A 74-year old, retired, white male with treatment refractory dysphagia following chemo-radiation therapy for treatment of a T3N1M0 base of tongue squamous cell carcinoma, enrolled in an experimental dysphagia treatment research program at a university-affiliated speech and hearing clinic. Nine months prior to initiating this dysphagia therapy program he completed 40 doses of radiation therapy (7200 cGray to the primary tumor site and 5040 cGray to a secondary neck lymphnode) in conjunction with cisplatin chemotherapy. Following chemoradiation he completed two separate courses of dysphagia therapy. At time of enrollment in this dysphagia treatment program, he received nearly all nutrition and hydration via percutaneous gastrostomy tube. Daily oral intake was limited to approximately two ounces of pureed foods. Vocally, he complained of fatigue, pitch breaks, difficulty being understood on the telephone, and reduced singing ability.
He completed 15 days of an experimental swallowing exercise program supported with transcutaneous neuromuscular electrical stimulation (NMES) [
5]. Swallowing exercise involved performing the effortful swallow technique while ingesting a progressive hierarchy of liquids and solids. NMES was delivered via the VitalStim
® NMES unit (Chattanooga Group, Hixson, TN). The VitalStim device uses two pairs of electrodes placed vertically along the midline of the anterior neck to supply the electro-stimulation.
On treatment day five, perceptual changes in the subject's voice were noted, prompting daily acoustic measurements of three vocal parameters beginning on treatment day six. Acoustic measurements of maximum phonation time (MPT), pitch range (highest and lowest attainable pitch), and habitual pitch while reading were obtained using the VisiPitch IV (KayPentax: Lincoln Park, NJ). Each parameter was measured three times before and after all therapy sessions and during the three follow-up sessions. Mean and standard deviation of the three trials were calculated for each task. A total of 10 treatment and three follow-up sessions were recorded.
Pre-, post, and follow-up swallowing function was evaluated via standardized clinical, endoscopic, and videofluoroscopic evaluations including completion of the Mann Assessment of Swallowing Ability (MASA) [
6] and the Functional Oral Intake Scale (FOIS) [
7], as well as self perception of swallow function measured via bisection of a 100 mm visual analog scale (VAS). Each voice parameter was analyzed using repeated measures ANOVA with Bonferroni correction for multiple comparisons. When significant interaction or main effects were identified for any voice parameter, Tukey pairwise post-hoc analysis was employed. Of primary focus were comparisons made between baseline vocal measurements (session 6) and end of treatment measurements (session 15).
Endoscopic, perceptual, and instrumental vocal results supported the existence of a cross-system interaction between dysphagia rehabilitation and improved laryngeal function. Baseline endoscopic examination revealed significant supraglottic compression and a glottal gap during phonation, while post therapy examination revealed decreased supraglottic compression and glottal closure during phonation. Perceptually, the subject reported that his voice was louder, that he was able to sing in church again, and that others reported being able to better understand him on the telephone. Instrumentally, significant between session main effects were observed in both MPT (F
(9,20) = 7.993, p < .001) and highest attainable pitch (F
(9,20) = 3.620, p = .008). Significant within session interaction effects were observed in habitual pitch [(F
(9,20) = 14.215, p < .001)]. No significant effects were observed for lowest attainable pitch (F
(9,20) = 0.949, p = .513). Mean scores for those voice parameters demonstrating significant change within or across treatment sessions are presented graphically in Figures
1,
2,
3. Scores for baseline, post therapy, and follow up measures for each vocal parameter are shown in Table
1.
| Figure 1Maximum Phonation Time. The longest length of continuous phonation produced while sustaining/a/at a comfortable pitch and volume. Each Pre-stim and Post-stim data point represents the mean of three trials of the task. The mean data point represents the (more ...) |
| Figure 2Highest Attainable Pitch. The highest attainable pitch produced without straining, while sustaining/a/in an upward glissando. Each Pre-stim and Post-stim data point represents the mean of three trials of the task. The mean data point represents the mean (more ...) |
| Figure 3Habitual Pitch. Subject's habitual pitch produced while reading the first sentence of the Rainbow Passage. Each Pre-stim and Post-stim data point represents the mean of three trials of the task. The mean data point represents the mean performance during (more ...) |
| Table 1 Average Performance for Each Vocal Parameter at Baseline, Post Treatment and During the Three Follow-up Sessions. |
Since comparisons between the baseline measurements (session 6) and the end of treatment measurements (session 15) were of primary focus, post hoc analysis was completed for MPT, highest attainable pitch, and habitual pitch. Analysis revealed that mean MPT increased significantly between session 6 and session 11 (p = .003) with this increase maintained through session 15 and the three follow-up sessions; mean highest attainable pitch increased significantly between treatment sessions 6 and 11 (p = .004), and demonstrated a trend toward significance between sessions 6 and 15 (p = .017) with the increase maintained through the one-week and one-month follow-up sessions, but not at six months post treatment. Furthermore, the increase in habitual pitch noted in session 11 was significant when compared to both baseline session 6 (p < .001) and last treatment session 15 (p < .001), but no significant change was noted between sessions 6 and 15 (p = 1.00).
This patient demonstrated improvement in all swallowing measures immediately after completion of the therapy program (Table
2). Total oral intake increased from a few bites of pureed or soft food each day, to three to five daily meals of solid food including waffles, sandwiches, and steak, while tube feeding simultaneously decreased from 8–9 cans daily, to 2–3 cans daily. MASA score increased 19 points; an increase of 10 points has been shown to be clinically significant [
6]. FOIS increased one level reflecting the increase in oral diet and simultaneous decrease in tube feeding. Lastly, the subject self-reported a 77 mm increase in self-perception of swallowing ability on the VAS. Swallowing improvements observed immediately post therapy were not maintained through the 6-month follow-up due to post-radiation changes, specifically radionecrosis of his mandible and complications affecting his esophagus. Consequently, by the sixth month post therapy, nutrition was delivered via full enteral feeding; oral intake was limited to occasional sips of liquid.