.05

^#vs. 6 months postop, P < 0.05

^{^}vs. 12 months postop, P < 0.05

Comparison of swallowing function between the two groups

Preoperatively, there was no significant difference in SSA scores between the two groups (P > 0.05). At all postoperative time points, the SSA scores in the rehabilitation training group were significantly lower than those in the routine care group (P < 0.05). The incidence of dysphagia in the rehabilitation training group was significantly lower than that in the routine care group (RR = 0.33, 95% CI: 0.14–0.78, P < 0.05). See Table 4.

Table 4.

Comparison of postoperative swallowing function between the two groups ([$\overline{\chi }\pm s$]/n,%)

Group	n	SSA score (points)				Dysphagia [n (%)]
		Preoperative	6 months Postop	12 months Postop	24 months Postop	6 months Postop	12 months Postop	24 months Postop
Rehabilitation training group	60	24.85 ± 4.24	23.54 ± 4.75*	15.12 ± 3.28*#	12.27 ± 2.86*#^	12 (20.00)	6 (10.00)	4 (6.67)
Routine care group	60	25.12 ± 4.54	25.22 ± 3.40*	19.54 ± 4.25*#	15.51 ± 3.24*#^	25 (41.67)	18 (30.00)	15 (25.00)
t/χ2		0.337	2.228	6.377	5.807	6.604	7.500	7.566
P		0.737	< 0.05	< 0.001	< 0.05	< 0.05	< 0.05	< 0.05

^*vs. preoperative in the same group, P < 0.05

^#vs. 6 months postop, P < 0.05

^{^}vs. 12 months postop, P < 0.05

Comparison of EQ-5D-3L index between the two groups

Preoperatively, there was no significant difference in the EQ-5D-3L index between the two groups (P > 0.05). At all postoperative time points, the EQ-5D-3L index in the rehabilitation training group was significantly higher than that in the routine care group (P < 0.05). By 12 months postoperatively, the score in the rehabilitation training group had reached 0.82 ± 0.09, close to the level of the normal population, indicating that comprehensive rehabilitation training can significantly improve patients’ long-term overall health status. See Table 5.

Table 5.

Comparison of EQ-5D-3L index between the two groups [$\overline{\chi }\pm s$]

Group	n	Preoperative	6 months Postop	12 months Postop	24 months Postop
Rehabilitation training group	60	0.75 ± 0.06	0.72 ± 0.08*	0.82 ± 0.09*#	0.84 ± 0.07*#^
Routine care group	60	0.73 ± 0.15	0.61 ± 0.10*	0.68 ± 0.11*#	0.70 ± 0.09*#^
t		0.959	6.653	7.630	9.511
P		0.340	< 0.001	< 0.001	< 0.001

^*vs. preoperative in the same group, P < 0.05

^#vs. 6 months postop, P < 0.05

^{^}vs. 12 months postop, P < 0.05

Comparison of postoperative complications between the two groups

The total postoperative complication rate in the rehabilitation training group (11.67%) was significantly lower than that in the routine care group (26.67%), and the difference was statistically significant (P < 0.001). The specific occurrence of various complications is shown in Table 6.

Table 6.

Comparison of postoperative complications between the two groups [n (%), cases]

Complication type	n	Pulmonary infection	Pharyngocutaneous fistula	Incision bleeding	Malnutrition	Other complications	Total
Rehabilitation training group	60	3 (5.00)	2 (3.33)	1 (1.67)	1 (1.67)	0 (0.00)	7 (11.67)
Routine care group	60	9 (15.00)	4 (6.67)	2 (3.33)	5 (8.33)	2 (3.33)#	22 (36.67)
χ2		3.333	0.073	0.342	2.807		10.231
P		0.068	0.787	0.559	0.094	0.496*	< 0.001

^*Fisher’s exact test was used

^#Neck seroma (1 case) and delayed wound healing (1 case)

Comparison of postoperative satisfaction between the two groups

As shown in Fig. 1, the satisfaction scores of patients in the rehabilitation training group regarding voice function, swallowing function, pain management, and nursing service were all significantly higher than those in the routine care group (P < 0.05). The total satisfaction scores at 6, 12, and 24 months postoperatively were all significantly higher in the rehabilitation training group than in the routine care group (P < 0.001). See Table 7.

Fig. 1.

Comparison of postoperative satisfaction scores for voice function, swallowing function, pain management, and nursing service between the two groups. * vs. 6 months postop in the same group, P < 0.05; # vs. 12 months postop, P < 0.05

Table 7.

Comparison of total postoperative satisfaction scores between the two groups ([$\overline{\chi }\pm s$], points)

Group	n	6 months Postop	12 months Postop	24 months Postop
Rehabilitation training group	60	79.67 ± 7.85	90.59 ± 7.60*	87.80 ± 7.93*#
Routine care group	60	68.03 ± 8.63	75.17 ± 8.80*	72.21 ± 9.22*#
t		7.729	12.439	13.813
P		< 0.001	< 0.001	< 0.001

^*vs. 6 months postop in the same group, P < 0.05

^#vs. 12 months postop, P < 0.05

Multiple regression analysis

To control for confounding factors, we performed multiple regression analysis with functional indicators at 24 months postoperatively as the dependent variables. The results showed that after adjusting for age, gender, TNM stage, and surgical method, receiving comprehensive rehabilitation training remained an independent protective factor for improved voice function (β = −9.215, P < 0.001), improved swallowing function (β = −3.268, P < 0.001), and improved quality of life (β = 0.138, P < 0.001), and could significantly reduce the incidence of dysphagia (OR = 0.176, P < 0.05) and the overall risk of complications (OR = 0.218, P < 0.001). Higher TNM stage and total laryngectomy were independent risk factors for functional recovery (P < 0.05). Furthermore, radiotherapy was significantly associated with worse voice function (β = 2.15, 95% CI: 0.45 to 3.85, P < 0.05) and swallowing function (β = 1.20, 95% CI: 0.15 to 2.25, P < 0.05), not significantly increasing the risk of dysphagia and complications or reducing quality of life (all P > 0.05). See Table 8.

Table 8.

Results of multiple regression analysis at 24 months postoperatively

Variable	Category	VHI score (β, 95% CI)	SSA score (β, 95% CI)	Dysphagia (OR, 95% CI)	EQ-5D index (β, 95% CI)	Complications (OR, 95% CI)
Gender (male vs female)		−1.50 (−3.55, 0.55)	0.82 (−0.75, 2.39)	1.25 (0.88, 1.78)	−0.02 (−0.07, 0.03)	1.15 (0.80, 1.65)
Age (per 1-year increase)		−0.10 (−0.19, −0.01)*	0.05 (−0.01, 0.11)	1.04 (0.99, 1.09)	−0.002 (−0.01, 0.01)	1.03 (0.98, 1.08)
TNM stage (vs stage I)	Stage II	−1.95 (−4.10, 0.20)	−0.90 (−2.50, 0.70)	1.55 (0.92, 2.61)	−0.03 (−0.08, 0.02)	1.40 (0.82, 2.38)
	Stage II	−3.60 (−5.95, −1.25)**	−1.85 (−3.55, −0.15)*	2.05 (1.16, 3.62)*	−0.05 (−0.10, 0.00)	1.80 (1.02, 3.18)*
	Stage IV	−5.65 (−8.75, −2.55)***	−2.95 (−5.10, −0.80)**	3.40 (1.72, 6.72)***	−0.08 (−0.14, −0.02)*	2.90 (1.45, 5.80)**
Surgical method (total vs partial)		−4.40 (−6.70, −2.10)***	−2.10 (−3.70, −0.50)**	2.70 (1.55, 4.70)***	−0.06 (−0.11, −0.01)*	2.50 (1.42, 4.40)**
Radiotherapy (yes vs. no)		2.15 (0.45, 3.85)*	1.20 (0.15, 2.25)*	1.65 (0.92, 2.96)	−0.03 (−0.07, 0.01)	1.70 (0.95, 3.05)
Rehabilitation training (yes vs no)		−9.215 (−12.07, −6.36)***	−3.268 (−5.08, −1.46)***	0.176 (0.07, 0.44)**	0.138 (0.10, 0.18)***	0.218 (0.09, 0.53)***

*P < 0.05, **P < 0.01, **P < 0.001

Discussion

This retrospective cohort study confirms that a systematic rehabilitation protocol integrating voice and swallowing training can significantly improve the recovery of voice and swallowing function in patients after laryngeal cancer surgery. The sustained benefits at 24 months postoperatively indicate that this rehabilitation strategy promotes physiological compensation and maintains patients' long-term quality of life.

The results showed significantly improved voice function in the rehabilitation training group compared to the routine care group (decreased VHI score), which may stem from neuromuscular functional reorganization and optimization of glottic biomechanics. Studies suggest that breathing support training (e.g., abdominal breathing) enhances subglottic pressure and improves vocal efficiency, thereby partially compensating for postoperative changes in vocal fold mass or structure [12]. Concurrently, laryngeal relaxation techniques alleviate potential compensatory muscle hypertonicity, reducing vocal fatigue. Furthermore, TEP training reconstructs the vibration source via the tracheo-esophageal fistula, leading to significant voice improvement [13]. Cocuzza et al. evaluated voice-related quality of life in postlaryngectomy rehabilitation and reported that tracheoesophageal speech provided significantly better socio-emotional and functional outcomes compared with esophageal speech in total laryngectomy patients. These findings support the value of structured voice rehabilitation in improving communication-related quality of life after laryngectomy [14]. Regarding swallowing function, employing the “sensory stimulation combined with resistance training” (e.g., tongue protrusion against resistance) recommended by the American Academy of Otolaryngology-Head and Neck Surgery guidelines can significantly reduce aspiration rates and improve swallowing function [15]. Orofacial and tongue resistance training directly enhances the lifting force of the hyolaryngeal complex and the propulsive force of the tongue, improving bolus transport [16]. The Mendelsohn maneuver prolongs laryngeal elevation, facilitating the opening of the upper esophageal sphincter [17]. Sensory stimulation (e.g., thermal-acid stimulation) may modulate cortical excitability, lowering the swallowing reflex threshold and thus improving delayed reflex initiation [18]. This multimodal rehabilitation approach effectively promotes sensorimotor cortex reorganization related to swallowing and phonation, which is a key mechanism for long-term functional recovery. The significant improvement in the EQ-5D index and satisfaction scores in the rehabilitation training group compared to the routine care group highlights the profound impact of functional rehabilitation on psychosocial health. Voice and swallowing disorders often lead to social withdrawal and anxiety [19]. Structured rehabilitation breaks this vicious cycle through the following pathways: First, communication strategy training (e.g., adaptive practice in noisy environments) enhances patients’ communication effectiveness in real-world situations, alleviating psychological burden. Research indicates that advanced communication training incorporating scenario simulation (e.g., communication strategies in noise) can significantly boost patients’ confidence in public communication [20]. Second, group counseling and family support enhance patients’ self-efficacy [21]. The postoperative EQ-5D index surpassing preoperative levels at 12 and 24 months indicates recovery from disease-related decline in quality of life, encompassing both functional restoration and psychosocial adaptation. Cocuzza et al. further demonstrated that successful voice rehabilitation is closely linked to quality of life outcomes in laryngectomy patients, with fistula complications serving as critical determinants of rehabilitative prognosis[14]. These observations highlight the importance of structured, multimodal rehabilitation—addressing not only voice but also swallowing, nutrition, and psychological support—in achieving meaningful long-term quality of life improvements. Notably, the utility index in the rehabilitation training group at 24 months postoperatively (0.84 ± 0.07) approached that of the general population, suggesting that comprehensive rehabilitation may achieve a virtuous cycle of overall health through “functional recovery-psychological adaptation-social participation,” fully aligning with the World Health Organization (WHO) three-dimensional “Function-Psychology-Society” rehabilitation model [22].

The total complication rate was significantly lower in the rehabilitation training group (11.67% vs. 26.67%, P < 0.001), particularly the reduction in aspiration pneumonia and malnutrition, which is directly related to improved swallowing safety. Studies show that targeted swallowing exercises for head and neck cancer patients significantly reduce objective signs of aspiration [23]. The swallowing posture training in this study (e.g., chin tuck swallowing at 15–30°) physically elevates the epiglottis, protecting the airway and reducing the risk of pulmonary infection. Concurrently, the dynamic nutritional management strategy based on SSA assessment in this study is consistent with the preventive measures recommended by the Chinese Guidelines for the Diagnosis of Hospital-Acquired Pneumonia in Adults (2018) [11]. Dynamic nutritional support based on swallowing function assessment ensures adequate caloric and protein intake, which is crucial for tissue repair, immune function, and overall recovery, interrupting the vicious cycle of “swallowing disorder→malnutrition→decreased immunity→infection” [24]. Furthermore, optimized pain management (combining pharmacological and non-pharmacological methods) may have indirectly promoted overall rehabilitation outcomes by improving patient compliance [25]. Multiple regression analysis showed that comprehensive rehabilitation training was an independent protective factor for postoperative functional recovery and complication prevention, not dependent on patients’ baseline status, providing crucial “de-biased” evidence for the clinical promotion of this protocol. Secondly, rehabilitation training reduced the risk of long-term postoperative dysphagia by 82.4% (OR = 0.176), indicating that rehabilitation, particularly its swallowing exercise component, directly breaks the vicious cycle of “swallowing disorder→aspiration→pulmonary infection→malnutrition” by improving physiological functions (e.g., enhancing laryngeal elevation force, optimizing swallowing timing). This underscores the importance of long-term adherence to training postoperatively, especially for patients with persistent dysphagia (e.g., SSA score > 18) at 6 months postoperatively, for whom intensified training may further reduce risks. The SSA trajectory significant group difference emerging at 6 months and further widening at 12 months also reflects the cumulative, time-dependent nature of swallowing rehabilitation, with early postoperative homogenization from surgical trauma gradually resolving as training effects accumulate. In subgroup analysis, partial laryngectomy patients demonstrated better functional outcomes compared with total laryngectomy patients. Marcotullio et al. reported 10-year absolute survival rates of 56.25% for cordectomy and 58.82–66.6% for supracricoid partial laryngectomy in T1b glottic tumors, demonstrating that both approaches yield comparable long-term oncological control [26]. Their findings support the consideration of functional outcomes alongside oncological efficacy when selecting surgical approaches for laryngeal cancer.

In conclusion, systematic comprehensive voice and swallowing rehabilitation training is an effective means to promote long-term functional recovery, improve quality of life, and reduce complication risks in patients after laryngeal cancer surgery, providing direct empirical evidence for optimizing postoperative rehabilitation pathways in clinical practice. This study has the inherent limitations of a retrospective design and a single-center sample source. Although significant differences were observed in the complication and subgroup analyses, the relatively small sample size in these secondary analyses warrants cautious interpretation of the results. Furthermore, the inherent heterogeneity in phonatory and swallowing physiology between patients undergoing partial versus total laryngectomy may constrain the direct comparability of absolute scores. Future multicenter randomized controlled trials with larger sample sizes are needed to further validate the generalizability of our findings and to conduct subgroup analyses stratified by surgical type, which would facilitate the development of tailored rehabilitation protocols to better meet the long-term needs of patients after laryngeal cancer surgery.

Author contribution

Jing Liu: Conceptualization; Data curation; Formal analysis; Writing—original draft; Gaiping Fan: Investigation; Methodology; Project administration;Writing—original draft; Wenjuan Li: Resources; Software; Supervision; Writing—review & editing.

Funding

The authors declare that no funds, grants, or other financial support were received from any public, commercial, or not-for-profit entities for the conduct of this research, the preparation of this manuscript, or the decision to submit it for publication. All authors confirm that there are no financial or non-financial conflicts of interest related to this work.

Data availability

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

Declarations

Ethics approval

All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki Declaration and its later amendments or comparable ethical standards. The study was approved by Ethics Committee of Shanxi Province Cancer Hospital (202404030401169).

Consent to participate

Informed consent was obtained from all individual participants included in the study.

Consent for publication

The authors affirm that human research participants provided informed consent for publication of the images in Fig. 1 and Tables 1, 3, 4, 5, and 6.

Competing interests

The authors declare no competing interests.