Dysphagia, Speech, Voice, and Trismus following Radiotherapy and/or Chemotherapy in Patients with Head and Neck Carcinoma: Review of the Literature

Introduction. Patients with head and neck cancer suffer from various impairments due to the primary illness, as well as secondary consequences of the oncological treatment. This systematic review describes the effects of radiotherapy and/or chemotherapy on the functions of the upper aerodigestive tract in patients with head and neck cancer. Methods. A systematic literature search was performed by two independent reviewers using the electronic databases PubMed and Embase. All dates up to May 2016 were included. Results. Of the 947 abstracts, sixty articles met the inclusion criteria and described one or more aspects of the sequelae of radiotherapy and/or chemotherapy. Forty studies described swallowing-related problems, 24 described voice-related problems, seven described trismus, and 25 studies described general quality of life. Only 14 articles reported that speech pathologists conducted the interventions, of which only six articles described in detail what the interventions involved. Conclusion. In general, voice quality improved following intervention, whereas quality of life, dysphagia, and oral intake deteriorated during and after treatment. However, as a consequence of the diversity in treatment protocols and patient characteristics, the conclusions of most studies cannot be easily generalised. Further research on the effects of oncological interventions on the upper aerodigestive tract is needed.


Introduction
Head and neck oncological patients suffer from various functional, physical, and emotional impairments due to both the primary illness and the secondary consequences of the tumor treatment [1]. The oncological treatment of head and neck tumors depends on the location and the stage of the tumor, as well as the treatment preferences of the individual patient. Head and neck oncological treatment can include surgery, radiotherapy, chemotherapy, or combinations of these. The impact of head and neck oncological treatments on the anatomical structures, organ function, and the quality of life (QoL) should not be underestimated [2]. For instance, the implications of loss of function for people treated nonsurgically for head and neck cancer (HNC) and its detrimental effects on functioning and QoL are well documented [3].
In order to assist people with dysphagia to adjust to and live successfully with the sequelae of the primary condition, speech pathologists managing this caseload need to ensure posttreatment services are available [4] that address not only the physical but also the emotional and psychosocial needs. A qualitative study by Nund et al. [5] exploring dysphagia management by speech pathologists suggests that care givers generally feel ill-prepared for their role. Furthermore, this study suggests that clinicians should provide adequate and timely training and support to carers. Furthermore, Krisciunas et al. [6] concluded that within speech pathology there is no standardised therapy for HNC patients and scant evidence to support any particular protocol. As a result, institutions and individual speech pathologists need to develop their own protocols based on "standard" practices or anecdotal evidence.
Evidence-based practice (EBP) is hailed to be paramount in the practice of speech pathology [7]. The American Speech-Language-Hearing Association (ASHA) defines evidence-based practice as ". . .an approach in which current, high-quality research evidence is integrated with practitioner expertise and client preferences and values, into the process of making clinical decisions" [8]. Essentially, EBP involves moving the foundation for clinical decisions from clinical protocols centered solely on expert opinion to the integration of clinical expertise, the best current research evidence, and individual client values. To facilitate EBP in healthcare, clinical practice guidelines can be developed to summarise clinically relevant evidence [9].
Several reviews have been published about the outcomes after radiotherapy and/or chemotherapy in HNC patients (e.g., Frowen and Perry [10]; Jacobi et al. [11]; van der Molen et al. [12]; Paleri et al. [13]; Roe et al. [14]). Most of the reviews focused on selected functional domains in populations with HNC: health-related QoL [15], swallowing [13,14,16], and voice and speech [11]. Only the review by van der Molen et al. [12] covered a wider range of functional outcomes in patients with advanced HNC, including swallowing, mouth opening, nutrition, pain, and QoL. Further, the purpose of some studies was to provide evidence-based clinical guidelines (e.g., Paleri et al. [13]) and they did not perform systematic literature searches in line with the PRISMA guidelines [17]. As such, even though a number of reviews have been published over the last ten years, a comprehensive updated systematic review is needed that includes all functional domains affected by radiotherapy and/or chemotherapy in patients with head and neck carcinoma.
A systematic review was conducted to describe the effects of radiotherapy and/or chemotherapy on functions of the upper aerodigestive tract in patients with HNC and examined the evidence of interventions by speech pathologists.

Methods
A systematic literature search was performed by two independent reviewers. The electronic biomedical databases PubMed and Embase were used (search period from start of database until 5 May 2016). The searches were limited to English language publications. In PubMed the MeSH terms larynx or hypopharynx were combined with all MeSH terms related to head and neck neoplasms (Table 1). Next, the results were linked to all MeSH terms for chemotherapy or radiotherapy, after which the outcome was combined with all MeSH terms found for dysfunctions of the upper aerodigestive tract and limited with adults +19 years. The exact syntax of the literature search is presented in Table 1.
In Embase the thesaurus terms larynx or hypopharynx were linked to neoplasm and radiotherapy or chemotherapy. Next, the search outcome was combined with the following terms: dysphagia, speech, speech disorder, voice, dysphonia, xerostomia, quality of life, dysarthria, or trismus (see Table 1).
Only articles presenting both pre-and postintervention data of the upper aerodigestive tract functions of the participants were included. Review articles and studies with a population sample of less than 20 patients were excluded, as well as experiments on animals or articles not published in English. Furthermore, studies published before 1990, case reports, expert opinions, and articles describing combinations of therapy including surgical interventions were excluded.
Final decisions on inclusion were made based on the original articles by consensus between two expert reviewers in accordance with the PRISMA statement [17]. The reference lists of all the included articles were searched for additional literature. Next, the standard quality assessment QualSyst as described by Kmet et al. [18] was performed in order to evaluate the methodological strength and weaknesses of the included studies. All ratings were performed by two independent reviewers. After consensus, studies with poor methodology scores (<50%) were excluded. All included articles were classified according to the Australian National Health and Medical Research Council (NHMRC) Evidence Hierarchy [19]. Data were retrieved from all studies and tabulated; further details on selected speech pathology interventions were summarised separately.

Results
Using MeSH or thesaurus terms, 304 articles were located in PubMed and 201 in Embase. Free text word searches resulted in another 148 articles in PubMed and 397 in Embase. The combination of these searches, without overlap, yielded 947 articles. Figure 1 outlines the PRISMA reviewing process according to Moher et al. [20]. Sixty articles met all inclusion criteria. Table 2 shows the outcomes of the QualSyst critical appraisal tool by Kmet et al. [18]. As all studies had sufficient methodological quality, no further studies were excluded; the overall methodological quality ranged from adequate to good with 0 studies ranked as poor, 3 studies as adequate, 3 studies as good, and 54 studies as strong. Based on the NHMRC Evidence Hierarchy [19], 6 studies were classified as level II evidence and 54 studies as level III evidence. All 60 studies focused on different functions of the upper aerodigestive tract following radiotherapy and/or chemotherapy for HNC. The following constructs were evaluated across the different studies: communication (voice and speech), functions of the digestive tract (oral intake, weight loss, dysphagia, trismus, xerostomia, and tube dependency), QoL, and overall survival rates. Table 3 provides a summary of the 60 retrieved observational and intervention studies that met the inclusion criteria. The first column presents the reference of the author(s). The second column represents the number of subjects, the third column represents the etiology of the head and neck  malignancies, and the 4th column displays the staging of the malignancies. The 5th column shows whether voice and/or speech, digestive tract, and QoL were studied. The 6th and 7th columns show the evaluation techniques and the treatment, respectively. The 8th column presents the follow-up and the last column describes the author's key findings.
All the studies reported good to excellent outcomes for voice quality at long-term follow-up. Some studies specifically reported pre-to posttreatment improvements of voice or speech quality following radiotherapy and/or chemotherapy [22,23,28]. However, other studies [36,40,42,44] reported a deterioration after therapy at long-term follow-up. Al-Mamgani et al. [26] found a better voice outcome in case of single vocal cord irradiation compared with irradiation of the whole larynx. Mittal et al. [37] concluded that radiation with tissue/dose compensation (TDC) improved articulatory outcome compared to radiation without TDC.
Follow-up times in these studies range from immediately after therapy [62] to 6 years after therapy [35], describing both low stage tumors and more advanced tumors. Thirtytwo studies used the TNM-classification system, stage was described in six other studies, and the remaining two studies did not report on tumor stage or grade. However, it was unclear whether the clinical TNM-score or the pathological TNM-score was used to describe the severity of the disease. Eight studies [16,35,46,48,52,56,57,69,73,74] described whether the patients received functional treatment (by a speech pathologist); the remainder of the articles did not mention whether the patient received any additional treatment.
Five studies [16,51,59,61,66] showed that swallowing was least affected at baseline, worst immediately following posttreatment (0-3 months after treatment), and improved by 6-12 months after treatment and later. However, swallowing usually did not return to pretreatment functioning level. In four studies [49,53,58,74], a relation between dosevolume, dysphagia, and aspiration was found. Caudell et al. [49] showed that a mean radiation dosage >41 Gy with >24% volume of the larynx being radiated was associated with increased percutaneous endoscopic gastrostomy (PEG) dependency and aspiration. Akst et al. [46] correlated advanced tumor stage and age >60 years with a deterioration of swallowing.
Ackerstaff et al. [22] demonstrated improved oral intake postradiotherapy and/or chemotherapy. Stenson et al. [70] stated that weight remained unchanged after treatment (via oral route), whereas Nourissat et al. [62] described a mean weight loss of 2.2 kg posttreatment.
Although three studies [28,72,77] demonstrated improvements in QoL, four studies [22,36,42,54] reported a decrease in general QoL as a result of radiotherapy and/or                 chemotherapy. Bansal et al. [27] found a significant decline in physical, social, and emotional functioning as well as in global health scores following a course of radiotherapy. However, the patients' functional scores improved one month after treatment but did not reach pretreatment levels.
The health-related QoL (HRQoL) scores of the majority of patients in the Bottomley et al. [78] study returned to baseline at 48-month follow-up. These findings support the findings of Ackerstaff et al. [22], Cohen et al. [51], Karlsson et al. [32], List et al. [59], and Wilson et al. [76], who suggested that HRQoL deteriorates significantly immediately after treatment, with variable degrees of improvement 3-72 months after treatment.

Reported (Efficacy of) Speech Pathology Interventions.
We assessed the speech pathology interventions against the following criteria: (a) whether a detailed description of the intervention was provided; (b) whether the authors provided a description of treatment duration and intensity; and (c) what the speech pathology intervention outcomes were. The reported efficacy of 14 speech pathology intervention studies aimed at addressing problems in dysphagia, speech, voice, and trismus is summarised in Table 4.
The three studies that investigated the treatment of trismus [48,52,74] presented the most detailed information on what the interventions involved. The study by Dijkstra et al. [52] described a wide variety of trismus-specific therapies, suggesting that most patients received a combination of therapies. The patients in the van der Molen et al. [44] and Kraaijenga et al. [35] studies did not receive any speech therapy. The remainder of the studies reported that patients received speech therapy; however, most of these studies did not provide specific data on treatment duration or intensity. None of the voice-related studies provided information on the specific exercises prescribed to patients except Karlsson et al. [31].
Of the eight studies on swallowing disorders, only Kotz et al. [57] and van der Molen et al. [73] described the prescribed exercises in detail. The aim of the latter study was to compare the effectiveness of experimental rehabilitation to standard rehabilitation in 49 advanced HNC patients. The authors concluded that preventive rehabilitation is feasible and effective in reducing the extent and/or severity of various functional short-term effects of chemoradiotherapy [74]. This finding is supported by the 6-year follow-up study by Kraaijenga et al. [35]. Kotz et al. [57] described a temporary improvement. These are the only studies that provided detailed information about the speech pathology intervention and reported on the effectiveness of the intervention.

Discussion
In total, 60 studies met the inclusion criteria. The studies described the effects of radiotherapy and/or chemotherapy on the functions of the upper aerodigestive tract in patients with HNC. The articles yielded by this systematic review vary in their findings regarding tumor characteristics and treatment modalities. As a result of this variability, no statistical pooling was possible. We also set out to investigate the involvement of speech pathologists in treating patients with HNC.
When considering treatment outcomes, voice quality worsened at the start of radiotherapy and/or chemotherapy but eventually improved after therapy finished. Dysphagia can be a major side effect of HNC and its treatment. The high incidence of dysphagia in this study population can cause serious secondary consequences, such as malnutrition, dehydration, an increased risk of aspiration, and, at worst, death [80]. As dysphagia is a common sequela to oncological treatment, early detection and treatment are needed to avoid or minimise serious secondary complications [81].
The general description of the study population in Table 3 shows that there was great variability in both the location of the tumor and the grading/staging, making comparisons of these studies difficult. As the follow-up times varied in each study, the outcomes may be noncomparable. Thus, this review shows that there is a need for more standardised approaches to research in this field.
Additionally, a large range of outcome measures were used, some of which are not validated. This calls into question the reliability of results reported in some of the studies. The use of validated and standardised assessments in future research would provide more robust findings.
When considering the functional outcomes of radiotherapy and/or chemotherapy, one of the most important factors is whether the patient had received voice or swallowing therapy. Interestingly, only 14 of the 60 included studies reported whether the patients received any speech therapy. Thus, in 46 articles functional results, such as voice quality, are presented with no specification of whether the patient received therapy. As some of these studies have a follow-up of >2 years, it is fair to assume that patients sought help for voice or swallowing problems. Therefore, the involvement of speech therapy may be underreported, suggesting that the presented outcomes in these studies are biased and raise questions about their reliability.
When information was provided about treatment, only six articles [31,48,52,57,73,74] described in detail the treatment intensity as the actual treatment. Furthermore, these studies are the only five that include conclusions about the efficacy of speech therapy in this specific population. In the context of EBP, this finding demonstrates the need for more research into the efficacy of speech pathology interventions for patients with HNC receiving radiotherapy and/or chemotherapy.
To enable the objective reporting of the effectiveness of radiation and/or chemotherapy, baseline measurements of different aspects of voice quality and swallowing are required. To manage expectations, healthcare professionals and patients need to be made aware that some aspects of both  voice and swallowing commonly do not recover to the level prior to the oncological intervention [16]. Regarding effectiveness of voice treatments, the following multidimensional assessment is recommended [82]: a videolaryngostroboscopy recording of the laryngeal structures and the vocal fold vibration; an acoustic and a perceptual analysis of voice; a voice-related questionnaire on QoL (e.g., the Voice Handicap Index) [83]; and a functional health status questionnaire. Such a protocol would be in line with the recommendations for functional assessment of voice pathology as described by the Committee on Phoniatrics of the European Laryngological Society [84]. When describing aspects of swallowing function, both fiber optic endoscopic evaluation of swallowing and videofluoroscopy are considered to be the gold standard in dysphagia assessment [85]. In addition to these tools, questionnaires on HRQoL and functional health status are recommended and should be integrated in the overall swallowing assessment protocol. Repeated measurements of outcome measures should be performed in order to monitor any side effects of the oncological intervention, to detect spontaneous recovery, and to measure the effects of the speech pathology interventions. Apart from baseline measurements, posttreatment and follow-up measurements should be used to monitor functional and QoL outcomes.
Additional research is needed to develop clinical practice guidelines to support evidence-based practice in the area of dysphagia, speech, voice, and trismus following radiotherapy and/or chemotherapy in patients with head and neck carcinoma. These practice guidelines should bring together the best available current evidence within a specific clinical area, formulating evidence-based recommendations for clinicians and present choices between different interventions that have an impact on health and use of resources [86]. This systematic review summarised the effects of radiotherapy and/or chemotherapy on the function of the upper aerodigestive tract in patients with head and neck cancer. However, because of the marked variation in treatment protocols and patient characteristics, outcome data from the included studies cannot be easily generalised. Recommendations for future studies advocate the use of a multidimensional assessment protocol, using well-validated measures and standardised pretreatment, posttreatment, and follow-up measurements, thus allowing for future meta-analysis of homogeneous outcomes.

Conclusion
The studies included in this systematic review described a wide variety of outcomes in patients with HNC following radiotherapy and/or chemotherapy. The findings about the long-term functional implications of radiotherapy and/or chemotherapy in patients with HNC are inconclusive as a result of the wide range of outcome measures used and the possible influence of underreported speech therapy.
Future researchers need to consider targeting more homogeneous groups using standardised treatment protocols to improve the treatment outcomes, thereby decreasing the side effects of the oncological treatments. Findings of these studies need to inform the decision-making process in the treatment of HNC so complications can be better predicted with due consideration of the possible negative side effects to the upper aerodigestive tract. Although the main objective of most studies was to determine curing rates, the importance of the functional implications of the side effects of oncology treatments should not be overlooked, particularly their impact on QoL. Finally, more research is needed to gain a full understanding of the complexity and variety in the effects of radiotherapy and/or chemotherapy on the functions of the upper aerodigestive tract following HNC.