Introduction

Laryngeal carcinoma is one of the most common malignant tumors of the head and neck region. In many parts of the world, such as the United States, Canada, and England for example, laryngeal carcinoma occurs in the glottic region in more than 50 % of cases [1]. Glottic carcinoma usually presents early with hoarseness of voice, and because the vocal cords have little lymphatic drainage system, the risk of lymph node metastasis is exceedingly rare. Early glottic carcinoma is a highly curable disease, and many modalities of treatment are available (radiotherapy, partial laryngectomy, and transoral laser microsurgery) [2–5]. In function preserving surgical interventions, normal tissues that are not infiltrated by tumor are preserved. Many studies have revealed that oncological outcomes after surgical management of T2 glottic carcinoma are better than those after radiotherapy.

In all of these modalities, both oncological and functional outcomes must be considered. Functional outcomes after management of laryngeal carcinoma include: voice, swallowing, and perioperative morbidity. Other factors that are also taken into consideration during treatment of early laryngeal carcinoma are quality of life and cost of each treatment modality. Swallowing problems, and in particular aspiration, are very common after different therapeutic approaches especially surgical interventions.

They also have a significant impact on recovery and regaining a normal daily life style after treatment.

Although occurring after almost all types of surgical approaches, the degree and duration of swallowing problems are different among these approaches. Few studies have prospectively analyzed swallowing functional problems after different types of surgical management. In this study, we compare swallowing functional outcomes, using both objective and subjective methods, between two groups of patients with T2 glottic carcinoma: one undergoing open vertical hemilaryngectomy, and the other undergoing transoral laser microsurgery (TLM).

Materials and methods

This prospective study was performed on patients with early laryngeal carcinoma, admitted and treated at the Otorhinolaryngology, Head and Neck Surgery Department, Alexandria Main University Hospital, Egypt and the University Hospital of Brescia, Italy. The clinical study was performed during the period from September 2012 to September 2015. An informed consent has been obtained from all patients according to the Ethics Committee of the institution.

Inclusion criteria for enrolment in the study were: diagnosis of T2 glottic tumor according to the 7th Edition of the Union for International Cancer Control–American Joint Committee on Cancer (UICC–AJCC) TNM staging system [6], without distant metastasis at the time of diagnosis, without preoperative or postoperative radiotherapy or chemotherapy, and with good pulmonary function. Exclusion criteria were diagnosis of other than T2 tumors (T1, T3, or T4), poor pulmonary function, previous oncological management, or the presence of any swallowing problems before surgery.

Preoperative assessment

Endoscopic evaluation and CT scan of patients was done to assess vocal cord mobility and for proper evaluation of the extent of tumor. Pulmonary function assessment was done using spirometry (FEV1 was at least 60 %). Swallowing function assessment was done using barium swallow. Examination under anesthesia, accurate assessment of tumor extent, mobility of the arytenoid cartilage, and biopsy were performed in all cases.

If laser surgery was not contraindicated due to difficult exposure, patients were randomized to either group. At the end of the study, the number of cases was limited, so matching of all possible confounding factors has been confirmed and some cases have been excluded to ensure balance between both groups. These factors include age, sex, intellectual status, extent of the tumor, and associated medical co-morbidities.

At the end of the study, 40 patients were included in the laser group, and 16 patients were included in the open vertical hemilaryngectomy group (4 with difficult exposure).

Operative procedures

After proper investigations and staging, patients underwent either: laser cordectomy type V, or open vertical hemilaryngectomy.

Laser cordectomy type V according to the European Laryngological Society (ELS) classification [7] is extended cordectomy with four subtypes: type Va is resection extending to the other cord, type Vb is resection extending to the vocal process, type Vc is resection extending to the supraglottic region, and type Vd is resection extending to the subglottic region. Laser tumor resection was done using either an en bloc technique if possible, or a piece-meal technique.

Open vertical hemilaryngectomy was done by performing tracheostomy, followed by an open surgical approach to the thyroid cartilage and dissection of thyroid perichondrium, and access to the laryngeal lumen followed by resection of the involved part with the adjoining cartilage and reconstruction using a sternohyoid muscle flap.

We collected the following data in all cases: blood loss, need and duration of ICU admission, duration of tracheostomy tube and nasogastric tube if needed, duration of hospital stay, and postoperative complications.

After surgery all patients had postoperative swallowing rehabilitation to help recovery of normal swallowing function. This was done according to Logemann [8] recommendations using compensatory procedures, therapeutic procedures, and changing food consistency.

Swallowing functional analysis

In order to compare swallowing functional outcomes after either laser or open surgical intervention, both subjective and objective evaluation techniques were used.

Subjective evaluation of the effects of swallowing problems on the life style of patients was done using a questionnaire about symptoms of swallowing difficulties and the MDADI (MD. Anderson Dysphagia Inventory) scoring system.

The MDADI scoring system is a validated dysphagia-specific quality of life questionnaire composed of 20 items, including one global question evaluating the global impact of swallowing problems on overall daily life, and three subscales (emotional, functional, and physical). Based on the answers given by patients, a score is calculated up to 100 (high functioning). This scoring was done 1 week after recovery of normal swallowing functions, and 3 months after the operation. The mean of overall score and the different subscale scores were calculated for each patient.

A team of otolaryngologists, speech pathologists and radiologists assessed swallowing outcomes using clinical assessment, fiberoptic endoscopic evaluation of swallowing (FEES) while swallowing colored fluid and semisolid food, and videofluoroscopy (VFS) using different consistencies of barium swallow (thin liquid barium, thick liquid barium, and barium paste) on days 1, 5, 10, 15, and 90 after the operation; the degree of aspiration was graded from 0 to 3, according to the consistency of aspirated material (Table 1).

Table 1 Grading of aspiration
Full size table

Precise evaluation of swallowing function of patients was objectively done using various measures obtained during VFS according to the technique described by Leonard et al. [9]: airway closure duration [10], opening of the upper esophageal sphincter (size and duration) [11], pharyngeal function measures such as transit time and pharyngeal constriction ratio [12], and hyoid displacement distance [11]. All these measures were obtained from lateral views, and the timing information at 0.01-s intervals was recorded.

Statistical analysis

The SPSS package was used for statistical analysis. Statistical analysis was performed with the Mann–Whitney U test for comparison of numerical variables (MDADI scores, degree of aspiration, airway closure duration, upper esophageal sphincter size and duration of opening, pharyngeal transit time, pharyngeal constriction ratio, hyoid displacement distance, duration of aspiration and duration of hospital admission).

The Pearson Chi-square test was used to compare categorical variables (percentage of incomplete airway closure, ICU admission, nasogastric tube insertion, and tracheostomy tube insertion). A p value ≤0.05 was considered statistically significant for both tests.

Results

Demographic data

The study included 56 patients (53 males and 3 females) with a mean age of 56 years (range 40–70).

In the beginning of the study, we had 63 patients diagnosed as T2 glottic carcinoma; two patients were excluded due to poor pulmonary function, two patients for the presence of aspiration before surgery and three patients for age ≥75 years. 53 patients presented with conventional squamous cell carcinoma and three with verrucous carcinoma.

Of the remaining 56 patients 40 patients underwent laser cordectomy (type V), and 16 underwent vertical hemilaryngectomy.

In the laser group all patients underwent different subtypes of type V cordectomies. All cases in the vertical hemilaryngectomy group underwent classical resection and extended to include part of the arytenoid cartilage in two cases and part of the epiglottis in another two cases. The extents of resections in both groups are summarized in Table 2.

Table 2 Distribution of cases according to the extent of resection in both vertical and laser groups and the mean duration of aspiration in each subgroup
Full size table

In the laser group, the ventricular fold was removed in all cases for better exposure.

In the vertical hemilaryngectomy group, all patients were admitted to the ICU in the postoperative period, had tracheostomy, 15 out of 16 patients needed nasogastric tube. The mean duration of hospital stay was 13 days ranging from 5 to 25 days.

In the laser group, only three patients were admitted to the ICU, and none had a tracheostomy, while 5 of 40 needed a nasogastric tube. The mean duration of hospital stay was 3.3 days (range 1–10 days).

All these variables were significantly lower in the laser group than in the vertical hemilaryngectomy group (p < 0.001). This was done using Pearson Chi-square test for comparison of nasogastric tube insertion and Fisher’s exact test for comparison of tracheostomy tube insertion and ICU admission. Aspiration pneumonia occurred only in one case in the vertical hemilaryngectomy group. No other postoperative complications occurred in either group.

Regarding the evaluation of swallowing outcomes, both FEES and VFS showed similar results regarding the presence and degree of aspiration, and other swallowing problems. In days 1, 5, and 10 after the operation, the degree of aspiration was significantly lower (p < 0.001) in the laser group than in the vertical hemilaryngectomy group, while on day 15 the degree of aspiration was lower in the laser group but did not reach statistical significance (Table 3).

Table 3 Mean and median of grading of aspiration in the vertical and laser groups after the operation using VFS and FEES
Full size table

The aspiration was evident both during deglutition and after deglutition. Of the cases that showed aspiration, it happened during deglutition in 58 % of the laser group and 56 % of the open surgery group. After deglutition it was present in 42 % of the laser group and 44 % of the open surgery group.

The duration of aspiration was also significantly shorter in the laser group than in the open surgery group (p < 0.001). The mean duration of aspiration was 3.4 days in the laser group and 11.4 days in the vertical group.

The detailed examination of swallowing function using different measures obtained from lateral view of VFS (Table 4) showed that the pharyngeal constriction ratio was significantly elevated in the vertical hemilaryngectomy group in days 1, 5, 10, and 15, but not 3 months after operation. Furthermore, the percentage of cases with incomplete airway closure was significantly higher in the vertical hemilaryngectomy group in days 1, 5, and 10 after operation. Regarding the pharyngeal transit time, although it was longer in the vertical hemilaryngectomy group, this difference did not reach statistical significance at any time the assessment was made. No other statistically significant difference was found between both groups using VFS.

Table 4 Detailed swallowing assessment results using VFS lateral view in both groups
Full size table

Using the MDADI scoring system for subjective evaluation of swallowing (Table 5) after 1 week of recovery of normal swallowing function, there was a significantly better score in the laser group in the global, mean score and all subscales of the questionnaire (p < 0.001). After 3 months, it did not show a significant difference between groups (p > 0.470).

Table 5 Mean and median MDADI scores of the vertical and laser group 1 week and 3 months after operation and the p value of the difference
Full size table

Discussion

In this study swallowing outcome was significantly better in the laser group than in the open surgery group. This was evident in the early post-operative period when swallowing was evaluated using FEES and VFS and subjectively using MDADI score. Indeed, the p value was ≤0.001 in the degree of aspiration in days 1, 5, and 10 and in the MDADI score at 1 week after recovery of swallowing function. The difference was not evident 3 months after surgery with all modalities of evaluation of swallowing. Mostly, this is due to the complete recovery of swallowing functions after both approaches.

Concerning other variables of comparison, the statistically significant differences show that laser cordectomy has a less morbid postoperative course than its counterpart, and confirm that this technique reduces hospital stay and costs.

The results for swallowing confirm that endoscopic resection shows faster recovery than the open surgical group with only 12.5 % of patients requiring a nasogastric tube for feeding after the operation, which was removed after less than 5 days (in most cases). In the open surgery group, the nasogastric tube was inserted for at least 7 days in almost all cases.

Regarding the objective evaluation of swallowing using VFS, the early significant difference in the pharyngeal constriction ratio shows that open approach affects the pharyngeal motor function more significantly than endolaryngeal approach. Incomplete laryngeal closure was also more prevalent in vertical group than in the laser group in the first 10 days after operation. These findings are in accordance with findings from other studies [13, 14], namely that laser resection of early glottic carcinoma has better swallowing function outcome in the postoperative period.

These differences in swallowing outcomes may be explained by the fact that in laser cordectomy the amount of resected tissue is tailored according to the extent of tumor, leading to more preservation of normal uninvolved tissues. Also, in laser resection there is no disruption of laryngeal skeleton integrity to gain access to the laryngeal lumen, which minimizes resulting edema and subsequent need for tracheostomy. The findings observed during VFS may also help better understanding of these differences. As the measures in the study show, after vertical hemilaryngectomy the protective action of vocal cords is more disturbed than after laser resection, and the pharyngeal motor function that assist food transport to the esophagus is also more affected in the vertical group.

In addition, the presence of tracheostomy prevents normal laryngeal elevation which is one of the protective physiological mechanisms against aspiration. The probability of nerve injury (which leads to loss of sensation) is much lower in laser resection than in an open surgical approach [15].

The longer duration of aspiration in the vertical group was associated with greater impact on the life style of patients, which was evident in the significant difference of the first MDADI score, although it was done 1 week after complete recovery of swallowing function in both groups.

The rapid, complete recovery of swallowing after both approaches may be explained by leaving the hyoid bone, epiglottis, and arytenoids in place in patients undergoing classical vertical hemilaryngectomy and laser resection. However, if the resection is extended to include the arytenoid cartilage or part of the supraglottic region, important components of airway protection are affected, and the risk of aspiration increases [16].

This complete recovery explains the similarity in MDADI score after 3 months in both groups, when all patients regained normal swallowing function.

Our overall results seem to confirm what was found by other studies [13, 14], that laser resection of early glottic carcinoma has less morbid outcomes in the postoperative period. This was also evident in other variables such as length of hospital stay and the need for tracheostomy. These results broaden the indications for conservative laryngeal surgery to include patients with a more advanced age, which is considered a contraindication for partial laryngeal surgery.

Studying functional outcomes after surgical treatment of T2 glottic carcinoma is important, since it is reported in the literature that surgical management is oncologically better than radiotherapy. RT achieves local control in 64–87 % with laryngeal preservation rates of 75–87 %. T2 tumors with impaired vocal cord mobility have even a worse prognosis and lower oncological outcomes after radiotherapy with local failure rates up to 30 % [17, 18]. On the other hand, open surgical treatment oncologic results are reported as having a local control rate of 90–98 % with a 93–98 % laryngeal preservation rate [19, 20]. Another problem after radiotherapy is that a second course of radiation for a recurrence or for a second tumor cannot be offered. Moreover, tumors may not be amenable to conservative surgery after previous RT [21–24].

It should be kept in mind that severe dysphagia often results from radiation induced fibrosis, which may be exacerbated by the addition of chemotherapy [25, 26].

Many other studies [27–29] have also shown similar results in the management of supraglottic laryngeal tumors, where endoscopic management led to a significantly lower incidence and degree of aspiration. It also showed significantly lower rates of tracheostomy, nasogastric tube, and duration of hospital stay.

A special deglutition rehabilitation management program is recommended for rapid recovery after partial laryngeal surgery. This program should include frequent swallowing assessments, which allow tailoring rehabilitation techniques to the progress of each patient.

In this study, we used multiple techniques recommended by Logemann [8]: postural techniques (chin down and head rotation to the damaged side), changing food consistency (avoidance of fluids in the early postoperative period), and therapy procedures (swallow maneuvers as supraglottic and super supraglottic swallowing). Other studies have also shown the importance and effectiveness of these rehabilitation maneuvers and have recommended their use.

FEES and VFS [30, 31] are the two widely used techniques for proper evaluation of swallowing. Because they differ in many aspects, it is better to use both to properly evaluate swallowing. FEES is a bed side test that does not include exposure to radiation, and also helps to evaluate the anatomical and functional aspects of recovery, but may miss minor degree of aspiration as the evaluation is mainly made before and after not during deglutition.

On the other hand, VFS is a very sensitive method that can detect and document even a minimal degree of aspiration in all phases of deglutition. Both methods can detect aspiration for fluids or solids, and can show the improvement after rehabilitation program.

Objective swallowing analysis using measures obtained during VFS is a very accurate and reproducible tool for evaluation of swallowing function. These measures have been used many times by other studies for detailed assessment of swallowing in different medical conditions [32, 33]. Only few studies have used these measures to examine swallowing after conservative laryngeal surgery. Precise evaluation of progression of swallowing recovery after supracricoid laryngectomy has been done using various VFS measures [34, 35].

In the literature, there are different grading systems for dysphagia and aspiration [36–38]. None of them have taken into consideration the consistency of the aspirated material; whether the patient aspirates fluids, solid food, or both. In cases of partial laryngeal surgery, this point is important because it shows the progress of recovery, and guides the rehabilitation. Although Bergamini et al. [39] proposed a dysphagia score that depends on the symptoms experienced by patients in relation to different consistencies of food, this scoring system is only subjective.

Accordingly, we used a new grading system for aspiration based on the consistency of the aspirated material as shown by VFS and FEES. Early in the postoperative period after vertical hemilaryngectomy the patient usually aspirates both fluids and solids even in the saliva. If this is the condition, it is advisable to keep the nasogastric tube.

In this study the aspiration was observed both during and after deglutition. This is different from what was reported by other studies on swallowing after resection of supraglottic tumors which was mainly after deglutition [40, 41]. This difference may be explained by the fact that the resection of vocal cords in the management of glottic tumors leads to loss of airway closure during swallowing with ensuring aspiration.

The study of voice functional outcome after surgical management of early glottic carcinoma was done in our institute and revealed better functional voice outcome in cases managed by an open surgical approach than in cases managed endoscopically (especially when both cords were removed by laser) [42].

Herein, we did not compare the oncologic results between the two groups of patients because of the insufficient follow-up. Additionally, the small number of cases did not allow us to perform other parametric tests. This was due to strict inclusion criteria (only T2 glottic tumors without preoperative and postoperative radiotherapy or chemotherapy).

Conclusion

This study revealed faster swallowing recovery, better swallowing outcomes, and reduced perioperative morbidity after management of T2 glottic carcinoma using TLM than after vertical hemilaryngectomy. Therefore, TLM should be considered the modality of choice in early glottic carcinoma, especially in patients with co-morbidities.

Because VFS is the definitive technique for anatomical and physiological study of swallowing, its usage after conservative laryngeal surgery is very useful for the detection and follow-up of postoperative aspiration and swallowing dysfunction. It is also helpful for the planning of rehabilitation programs and swallowing training after surgery.