Abstract
Laryngectomy is a disfiguring, albeit potentially life-saving, procedure with significant impact on the recipients' respiration, swallowing, and social interactions. Voice restoration can be achieved through various means, most commonly electrolaryngeal speech, esophageal speech, and tracheoesophageal speech, although some patients do not master any of these and are restricted to writing notes.
Only recently has attention begun to turn from the objective quality of the speech produced 1 , 2 to the patients' quality of life as influenced by their method of communication. 3 This study examines the quality of life of these patients, both specifically with regards to communication and generally with the European Organization for Research and Treatment of Cancer (EORTC) Quality of Life measurement tool. 4 , 5
METHODS
Seventy-four living patients who had undergone total laryngectomy for laryngeal cancer at the Halifax Infirmary Hospital before March 1996 were identified through the records of the local Laryngectomee Society. Each patient undergoing this procedure is automatically registered in the Society before discharge from the hospital. Each was mailed a questionnaire in a modified Dillman type of protocol. 6 The cover letter was signed by a junior-level otolaryngology resident (M.M.C.), whose name would not have been immediately recognizable to most of the respondents. They were asked to indicate their sex, number of years since laryngectomy, whether radiation therapy had been received, first language, education level, proximity to a speech-language pathologist, major occupation and current work status, method of communication, and how this had been taught. They were then asked to rate their ability to communicate in a number of common situations and to rate their difficulty with several communication problems on a 4-point Likert scale. 7 They were given space to write about what they believed was their greatest difficulty with communication and which, if any, activities they had given up because of voicing difficulties. Quality of life was assessed with the EORTC QLQ-C30 (version 2), which was used with permission from the EORTC Quality of Life Group.
Data were analyzed with an Excel 97 spreadsheet and SPSS 7.0 software.
RESULTS
Sixty-two patients responded, for a return rate of 84%. The mean age of respondents was 65.6 years (range 49–86 years), and there was a 9:1 male predominance. The average number of years since surgery was 3.7 (range 0.2–26). Seventy-five percent had undergone radiation therapy, 58% of these before surgery and 40% after (1 person had radiation both before and after surgery). Ninety-three percent considered English to be their first language. Forty-three percent had an education level of grade 8 or less, 45% had at least some high school, and 12% had at least some college or university education. Seven percent were working full-time, 2% were working part-time, and 2% were looking for work. Sixty-five percent were retired. Twenty-five percent were off work as a result of disability. There was a wide range of occupations.
Ratings for ability to communicate in various situations by laryngectomy patients
70141-0-table1.png)
Scale: 1, not at all; 2, a little; 3, quite a bit; 4, always.
∗Includes respondents using more than 1 voicing method (combination of electrolarynx and either esophageal or TEP voice prosthesis). Not all respondents indicated answers to all parts of the survey.
Seventy-three percent lived within a 30-minute drive to a speech-language pathologist, and 81% had worked with a speech-language pathologist for 6 months or less to master their current method of vocalizing. During their therapy, 67% of respondents visited every week, and 20% visited every other week.
Respondents were asked to indicate which methods of speech they had used and which they were currently using. Eighty-six percent (N = 53) had used electrolaryngeal speech, 32% (N = 20) had used esophageal speech, and 15% (N = 9) had used tracheoesophageal speech. Currently, 57% (N = 34) were using only the electrolarynx, 19% (N = 11) esophageal speech, and 8.5% (N = 5) tracheoesophageal speech. One respondent wrote notes without using any method of vocalizing. Eight respondents (13.5%) used a combination of electrolarynx and esophageal or tracheoesophageal speech. Three patients did not indicate their current method.
These 3 groups (electrolaryngeal, esophageal, and tracheoesophageal speakers) were comparable with respect to age, sex, first language, education level, whether they had undergone radiation therapy, and number of years since laryngectomy. There was no difference in respondent reports for how far they lived from speech therapy facilities or how long or how often they worked with a speech-language pathologist to learn their current method of vocalizing.
Mean ratings for respondents' ability to communicate in various situations are shown in Table 1. Comparison among the 3 groups with a Kruskal-Wallis test showed no significant differences. Electrolarynx users were compared with esophageal speech users (because these groups were the two largest) with a Mann-Whitney U test. Only one situation was different between these groups; the esophageal speech users felt more capable of expressing emotions than did the electrolaryngeal speakers (P = 0.032). The esophageal speakers tended to feel more able to communicate in most of the listed situations. Pearson's correlation coefficient among these parameters was highly significant in all cases except for ability to sing, suggesting a relationship between these areas.
Respondents were asked to rate the following problems on a 4-point Likert scale: embarrassment speaking to friends or family; embarrassment speaking to strangers; effort required to speak; discomfort in throat, neck, and mouth; inability to produce a sound; and need to give up activities because of communication difficulties. There were no significant differences between the 3 groups or between the esophageal and electrolaryngeal speakers. Eleven respondents listed activities that they had given up (Table 2; some had given up more than 1 activity).
Activities given up by laryngectomy patients because of communication difficulties (N = 11)
70141-0-table2.png)
A post hoc power calculation indicated that a clinically significant difference of 1 point on the Likert scale would be revealed statistically with a group of this size with approximately 70% certainty. 8
Respondents were asked to describe their greatest problem with their current method of communication, and 48 did so. These are listed in Table 3. The most frequently cited problem overall was difficulty being heard in a noisy environment.
The EORTC QLQ-C30 includes 6 functional scales: physical functioning (moving and self-care), role functioning (work and leisure pursuits), emotional functioning (anxiety and depression), cognitive functioning (concentrating and remembering), social functioning (family and social activities), and global health status/quality of life, and average ratings for the entire group are shown in Table 4. Comparisons between the 3 groups were made with a Kruskal-Wallis test, and there were no significant differences between them.
DISCUSSION
This study examines quality of life on a variety of dimensions. Respondents were asked to rate their communication ability in 14 different situations, and these ratings were highly uniform. The EORTC QLQ-C30 assesses general quality of life through several scales of function, without regard for communication ability, and was used to compare general health and functional status among our groups. Our groups were comparable in demographic characteristics, and there were few differences in any of these parameters. Our groups are relatively small but of sufficient size to reveal a clinically important difference with reasonable certainty. Regardless of the method of communication used by these laryngectomy patients, quality of life as assessed by these scales is comparable. Despite relatively high levels of functioning, these patients regard their overall quality of life as moderately low.
Greatest problem with communication as cited by laryngectomy patients
70141-0-table3.png)
Tracheoesophageal speech is thought to be more intelligible than either esophageal or electrolaryngeal speech. 1 , 2 , 9 , 10 A previous study of patient satisfaction with their method of communication and quality of life suggested that those using tracheoesophageal speech were more satisfied than those using other methods. 3 This study used a patient group smaller than ours but with a higher proportion of tracheoesophageal speakers and nonvocalizing writers. Use of analysis of variance to compare these 4 groups (including electrolaryngeal and esophageal speakers) on 5 parameters revealed significant differences for 4 of them. The data were graphed, showing that tracheoesophageal speakers rated these parameters highest, and writers rated them lowest, so a significant analysis of variance may be revealing a difference between the writers and the vocalizers in general.
In some centers, primary tracheoesophageal puncture (TEP) is performed, abrogating the need for a second hospitalization for this procedure. Risks of TEP include enlargement of or recurrent/persistent infection of the puncture site, aspiration pneumonia, stomal stenosis, and death. 11 – 11 The complication rate is high, estimated at 25% 12 to 40%. 13 In our group of TEP patients, there were minor complications in all but 1; complications ranged from leakage, local pain, stenosis requiring dilation and ultimately repeated puncture, and incorrect placement. One patient who had his TEP performed primarily elsewhere requested closure of the fistula, which proved a slow process. We know that not all patients who receive TEP succeed in voicing in this manner. In our group 88% succeeded in voicing, which is higher than the rate reported for a group of primary TEP patients (70%). 14 Our 1 nonvoicer succeeded in learning esophageal speech. Closure of unsuccessful TEPs may require local flaps 15 or may be delayed, as in our patient. In view of the suggestion that quality of life may not be different for electrolaryngeal or tracheoesophageal speakers, patients may benefit from detailed preoperative counseling, after which they can make a better informed decision regarding postlaryngectomy voicing method. The patients in this study did not receive primary TEP, and all were given access to an electrolarynx on loan immediately after surgery. Those who used the other 2 voicing methods had made this decision after a trial of electrolaryngeal speech. Patients happy with communication as afforded by the electrolarynx would likely not pursue TEP or esophageal speech.
Average ratings for EORTC QLQ-C30 functional scales
70141-0-table4.png)
Ratings are on a scale ranging from 0 to 100 where higher values represent a higher level of functioning.
Includes respondents currently using more than 1 method of voicing.
Estimated costs for each voicing system in Nova Scotia in 1997
70141-0-table5.png)
∗Costs taken from the MSI Fee Guide 1997, Nova Scotia's allowable charges for these procedures.
A cost-benefit analysis was not part of this study, but we speculate that electrolaryngeal voice rehabilitation may be least costly to our health care system (Table 5) and in terms of patient morbidity. Our results suggest that there may be little benefit to pursuing more complicated methods of voice rehabilitation, despite the objectively superior vocalization produced by tracheoesophageal prostheses.
Patients' most common problem is that they cannot be heard in environments with high levels of ambient noise. This may be related to several factors, the first of which is inability to produce sufficient volume to be heard. The acoustic signature in alaryngeal voicing, especially electrolaryngeal speech, is also very different from that of normal speech. In normal speech, the consonants (especially the stops) are produced by imposing a complex upper vocal tract filter function on the low-amplitude bursts and aspiration noise during carefully timed brief periods of nonvoicing. In electrolaryngeal speech, those voiceless periods are either missing or poorly timed. This means that the acoustic signature for the consonants residing in the aspiration noise is partially masked by the electrolaryngeal speech in quiet, but makes it highly vulnerable to further masking by environmental noise, especially if that noise has significant energy in the range of the acoustic signature of the consonants. Devising a better electrolarynx has been a goal since their invention; perhaps attention to this issue will make this type of speech more intelligible.
CONCLUSION
These results suggest that the method of voicing as chosen by laryngectomy patients does not appear to affect their overall quality of life, although some weaknesses of these alaryngeal voicing systems are apparent. Because there is morbidity associated with TEPs and costs associated with all voicing systems, patients may benefit most from detailed counseling about communication methods, and being free to choose a system compatible with their lifestyles.
We are very grateful to the Laryngectomee Society of Nova Scotia for their support of this project.