The Laryngoscope
C 2011 The American Laryngological, V
Rhinological and Otological Society, Inc.
Determining the Efficacy and Cost-Effectiveness of the ActiValve: Results of a Long-Term Prospective Trial Donna J. Graville, PhD; Andrew D. Palmer, MS; Peter E. Andersen, MD; James I. Cohen, MD, PhD Objectives/Hypothesis: To investigate 1) whether the Provox ActiValve results in increased device-life in individuals with below average device-life, 2) whether it is cost-effective, and 3) whether it has any impact on voice-related quality of life. Study Design: Prospective study. Methods: Individuals who experienced below-average tracheoesophageal prosthesis (TEP) life were studied. Results: Individuals with persistent below-average TEP life were enrolled in the study and underwent periodic re-evaluation. The majority (73%) experienced significant improvement as a result of use of the device. Those who continued to wear the device were followed for an average of 30.45 months (range, 14.70–43.49 months) and wore a total of 31 devices over this time. They demonstrated an average increase in device-life of more than 500%, going from an average of 1.93 months with a traditional indwelling device to 10.30 months with the ActiValve. The majority of individuals found that voicing with the ActiValve was either the same or better than with their previous indwelling TEP. Voice-related quality of life was not significantly different from that of a group of controls. Overall satisfaction with the device was high, and the majority would have chosen the ActiValve in the future. Overall, there were estimated to be cost savings to third-party payers through use of the ActiValve in this population. Conclusions: The ActiValve is effective in increasing device-life in selected patients who have failed conservative measures. Our protocol for use of the device requires individuals to meet several usage criteria before initial placement and to return for periodic monitoring. Key Words: Candidiasis, costs and cost analysis, laryngeal neoplasms, laryngectomy, larynx, artificial prosthesis failure, prosthetic voice rehabilitation, treatment outcome, voice disorders. Level of Evidence: 2c. Laryngoscope, 121:769–776, 2011
INTRODUCTION Since being introduced in the 1970s, the tracheoesophageal prosthesis (TEP) has become an extremely popular method of speech rehabilitation after laryngectomy,1 and success rates of over 90% have frequently been reported in the literature for selected patients.2–7 Over that same period, however, the laryngectomy population itself has changed dramatically. With the advent of organ preservation protocols as a treatment method for advanced laryngeal cancer, many individuals undergo surgical salvage as a secondary procedure. The combined effects of chemoradiation and surgery (often From the NW Clinic for Voice and Swallowing, Department of Otolaryngology–Head and Neck Surgery (D.J.G., A.D.P.), Portland, Oregon; Department of Otolaryngology–Head and Neck Surgery, Oregon Health and Science University, Portland, Oregon (P.E .A .), Department of Otolaryngology–Head and Neck Surgery, Portland Veterans Affairs Medical Center, Portland, Oregon (J.I.C.), U.S.A. Editor’s Note: This Manuscript was accepted for publication October 5, 2010. The devices used in this study were provided by Atos Medical at reduced cost in order to make this trial possible. The authors have no other funding, financial relationships, or conflicts of interest to disclose. Send correspondence to Donna J. Graville, Dept. of Otolaryngology–Head and Neck Surgery, Oregon Health and Science University, 3181 SW Sam Jackson Park Rd., Portland, OR 97239-3098. E-mail:
[email protected] DOI: 10.1002/lary.21380
Laryngoscope 121: April 2011
with reconstruction) mean that postoperative complications are common and may delay speech and swallowing rehabilitation.8,9 Moreover, individuals who undergo complex surgical reconstruction may be poorer candidates for a TEP.10 In an attempt to meet the needs of this changing population, manufacturers have produced a variety of new devices with a range of features designed to increase device-life and prevent frequent failure, but there is little scientific data about which patients are the best candidates for these devices. Moreover, given the fact that these devices vary substantially in cost, there is little information for clinicians and patients about whether the increased prices are justified or whether there are cost savings to third-party payers by their use. One of these devices, the Provox ActiValve (Atos Medical, Ho¨rby, Sweden) has shown significantly increased device life over traditional indwelling devices in studies in Europe.11 The device has two advantages over the first generation of indwelling TEPs. First, it uses two magnets inside the valve mechanism to increase resistance to inadvertent opening (e.g., during swallowing or inhalation), which is one cause of early replacement. Second, the valve and valve seat are manufactured from fluoroplastic, which are resistant to the build up of Candida on the device (another cause of failure). Frequent device failure can be a significant Graville et al.: ActiValve Efficacy and Cost-Effectiveness
769
problem for some individuals, as it may result in aspiration of liquids from the esophagus into the trachea, increased chest congestion, multiple clinic visits, and substantially increased device-related costs. Studies in Europe have found that the life of the ActiValve may be between 13 to 16 times greater than that of a conventional indwelling TEP.11,12 One of the barriers to its adoption in the United States, however, has been the fact that it is significantly more expensive than traditional devices. Consequently, a prospective trial of the device was undertaken in individuals who experienced below average device life despite traditional management methods. The first goal was to determine whether these individuals would experience benefits in terms of increased device life by use of the ActiValve. Second, we wished to determine whether users were satisfied with their voice quality with the new device. Third, we wished to determine whether there would be a significant reduction in cost to third-party payers as a result of the use of the device in this population.
MATERIALS AND METHODS The study was approved by the institutional review board. In order to be eligible for inclusion in this study, subjects were required to meet the following criteria: be medically stable; not currently be receiving cancer treatment; be at least 1 year postlaryngectomy; be able to independently manage their daily TEP care; have worn an indwelling TEP for at least 1 year and use it as their primary means of communication; not require upsizing or downsizing of the prosthesis due to granulation tissue or edema of the tracheoesophageal wall over the course of their last four TEP changes; be compliant with any prescribed medical treatment for gastroesophageal reflux disease (GERD), if appropriate; be compliant with any prescribed anti-candidal protocol in the past, if appropriate; and continue to experience device failure of 3.5 months or less (on average) over the course of their last six changes. Individuals who met the eligibility requirements and were deemed appropriate candidates for ActiValve placement underwent initial evaluation. This consisted of gathering relevant background information regarding previous device life and complications from the patient’s chart. Information regarding reflux and anti-candidal protocols was also gathered via interview. After placement of the ActiValve, follow-up occurred within 2 to 4 weeks of initial placement, and at subsequent clinic visits. A structured interview was conducted and, at long-term follow-up, the Voice-Related Quality of Life instrument (V-RQOL) was administered. The VRQOL is a 10-item self-administered questionnaire that evaluates the impact of a voice disorder on day-to-day activities.13 It has been shown to correlate with another widely used voice measure, the Voice Handicap Index14 and to be valid for use with TEP users.15 Between January 2005 and January 2008, 11 individuals who were being treated at the Department of Otolaryngology-Head and Neck Surgery at Oregon Health and Science University (OHSU) were enrolled Laryngoscope 121: April 2011
770
TABLE I. Background Characteristics of the Sample (N 5 11). Characteristic
No. (%)
Gender Male
10 (91)
Female Age, yr
1 (9)
Mean
64.3
Range 52–79 Months after laryngectomy at first ActiValve placement Mean Range Surgical closure Primary Radial/ulnar forearm free tissue transfer Nonsurgical treatment
45.1 22–108 8 (73) 3 (27)
Radiation treatment (before surgery)
6 (55)
Radiation treatment (after surgery) Chemotherapy
5 (45) 3 (27)
Other surgical procedures Dilation for esophageal stricture Fundoplication for GERD
3 (27) 2 (18)
Endoscopic diverticulectomy
1 (9)
GERD ¼ gastroesophageal reflux disease.
into the study on an ongoing basis. The study group consisted of 10 men (91%) and one woman (9%). The mean age was 64.3 years (range, 52–79 years). The time since total laryngectomy was an average of 45.1 months (range, 22–108 months). All subjects had had radiation treatment, either before (55%) or after surgery (45%). In addition, three individuals (27%) had also undergone chemotherapy. Eight individuals had had a total laryngectomy with primary closure (73%), whereas three had undergone partial pharyngectomy with radial forearm free tissue transfer (27%). In terms of other surgical procedures, three individuals required dilation of an esophageal stricture (27%), two individuals had had a fundoplication for gastroesophageal reflux (18%), and one individual had an endoscopic diverticulectomy for a diverticulum that occurred at the site of the pharyngeal closure (9%). These background data are summarized in Table I.
Statistics Paired-samples t tests were conducted to analyze whether there were significant mean differences in any of the following: 1) average duration of in situ prosthesis life for the last six indwelling TEPs as compared to the ActiValve, and 2) estimated cost of standard TEP replacement as compared to the ActiValve, and 3) overall score on the V-RQOL of the ActiValve users compared with a control group. Descriptive statistics were used to summarize responses with regard to the use, maintenance, and satisfaction with aspects of the ActiValve from the follow-up questionnaire. Graville et al.: ActiValve Efficacy and Cost-Effectiveness
RESULTS A total of 11 individuals were enrolled in the study and seen for follow-up. Of these, eight (73%) completed a successful trial of the ActiValve (i.e., wore it for at least 6 months). Of the remainder, two (18%) discontinued use of the ActiValve due to issues with granulation tissue requiring frequent replacement of the TEP, and one (9%) experienced accidental extrusion of two separate ActiValves and was not considered a good candidate for further devices as part of the study. All subjects were followed up either by phone or in person 2 to 4 weeks after initial placement. If subjects were experiencing difficulty with the strength of the magnet originally placed, it was replaced. Of the 11 original study participants, five (45%) had an ActiValve with a Light-strength magnet placed, and the remaining six (55%) had a device with a Strong magnet at their original visit. Three of the 11 participants (27%) required the device to be changed within the first 30 days due to issues with valve strength. Two (18%) required the valve strength to be decreased from Strong to Light due to difficulty voicing, and one individual (9%) required the device strength to be increased from Light to Strong due to leakage of liquids through the valve. Over the course of the study, the majority of the eight individuals who continued to use an ActiValve tended to remain with the strength of device that had been selected within this 30-day window, with the exception of one individual whose device life with the ActiValve was improved after switching from a Strong to an XtraStrong magnet. Thus, comparing the initial to the final strength of magnet for these eight individuals, three participants (37%) had a Light strength magnet placed, five participants (63%) had a Strong magnet, and none had an XtraStrong at their initial visit. By the conclusion of the study, three wore a Light magnet (38%), four (50%) were wearing a Strong magnet, and one (13%) wore an XtraStrong. Individuals who continued to wear the ActiValve were interviewed about issues related to maintenance, ease of communication, and overall satisfaction with the device, as summarized in Table II and Table III. The majority of individuals did experience the valve occasionally remaining resistant to opening for voicing (63%), and slightly more than a third were bothered by this (38%). On average, our subjects lubricated the valve slightly less than twice daily (mean, 1.88 times per day). None of our subjects were bothered by the need to do this. It should be noted that one individual did not lubricate the valve routinely, although he had been directed to do so as part of the daily management of the ActiValve. When the valve did become occluded the most common strategies were forcing air through the valve (25%), using the cleaning brush (25%), using the lubricant (25%), or drinking water (13%). In terms of the differences in voicing with the ActiValve compared to a traditional indwelling TEP, surprisingly, the majority of our respondents reported that either there was no difference or it was easier to initiate voicing with the ActiValve (75%). Speaking was the same or easier for the majority also (88%) and the Laryngoscope 121: April 2011
TABLE II. Issues Related to Valve Maintenance (n 5 8). Characteristic
No. (%)
Routine cleaning of the ActiValve Use brush Use flush
8 (100) 2 (25)
Issues relating to valve lubrication Use lubricant Frequency of lubrication per day (average)
7 (88) 1.88
Frequency of lubrication per day (range)
0–3
Management strategies when valve sticks Force air through valve
2 (25)
Use brush
2 (25)
Use lubricant Other: drink water
2 (25) 1 (13)
same percentage reported no difference in the number of words they could say on a single breath. The majority (88%) also noted no difference in how easily their speech was understood and considered their intelligibility to be the same or better. Most (75%) also said that the loudness of their voice was the same or better. In terms of the pitch of their voice, the majority (75%) said that it was the same, whereas one quarter reported that it was lower than before. Similarly, in terms of speech rate, the majority (75%) reported that their speech rate was unchanged, whereas one quarter reporting that their speech rate was slower than before. With regard to bloating of the stomach after speaking, the majority (75%) again reported that it was the same, whereas one quarter reported less bloating than with their previous TEP. Bloating after exercise was the same for seven individuals (88%) and less for one individual (13%). Asked about use of their usual stomal attachments (whether a heat-and-moisture exchanger or a hands-free valve), one half of the sample reported that use of the device was the same, whereas three individuals (38%) reported that use of the hands-free valve was easier. Overall, subjects adapted quickly to the new device, responding that it took somewhere between 0 to 4 days (mean, 1.07 days). When asked about which type of device they would prefer in the future, seven individuals (88%) preferred the ActiValve over a traditional TEP, and one individual (13%) was unsure. This individual felt more itching around the tract with the ActiValve but continued to wear the device without any complications. The reasons for choosing the ActiValve were fairly consistent across the rest of the sample, namely the reduced fear of choking, reduced aspiration, and reduced trips to the clinic for TEP changes without a significant decline in voice quality. One other observation was that individuals noted a significant reduction in the aspiration of saliva at night. This was not an area that had been included in the formal questionnaire, but for at least two subjects appeared to be associated with a perception of improved respiratory status and decreased secretions in the morning. Graville et al.: ActiValve Efficacy and Cost-Effectiveness
771
TABLE III. (Continued).
TABLE III. Issues Related to Voicing (n 5 8). Characteristic
Characteristic
No. (%)
Same
Does valve sometimes stick when trying to speak? Yes Does this bother you? Yes
5 (63) 3 (38)
Yes 0 Do you hear a clicking from the valve when talking?
Yes 0 Compared to your old TEP, how easy is it to get your voice started? 2 (25) 2 (25)
5 (63) 2 (25)
Harder
1 (13)
Compared to your old TEP, can you say more or less on a single breath? More Same
2 (25) 5 (63)
Less
1 (13)
Compared to your old TEP, is your voice easier or harder to understand? Easier 3 (38) Same
4 (50)
Harder 1 (13) Compared to your old TEP, is your voice louder or quieter? Louder
3 (38)
Same Quieter
3 (38) 2 (25)
Compared to your old TEP, is the pitch of your voice higher or lower? Higher Same
0 6 (75)
Lower
2 (25)
Compared to your old TEP, do you talk faster or slower? Faster 0 Same
6 (75)
Slower 2 (25) Compared to your old TEP, is there a difference in bloating after speaking? Less 2 (25) Same
6 (75)
More 0 Compared to your old TEP, is there a difference in bloating after physical exercise? Less
1 (13) (Continued)
Laryngoscope 121: April 2011
772
1.07 0–4
ActiValve Old TEP
7 (88) 0
Don’t Know
1 (13)
TEP ¼ tracheoesophageal voice prosthesis; HME ¼ heat-and-moisture exchanger; NA ¼ not applicable.
Compared to your old TEP, how easy is speaking easier or harder? Easier Same
3 (38) 1 (13)
Which prosthesis would you prefer in the future?
Have other people commented on the clicking?
4 (50)
4 (50)
Hands-free easier NA Mean Range
(If yes) How annoying do you find this? Not at all 2 (25)
Same Harder
Same
How many days did it take to get used to the ActiValve?
2 (25)
Easier
7 (88)
More 0 Do you notice any differences in using your HME or hands-free device?
Does it bother you that you need to lubricate valve?
Yes
No. (%)
Devices that have increased resistance to accidental opening and leakage tend also to require greater pulmonary exertion for speech. Thus we anticipated that increased device life would be associated with increased effort in speaking and possibly poorer vocal quality. In order to determine whether the ActiValve caused an increase in voice-related difficulties, subjects completed the V-RQOL at long-term follow-up. As there is no normative data for this instrument in an alaryngeal group, scores were compared with those from a group of controls. The control group was composed of a convenience sample of 29 TEP users recruited through the otolaryngology clinic at OHSU. All subjects in the control group had been a laryngectomee for at least 1 year, had had a total laryngectomy without additional reconstructive surgery, used the TEP as their primary method of communication, and were judged to be independent with its use and maintenance. For all three domains of the VRQOL, the mean scores of the ActiValve group were higher than those of the controls, including the SocialEmotional Subscale (75.78 6 20.44 vs. 70.69 6 24.78, respectively), Physical-Functional Subscale (66.88 6 17.92 vs. 62.07 6 20.24), and Total Score (67.01 6 19.10 vs. 62.45 6 19.86). These results are presented graphically in Figure 1. There were no significant differences between the two groups on any of the three subscales. For each of the eight subjects, the average device life for their last six indwelling TEPs was calculated. The eight participants wore 48 traditional indwelling devices for a combined total of 92.83 months. The participants’ average device life for a traditional indwelling TEP ranged from 0.99 to 3.38 months, and the group mean was 1.93 months (65.16 months). Excluding the three initial replacements due to difficulties with magnet strength in the first 4 weeks, the eight subjects in the study subsequently wore 28 ActiValves for a total of 243.59 months. The participants’ average device life with the ActiValve ranged from 5.98 to 14.70 months, with a group mean of 10.30 months (63.51). Using a Graville et al.: ActiValve Efficacy and Cost-Effectiveness
Fig. 1. Voice-Related Quality of Life scores for ActiValve wearers and controls.
paired-samples t test, the ActiValve was found to have significantly longer in situ life (P ¼ .001). These data are presented graphically in Figure 2. Reasons for replacement of the TEP were also compared. For each of the eight subjects, data were compiled from chart reviews regarding the reasons for replacement of the indwelling TEP. The TEPs worn during this time were either Provox2 devices (85%; Atos Medical) or the Advantage Indwelling TEP (15%; InHealth Technologies, Carpinteria, CA). Of the 48 devi-
ces worn, 44 devices (92%) were replaced for leakage through the device, three devices (6%) were replaced due to leakage around the TEP or for size, and one was replaced due to voicing difficulty (2%). There were no replacements due to granulation tissue or for other reasons during this time. These data are consistent with the entry criteria for the study, because individuals were required to have had a stable length over the course of their last four TEP changes and also have frequent failure of the device before they were enrolled. With regard
Fig. 2. Comparison of the average device life of the indwelling tracheoesophageal prosthesis (TEP) and the ActiValve by subject.
Laryngoscope 121: April 2011
Graville et al.: ActiValve Efficacy and Cost-Effectiveness
773
TABLE IV. Summary of Indwelling Tracheoesophageal Voice Prosthesis and ActiValve Data by Subject (n 5 8).
Subject
Age at A/V Placement (yr)
No. of Devices (TEP)
Mean TEP Life (mo)
Total Months (TEP)
Estimated Total Cost* (TEP)
Estimated Cost/ Month (TEP)
1890.00 1890.00
121.73 199.50
No. of Devices (A/V)
4 3
Mean A/V life (mo)
Total Months (A/V)
Estimated. Total Cost (A/V)
Estimated. Cost/Month (A/V)
7.70 13.72
23.09 41.15
3600.00 5100.00
155.90 123.93
Estimated. Annual Change in Cost
1 2
71 72
6 6
2.59 1.58
15.53 9.47
410.02 906.81
3
59
6
2.77
16.61
1890.00
113.77
3
8.87
17.73
2700.00
152.28
462.09
4 5
64 52
6 6
1.52 1.42
9.11 8.52
1890.00 2274.00
207.42 266.91
2 7
14.01 5.98
28.03 35.86
2600.00 7100.00
92.77 198.02
1375.83 826.70
6
78
6
0.99
5.92
1954.00
330.01
7
6.59
39.54
7100.00
179.57
1805.30
7 8
58 59
6 6
1.23 3.38
7.37 20.30
1890.00 1890.00
256.50 93.12
1 4
14.70 10.87
14.70 43.49
1700.00 6000.00
115.62 137.97
1690.62 538.21
Mean
64.13
6
1.94
11.60
1946.00
198.62
3.88
10.31
30.45
4487.50
144.51
649.37
SD
8.77
0
0.86
5.16
134.41
84.17
2.17
3.51
11.12
2125.65
34.35
986.61
*Costs estimated using prices on January 1, 2009. TEP ¼ tracheoesophageal voice prosthesis; A/V ¼ ActiValve; SD ¼ standard deviation.
to the 31 ActiValve replacements, the most common reason for replacement of the ActiValve was leakage through the device, which occurred in 12 cases (39%). Six devices (19%) were replaced due to problems with granulation tissue (which in one case had caused extrusion of the device). Five (16%) were replaced due to issues with leakage around the device or size, only three (10%) were replaced due to difficulty voicing, and an additional five devices (16%) were replaced for other reasons. In one case this was due to the one-way valve having become dislocated from inside the ActiValve causing frank aspiration of liquids. One was replaced due to curling of the esophageal flange over the device, which caused difficulty voicing, one was at the patient’s primary care physician’s request due to concerns about biofilm buildup on the device; one was changed prophylactically prior to device failure (a total of 448 days after original placement); and one was changed at the participant’s request due to complaints about the device ‘‘clicking.’’ A cost analysis of the relative cost of the ActiValve to a traditional indwelling device was performed. The goal was to determine whether the relatively significantly higher cost of the ActiValve was justified given the longer duration in individuals who required frequent clinic visits and replacements with existing devices. As the study had taken place over the course of several years, for simplicity we used the price of the indwelling devices on January 1, 2009 for the estimate (i.e., $279 for the Advantage Indwelling TEP and $215 for the Provox2). Because the cost charged for a clinic visit varies regionally, we used a standard cost of $100 per clinic visit, which is approximately equal to the value of Medicare reimbursement code 92597 (the code used for a voice prosthesis evaluation) as used in other cost analyses of the TEP.16 This analysis also took into account the current pricing structure for the ActiValve, which is prorated based on its in situ life. Thus, if the device is in place for 210 days. The estimate did not include individual costs incurred by the participant such as travel- or time-related costs (although these are often considerable), as the goal was to determine the relative cost to a third-party payer. Finally, the cost of accessories such as cleaning brushes and flushes were not included, as most patients used them with both devices. The cost of the lubricant for the ActiValve was also not included, as it was felt that this cost was likely more than compensated for by discontinuing use of nystatin (the antifungal agent used to counteract Candida buildup on a traditional indwelling device, which is not required when using an ActiValve). These data are summarized by subject in Table IV. With the traditional indwelling TEP, the estimated cost for the six devices was $1346.00 (6$134.41), and when the cost of six clinic visits for replacement were added (6 $100) the estimated total average cost was $1946.00 (6$134.41). Based on the device life and type of indwelling TEP used across the last six changes, average device-related costs were estimated to be $198.62 (6$84.17) per month for each individual. The device-related costs associated with the 31 ActiValves worn by the eight subjects was an average of $4100.00 (6$1933.17) per person for the devices themselves and an average of $387.50 (6$216.71) per person for clinic visits. The average estimated ActiValve-related cost was $4487.50 (6$2125.65) per person over the course of the study. Using the device-life data, the relative cost per month of the ActiValve for each individual was estimated to be $144.51 (6$34.35). Thus, overall, there was estimated to be an average annual cost saving of $649.36 per person as a result of the use of the ActiValve in this group. Cost savings were not projected for all individuals, however. There was an estimated increased annual cost for three participants of between $410.02 and $538.21 per person. For the remaining five participants, the annual cost savings to insurance was estimated to range from $826.70 to $1805.30 per person. It was noted that the ActiValve was not cost-effective in those with a higher indwelling Graville et al.: ActiValve Efficacy and Cost-Effectiveness
device life at baseline (range, 2.59–3.38 months), but that cost savings were projected in those whose baseline TEP life was shorter (range, 0.99–1.58 months).
DISCUSSION Since its introduction in the 1970s, the TEP has become an increasingly common method of alaryngeal speech rehabilitation.1 At many facilities, the success of this method has been documented as being over 90% in selected patients.2–7 The TEP has gained ground as a method of alaryngeal speech rehabilitation; however, it is not universally available. In a national survey of speech-language pathologists involved in alaryngeal speech rehabilitation in the United States, the majority said that
