Customize your JAMA Network experience by selecting one or more topics from the list below.
Ajmani GS, Nocon CC, Brockstein BE, et al. Association of a Proactive Swallowing Rehabilitation Program With Feeding Tube Placement in Patients Treated for Pharyngeal Cancer. JAMA Otolaryngol Head Neck Surg. 2018;144(6):483–488. doi:10.1001/jamaoto.2018.0278
What is the association of a proactive speech and language pathology (SLP) program with rates of feeding tube placement and oral intake in the treatment of patients with head and neck cancer?
In this cohort study of patients who received a diagnosis of head and neck squamous cell carcinoma between 2004 and 2015, diagnosis after implementation of a proactive SLP program was associated with an increase in pretreatment SLP referral, a decrease in feeding tube placement, and improvement in the rate of posttreatment oral intake tolerance.
Integration of an SLP program into the multidisciplinary care of patients with head and neck cancer can improve patient quality of life.
A proactive speech and language pathology (SLP) program is an important component of the multidisciplinary care of patients with head and neck squamous cell carcinoma (HNSCC). Swallowing rehabilitation can reduce the rate of feeding tube placement, thereby significantly improving quality of life.
To evaluate the initiation of a proactive SLP rehabilitation program at a single institution and its association with rates of feeding tube placement and dietary intake in patients with HNSCC.
Design, Setting, and Participants
Cohort study at a tertiary care and referral center for patients with HNSCC serving the northern Chicago region.Patients were treated for squamous cell carcinomas of the hypopharynx, oropharynx, and nasopharynx from 2004 to 2015 with radiation or chemoradiation therapy in the definitive or adjuvant setting. Patients who received less than 5000 cGy radiation or underwent reirradiation were excluded.
A proactive SLP program for patients with HNSCC was initiated in 2011. Study cohorts were divided into 2 groups: 2004 through 2010 and 2011 through 2015.
Main Outcomes and Measures
Primary outcome variables were SLP referral placement and timing of the referral. Secondary outcomes were feeding tube placement and ability to tolerate any oral intake.
A total of 254 patients met inclusion criteria (135 before and 119 after implementation of SLP program; median age, 60 years [range, 14-94 years]; 77% male). With the initiation of a proactive SLP program, pretreatment evaluations increased from 29 (21.5%) to 70 (58.8%; risk ratio [RR], 2.74; 95% CI, 1.92-3.91), and rate of referral overall at any time increased from 60.0% to 79.8% (RR, 1.33; 95% CI, 1.13-1.57). Feeding tube placement rates decreased from 45.9% (n = 62) to 29.4% (n = 35; RR, 0.64; 95% CI, 0.46-0.89). Among patients receiving a swallow evaluation, feeding tube requirements were less frequent for those receiving a pretreatment evaluation (31 of 99 [31%]) than for those referred during (11 of 18 [61%]) or after (38 of 59 [64%]) treatment. The rate of tolerating any oral intake at the end of treatment improved from 71.1% (n = 96) in the preimplementation period to 82.4% (n = 98; RR, 1.16; 95% CI, 1.01-1.33).
Conclusions and Relevance
A proactive SLP program can be successfully established as part of the multidisciplinary care of patients with HNSCC and improve patient quality of life.
Head and neck squamous cell carcinoma (HNSCC) represents approximately 3% to 5% of all solid tumors, with approximately 45 000 new cases diagnosed in the United States in 2015.1 The incidence has remained constant over time due to the increasing incidence of oropharyngeal carcinoma caused by oncogenic human papillomavirus (HPV).2 Patients with HPV-positive HNSCC, which is primarily located in the oropharynx, present at a younger age and have significantly better survival compared with their HPV-negative counterparts.3 As a result, patients with HPV-positive HNSCC are more likely to sustain long-term treatment-related adverse effects.
Many patients with HNSCC with primary pharyngeal tumors undergo radiation therapy (RT) with or without chemotherapy as part of treatment in a definitive or adjuvant setting. Treatment-related adverse effects such as mucositis and xerostomia may lead to inadequate oral intake, severe weight loss, and malnutrition in this high-risk population. Patients undergoing RT or chemoradiotherapy are commonly referred for placement of a feeding tube to avoid these issues. A feeding tube helps maintain adequate nutrition and prevent weight loss. Criteria for placement and dependence vary across institutions and primary tumor subsites. For example, many centers will place a feeding tube before initiation of therapy to avoid treatment interruptions.
Previous data have shown that patients who participate in an active speech and language pathology (SLP) rehabilitation program are more likely to avoid feeding tube placement altogether and maintain adequate oral intake.4,5 In addition, patients who maintain swallow function after SLP treatment are more likely to return to regular diet.6 The goal of this study was to evaluate the initiation of a proactive SLP rehabilitation program at a single institution and its association with feeding tube placement and dietary intake in patients with HNSCC.
NorthShore University HealthSystem (NSUHS) is a comprehensive, fully integrated health care delivery system serving the northern Chicago region. It encompasses 4 hospitals and is anchored by a tertiary care center and comprehensive acute care facility. This is a retrospective evaluation of patients treated for squamous cell carcinomas of the hypopharynx, oropharynx, and nasopharynx at NSUHS from 2004 to 2015 with RT or chemoradiotherapy in the definitive or adjuvant setting. We restricted our analysis to patients with pharyngeal primary tumors because sinonasal, oral cavity, and larynx cancers have different treatment fields. We selected 2004 as the starting point because this was when intensity-modulated radiation therapy (IMRT) became standard at NSUHS. Patients receiving less than 5000 cGy radiation and those who underwent reirradiation were excluded. This study was approved by the Institutional Review Board of NSUHS. Informed consent was waived due to the retrospective nature of the study.
A comprehensive swallowing rehabilitation program was implemented at NSUHS in 2011. Prior to this program, patients received SLP referrals based on clinical symptoms (ie, dysphagia). After initiation of the program, all patients received proactive SLP referrals after tissue diagnosis and prior to any cancer-directed therapy (surgery, chemotherapy, or RT). This proactive program entails SLP evaluation prior to initiation of RT with follow-up visits at 1 to 2 weeks after first SLP visit, at the midpoint of therapy, and 2 to 3 weeks after completion of therapy to ensure accuracy of exercise performance and program adherence. Patients are given a standard program involving Masako, super-supraglottic, lingual press, mandibular range of motion/stretching, pitch glides, Shaker, Mendelsohn maneuver, and posterior tongue base retraction exercises. A process map demonstrating patient SLP referral prior to and after program initiation is presented in the Figure.
Because the proactive SLP referral program was initiated in 2011, patients were divided into 2 treatment groups: those who received a diagnosis between 2004 and 2010 and who received a diagnosis between 2011 and 2015. The primary outcome variables were SLP referral placement and timing of the referral. Secondary outcomes were feeding tube placement and ability to tolerate any oral intake at the completion of treatment. Unadjusted risk ratios (RRs) with corresponding 95% confidence intervals were computed for each outcome. Adjusted RRs for each outcome were estimated using multivariable Poisson regression with robust error variance,7,8 adjusting for age, sex, smoking status, tumor site, tumor stage, receipt of chemotherapy, receipt of surgery, and radiation dose.
Statistical analysis was performed using Stata, version 15.1 (StataCorp LP).
We identified 254 patients meeting inclusion criteria, 135 diagnosed before implementation of the SLP program and 119 in the postimplementation period. The median age in our cohort was 60 years (range, 14-94 years). A majority of patients had tumors of the oropharynx (206 [81.1%]), with fewer patients having nasopharynx (21 [8.3%]) or hypopharynx (27 [10.6%]) primary tumors. Most patients also received chemotherapy (214 [84.3%]), but only a minority of patients (65 [25.6%]) received any surgery. This cohort included patients with disease of stages I to IVc, mostly III (47 [18.5%]) or IVa (158 [62.2%]). Overall, patients received a median dose of 7000 cGy (range, 5000-7700 cGy). Characteristics of the pre-SLP and post-SLP program cohorts are presented in Table 1. Notably, the median radiation dose in the pre-SLP cohort was 7020 cGy vs 7000 cGy in the post-SLP cohort.
Among the overall cohort, a majority of patients received an SLP referral (176 [69.3%]), but only 99 (39.0%) received a referral prior to initiating treatment. Ninety-seven (38.2%) patients received a feeding tube at some point, most often during treatment (70 [72%] of those receiving a feeding tube). A majority of patients (194 [76.4%]) were able to tolerate oral intake at the completion of treatment.
Prior to initiation of the proactive program, only 29 (21.5%) patients received a pretreatment SLP evaluation (Table 2). This increased to 70 (58.8%) in the postimplementation period (RR, 2.74; 95% CI, 1.92-3.91). Of note, there was also a significant increase in rate of referral to speech pathology overall at any time (from 81 [60.0%] to 95 [79.8%]; RR, 1.33; 95% CI, 1.13-1.57). We also observed a significant decrease in feeding tube placement rates following program initiation, from 45.9% (n = 62) to 29.4% (n = 35; RR, 0.64; 95% CI, 0.46-0.89). Among patients receiving a swallow evaluation, feeding tube requirements were less frequent for those receiving a pretreatment evaluation (31 of 99 [31%]) than for those referred during (11 of 18 [61%]) or after (38 of 59 [64%]) treatment (RR for pretreatment vs during or after treatment, 0.49; 95% CI, 0.35-0.69). Ultimately, the rate of tolerating any oral intake at the end of treatment improved from 71.1% (n = 96) in the preimplementation period to 82.4% in the postimplementation period (n = 98; RR, 1.16; 95% CI, 1.01-1.33).
The associations between SLP program initiation and improved SLP program referral, decreased rates of feeding tube placement, and improved oral intake were confirmed in a multivariable analysis (Table 2); in all cases, adjusted and unadjusted RRs were similar. Patients who received a diagnosis after implementation of the SLP referral program were more likely to receive an SLP referral at any time (adjusted risk ratio [aRR], 1.38; 95% CI, 1.17-1.63) and particularly before treatment (aRR, 2.82; 95% CI, 1.97-4.04). Postimplementation patients were also less likely to require feeding tube placement (aRR, 0.68; 95% CI, 0.48-0.96) and more likely to tolerate oral intake at completion of treatment (aRR, 1.14; 95% CI, 0.99-1.31).
This study demonstrates the efficacy of implementing a proactive swallowing rehabilitation program, with subsequent reduction in feeding tube placement rate and improved oral intake in patients with HNSCC receiving RT as part of their treatment. The program improved pretreatment SLP referral rates from one-fourth to almost two-thirds of patients, with a subsequent decrease in feeding tube rates from approximately one-half to almost one-fourth of the patients. We demonstrated the importance of timing of the SLP evaluation, with fewer feeding tube placements for patients who saw SLP prior to treatment. Overall, the rate of oral intake tolerance at the end of treatment was significantly increased after program initiation.
With the established use of RT for the treatment of HNSCC, radiation-induced dysphagia has become a well-known clinical problem. In addition to its therapeutic benefits, RT produces a range of physiological, structural, and functional deficits, all of which contribute to dysphagia.9 Early and acute adverse effects may include mucositis and desquamation, xerostomia, dysgeusia, and systemic fatigue and nausea, which can diminish motivation to eat. Late and chronic changes may include muscular fibrosis and atrophy, resulting in reduced base of tongue retraction, poor epiglottic retroflexion, reduced laryngeal elevation, and poor coordination of swallowing muscles.
In the current era of HPV-related HNSCC, with its increased treatment sensitivity and younger population, survivorship and quality of life (QoL) have become important considerations and have brought renewed attention to treatment toxicities. Prospective studies have identified swallowing function as a major contributor to QoL and top priority among patients with HNSCC.10,11 Several clinical trials are investigating the oncologic and functional outcomes of RT dose reduction in the setting of deintensifying treatment for HPV-positive disease, including ECOG 1308 (NCT01084083), NRG-HN002 (NCT02254278), ECOG 3311 (NCT01898494), and PATHOS (NCT02215265). In addition, the advent of IMRT minimizes the radiation dose to the pharyngeal musculature and salivary glands, thereby reducing treatment-related dysphagia and prolonged feeding tube dependence.5,9
Although feeding tube use can prevent treatment-related weight loss and allow patients to complete treatment without interruption, it has a significant negative impact on QoL. In a survey of patients with HNSCC, Terrell et al12 found that the presence of a feeding tube was the factor that had the most negative impact on QoL. Overall, 60% of patients in our study avoided feeding tube placement. This increased to more than 70% when a formal SLP program was initiated. Not surprisingly, this finding was driven by the cohort of patients who underwent pretreatment swallow evaluations.
Such findings underscore the importance of an SLP program as part of a comprehensive multidisciplinary head and neck cancer team. Referral to an SLP program is essential for functional rehabilitation of patients with HNSCC who undergo RT. Prophylactic swallowing therapy encourages ongoing use of the swallowing musculature during treatment by avoiding periods of no oral intake and providing targeted preventive swallowing exercises. The benefit of prophylactic swallowing exercise programs is increasingly supported by clinical studies,4-6,13-15 including 3 randomized clinical trials.16-18 A recent evidence-based review of dysphagia reduction strategies showed a nonsignificant finding of benefits for a preventative exercise program.19 Favorable outcomes reported in patients who receive preventive swallowing therapy include superior swallow-related QoL scores,13,18 reduced rate of feeding tube placement and dependence,4,5 maintenance of oral diet during treatment and return to regular diet after treatment completion,6 better base of tongue retraction and epiglottic inversion,14 reduced extent and severity of penetration and/or aspiration in the short and long term,15,16 and greater preservation of the genioglossus, mylohyoid, and hyoglossus muscles on magnetic resonance imaging.17 Patients also reported more physician engagement and greater satisfaction with swallowing function.20 Not only did feeding tube rates significantly decrease in the present study after SLP program initiation, but also the rate of any oral intake tolerance at the end of treatment significantly improved from 71.0% to 82.6%.
The present study demonstrates the efficacy of implementing a proactive SLP rehabilitation program within a comprehensive cancer care facility. A 2012 survey of SLPs in the United States revealed that only 18.3% intervene proactively and see their patients with HNSCC routinely prior to treatment.21 This contrasts with 50% of SLP teams in the United Kingdom, who provide pretreatment evaluation of patients with HNSCC. This difference may be accounted for by the appointment of speech and language pathologists in the National Health Service as core members of a multidisciplinary head and neck cancer team as stated in their national guidelines.22 Despite varying practice patterns and potential institutional and/or cultural barriers, we found the implementation of an SLP program to be practical and associated with improvement in functional outcomes.
This study is strengthened by its large sample size and comparable number and characteristics of patients who were treated before and after SLP program implementation. However, there are several important limitations to our analysis. This study included a mixed group of patients who underwent both adjuvant and definitive RT regimens, although we note there was no significant difference in timing of RT between patients before and after SLP program initiation. Furthermore, other patient, tumor, and treatment factors that could affect swallowing function were not analyzed extensively. We note that the inclusion of these factors in multivariable analysis did not alter our findings regarding the association of the SLP referral program with the studied outcomes. We were also unable to definitively determine the type, frequency, and duration of swallowing exercises or exercise nonadherence, which can diminish the benefits of a proactive SLP program.23,24 We did not specifically correlate oral intake or feeding tube placement to program adherence; this will need to be the subject of future studies prospectively evaluating the impact of SLP evaluations and exercises.
Increasing the rate of SLP referrals before the start of RT for patients with HNSCC was associated with decreased rates of feeding tube placement and improved overall oral intake. Implementation of a proactive SLP program can be successfully established as part of the multidisciplinary care of patients with HNSCC and improve patient QoL.
Accepted for Publication: February 22, 2018.
Corresponding Author: Mihir K. Bhayani, MD, Kellogg Cancer Center, Evanston Hospital, 2650 Ridge Ave, Evanston, IL 60201 (firstname.lastname@example.org).
Published Online: April 19, 2018. doi:10.1001/jamaoto.2018.0278
Author Contributions: Mr Ajmani and Dr Bhayani had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.
Study concept and design: Campbell, Bhayani.
Acquisition, analysis, or interpretation of data: Ajmani, Nocon, Brockstein, Campbell, Kelly, Allison.
Drafting of the manuscript: Ajmani, Nocon, Allison, Bhayani.
Critical revision of the manuscript for important intellectual content: Ajmani, Nocon, Brockstein, Campbell, Kelly.
Statistical analysis: Ajmani.
Administrative, technical, or material support: Nocon, Campbell, Kelly, Allison, Bhayani.
Study supervision: Nocon, Campbell, Bhayani.
Conflict of Interest Disclosures: All authors have completed and submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest and none were reported.
Additional Contributions: Charlotte Wilson, BS, NorthShore University HealthSystem, provided assistance with patient data extraction, and Jennifer Jaffe, MPH, NorthShore University HealthSystem, provided administrative support. They were not compensated beyond their salaries.