The use of a PG tube to prevent malnutrition in HNC patients receiving definitive CCRT has gained a lot of attention in recent literature; however, the appropriate guidelines in clinical practice have not been established yet. This survey reporting on the Belgian HNC radiation oncologists’ current clinical practice and their opinion on whether and when to consider a PG tube had a 100% response rate.
Ninety six percent of centers reported to yearly irradiate more than 20 HNC patients and 71% even treat more than 50. Facility volume improves a variety of clinical processes, including access to supportive care such as pain management, swallow/speech therapy and nutrition that increase the probability of treatment completion, minimize the likelihood of treatment interruptions, and mitigate morbidity. There is an emergent body of evidence that patients with HNC who are treated at high-volume centers also have better outcomes (16–19). Within a randomized trial of the Radiation Therapy Oncology Group (RTOG 0129), which compared cisplatin concurrent with standard versus accelerated fractionation radiotherapy Wuthrick et al. found the 5-year OS rate to be 69.1% vs 51.0% (p = .002), respectively, for patients treated at historically low- vs high-accruing centers (20). In 2019, the Belgian Health Care Knowledge Centre (KCE) published an evaluation report on the quality of care in HNC patients in Belgian hospitals according to quality indicators and objectives defined by a panel of experts. According to this KCE report, 9175 head and neck squamous cell carcinoma (SCC) were treated in 99 different centers during the six year study period. It was noted that the median survival of patients treated in high-volume centers (hospitals treating more than 20 patients per year) was 1.1 year longer than their peers treated in low-volume centers (5.1 versus 4.0 years) (21). Regarding radiotherapy volume, 4539 head and neck squamous cell carcinoma (HNSCC) were treated in Belgian radiotherapy (RT) centers between 2009 and 2014. The median RT center volume was 169 patients over the six year period (i.e. 28 patients per year) with a quarter of the centers treating less than 17 patients per year. There was no statistically significant association between RT center volume and overall survival among patients with HNSCC (p = 0.61). Assuming that our respondents’ answers are in agreement with the clinical practice, 96% of Belgian radiotherapy centers are in line with this quality indicator of offering personalized care and treatment to more than 20 HNC patients per year. The difference with KCE numbers of treated patients in RT departments can be explained by the KCE selection criteria which included only first treatments for SCC of the oral cavity, oropharynx, hypopharynx and larynx (nasal cavities, thyroid and salivary glands excluded) while 3287 patients (26%) with multiple synchronous tumors were left out of the analysis.
We found 83% of all gastrostomy tube indications to be discussed in a HNC dedicated multidisciplinary board. In the absence of a golden standard, the role of the interdisciplinary team is crucial to assess for each case the appropriateness of a nutritional intervention (22). A multidisciplinary approach provides more accurate treatment recommendations, communication, reinforces cooperation, coordination and adherence to clinical guidelines (23)(22). The combination of a HNC multidisciplinary expert team in a high volume referral cancer centre is considered an important indicator of quality of care for HNC and associated with better therapeutic decisions (24).
The most interesting results from our survey include the different factors and their importance in the decision for the PG in case of a HNC patient. Predicting which patient will benefit from PG is challenging. However, we were able to evaluate a number of factors that may correlate with the development of swallowing dysfunction during CRT. The foreseen irradiated mucosal volume, followed by anatomical site has been considered most important by the respondents. It is clear that the irradiated mucosal volume is different in function of anatomical site and lymph node involvement. Many studies have demonstrated a relationship between the dose received by anatomical structures involved in swallowing (e.g. the superior pharyngeal constrictor muscle) and radiation induced acute and late dysphagia. (25–31). Subsequently, several studies are currently focusing on reducing the elective radiation dose and the irradiated volume in order to decrease acute and late swallowing dysfunction (32–35). Langendijk et al. developed a predictive model to identify patients at high risk of radio-induced dysphagia (14). Advanced tumour stage (T3-T4), oropharyngeal and nasopharyngeal tumour site, primary and bilateral neck irradiation, weight loss at baseline, and treatment modality (accelerated RT or CCRT) were identified as independent factors predicting swallowing dysfunction. In Belgium, such studies are focussing on reduction of the elective dose and on volume individualisation of the prophylactic nodal target irradiated zone using the identification of the sentinel node (33, 35).
Another factor we evaluated is the patient’s choice. Despite the growing interest in supporting the patient’s participation in clinical decisions, there is no evidence to guide clinicians regarding the level of patients’ involvement in the decision-making process. Patient’s preference for involvement may vary between those preferring to take part on their own treatment, to those who prefer to leave treatment decisions to their medical team, largely as patients report lacking the specialized knowledge needed to make treatment decisions (36).
Swallowing dysfunction also has a significant impact on health-related quality of life, even more than xerostomia, as reported by patients (14). Prophylactic endoscopic gastrostomy (PEG) tube may also negatively affect the psychological status of the patients as it may interfere with family life, intimate relationships and social activities (37). However, a recent systematic review of the effect on enteral tube feeding on health related quality of life suggests PG placement to be effective in improving quality of life for patients with HNC cancers treated with CCRT (38).
The respondents accorded moderate importance in the decision making regarding PG tube to the sort of multimodality treatment, more precisely the postoperative versus definitive setting. Surgery before radiotherapy and extent of reconstruction appear to be important factors to develop swallowing problems during postoperative treatment (39–41). Finally, local expertise and available techniques are important in the planning of the PEG tube. Different techniques of PEG placement are available across the Belgian radiotherapy centres, with endoscopic placement being the most commonly used (68%). There are, however, no randomised trials comparing these different techniques, hence current evidence is only based on retrospective and non-randomized controlled studies (42). It is recommended to base the choice of the technique on indications and contraindications, local experience and the available techniques (43). Complications related to the PEG tube placement by different techniques are quite rare and range from minor infections and bleeding to peritonitis (44). One of the most serious complications is abdominal wall metastasis following PEG placement. This risk is correlated with advanced tumour stage, tumour biology and the technique (45). As such, the “Pull” technique instead of the “Push” technique was identified as a risk factor in a large retrospective study with 777 HNC patients where the incidence of abdominal wall metastasis was 0.64% (46).
Regarding optimal starting of the enteral nutrition, the majority of respondents opted to start nutritional support when clinically indicated, more specifically in case of deterioration of swallowing or nutritional status including weight loss. As the CCRT side effects usually start around the second or third week of treatment, patients could continue oral food intake these first couple of weeks of radiotherapy. The benefit of the maintenance of oral intake was demonstrated by Hutcheson et al. who retrospectively analysed swallowing outcome in 497 HNC patients treated with CCRT. Maintenance of oral intake throughout treatment was associated with better long term swallowing function and less long term gastrostomy dependency (15). Brown et al conducted a randomized controlled study comparing early versus postponed feeding nutrition (i.e. when clinically indicated) in HNC patients with PG tube placement. They concluded that early use of the PG tube did not result in an increase in long term dependency. However, swallowing outcome measurements were not included in this study (47). As there are no general guidelines on when to remove the gastrostomy the majority of survey respondents opted to wait until the patient is able to assume an adequate oral feeding. Alternatively, it is defendable to wait for the first follow up exams to ensure that no further salvage therapy is required. Salvage treatment such as neck dissection could increase the risk of PEG tube dependency in HNC patients (48).
One of the most interesting results from the data presented is the disparity with regard to the influence of PG tube on swallowing outcome. This is consistent with published evidence on the possible negative effect of PG tube placement on long-term swallowing function (49–51). Shaw et al. conducted a systematic review on this subject and concluded to a lack of consensus in literature regarding the impact on late swallowing function of the use of a PG (13). We found the main debate to be about the importance of maintaining adequate nutrition during treatment versus maintaining swallowing function.
There are some limitations of this study. The first and most important is the subjective matter given the survey is opinion based. The questions were designed with the multidisciplinary team for clarity and reliability however they have not been tested before being sent to the participants. Additionally, only a single national speciality was surveyed (radiation oncology), while the opinions of head and neck surgeons, medical oncologists and other physicians who care for HNC patients being clearly important, they were outside of the scope of this survey.