Abstract
Background/Aim: Head and neck cancer (HNC) is associated with a high risk of malnutrition. Malnutrition is defined as acute weight loss greater than 5% and increases mortality 1.7-fold for HNC patients. The aim of the study was to investigate the social and nutritional impairments that patients face throughout cancer-survivorship. Patients and Methods: The study was conducted nationwide via the self-help network and in a single oncological center. We analysed 134 patient reported outcome (PRO) questionnaires with a mean age of 65.5±9.4 years, including 88 males, 36 females, and ten participants of undisclosed sex. The questionnaire contained 47 items enquiring about demography, weight development, and how treatment impaired nutrition. Results: The patient data showed a weight loss of 8% after surgery, 13% after (chemo)radiotherapy [(C)RT] and a return to baseline weight in convalescence. However, patients with a baseline weight >100 kilogram (kg), had a 22% weight loss after (C)RT (p<0.0001) and this remained permanent at 11% (p=0.0041). Treatment-associated side-effects gradually decreased in the course compared to the time of treatment: loss of taste (55% to 21%), xerostomia (56% to 42%), dysphagia (57% to 43%), and dental problems (33%/ to 21%). (C)RT immediately led to more loss of taste (p=0.0461) and dysphagia (p=0.0334), and surgery as a singular modality scored the lowest odds ratio for side-effects. Social Impact: mood, supporting networks, and supplement satisfaction were rated “good” (Likert Scale). Conclusion: Malnutrition is common among HNC patients. High baseline weight and extensive multimodal treatment are important risk factors that require enhanced stewardship.
- Nutrition
- malnutrition
- head and neck cancer
- carcinoma
- narrative interview
- weight trajectory
- nutrition impact symptom
- patient reported outcome
- PRO
- self-help group
Even before treatment, Head and Neck Cancer (HNC) patients are at a high risk of malnutrition (1-4). Reasons vary from tumor-induced wasting, inflammation, and mechanical obstruction but also include nutrition impact symptom(s) (NIS) (5-7). NIS are defined as impairments due to treatment which affects weight and the level of malnutrition (8, 9). NIS include alterations of smell and taste, dysphagia, mucositis, and xerostomia. Furthermore, vomiting, nausea, pain, dental and respiratory problems may cause malnutrition (10-17). Detecting and treating malnutrition is a challenge (18-20). According to Cederholm et al. the European Society for Clinical Nutrition and Metabolism (ESPEN) guidelines define malnutrition as “weight loss (unintentional) >10% indefinite of time, or >5% over the last 3 months combined with either” (21). Detecting malnutrition is essential as untreated malnutrition increases mortality, with an estimated rate of 10-30% of malnutrition-related deaths (21-25). For HNC patients, losing more than 5% bodyweight increases mortality 1.7 times (26).
Malnutrition is prevalent in HNC patients before they are diagnosed and continues to affect them throughout their lives. It affects three out of four HNC-patients at the time of diagnosis (2, 6, 27). However, only 30-60% of cancer patients at risk of malnutrition receive nutritional support (28). The effect of NIS on nutrition is often underestimated by physicians, patients, and their support systems (24, 28-36). Moreover, social, and public life impairments caused by tumors and treatment has yet to be researched.
Our objective was to narrow the existing gap in understanding, by using a questionnaire directed to HNC survivors. Therefore, we designed a prospective study that captures cancer survivors’ retrospective notions of eating and nutrition after surgery, (chemo)radiotherapy (C)RT, and at full recovery. With etiological mechanisms and risk factors for an excess mortality as subject of current research, the patient’s perspective has long been overlooked. It adds unique insight to the complex problem of malnutrition in HNC patients and provides a foundation for inclusive solutions.
Patients and Methods
Patients. A total of 154 questionnaires were included. They comprised of 50 questionnaires from the nationwide return of the self-help groups and 104 questionnaires submitted by outpatients from a single oncological center. Twenty outpatient questionnaires were excluded due to insufficient data. Inclusion criteria were existing or prior head and neck cancer and age over 18 years. The 47 items enquired patient demography, their course of weight and NIS after treatment and at the time of survey, as well as the quality of supporting structures.
Narrative interviews. The items of the patient reported outcomes (PRO) questionnaire were designed based on three semi-structured narrative interviews facilitated by the Bundesverband für Kehlkopfoperierte e.V. (Federal Association of Laryngectomized). The interviewees were of different age, sex, and time after treatment. The objective was to identify important topics and to map questions from the matrix of information obtained. We conducted three semi-structured narrative interviews. All interviewees were contacted for their consent before analyzing the protocols. The structured part of the interviews contained open-ended questions about treatment and weight loss, nutrition after surgery and oral supplements, additional costs of nutrition, taste changes after treatment, and impact on food intake.
Questionnaire. As a result of the interviews, five main topics emerged: personal data and weight development, food intake, management of nutrition, ongoing cancer symptoms, and the concerns of relatives.
Based on these topics, 47 items were formulated as 25 multiple-choice (MC) questions, five open-ended questions, and 17 five-point Likert scales (LS) illustrated with smiley faces to demonstrate a range of emotion. The results were averaged and presented with standard deviation (SD). The LS were labeled in one direction: e.g. a score of one indicates disapproval, dissatisfaction, or disagreement, whereas five denotes approval, satisfaction, or very good. The content is detailed in Table I.
Summary of the aspects enquired in the questionnaire with quantitative presentation of three main elements: multiple choice (MC), Likert scale (LS) and open question (OQ).
The original questionnaire was reviewed and improved in a pilot group of five patients. The final questionnaire (in German) is available on request to the first author.
Distribution of questionnaires. We distributed 192 questionnaires nationwide across the 15 Federal self-help groups. The response rate was relatively low due to COVID-19 restrictions (N=50; 26%). Furthermore, we recruited data of questionnaires distributed to outpatients from the Südharz Clinic Nordhausen (N=104; 100%). A flow chart displays the data for the respective fields (Figure 1). Ethics. The amended questionnaire prototype obtained the approval of the local ethics committee (Reg.-Nr.: 2020-1857-Bef) for study and questionnaire.
The flowchart illustrates the distribution, return, and how many questionnaires were included for weight-, side effect-, and demographic analysis.
Statistics. The data were collected in Microsoft Excel software (Version 16.21, Microsoft Corporation, Redmond, VA, USA) plotted and descriptively analyzed. Further graphic and statistical analysis was performed via GraphPad Prism 8.1.1. The statistical tests applied were one-way analysis of variance (one-way ANOVA) and Dunnett’s multiple comparisons test, the chi-square (χ2) test, and the independent samples t-test. Results and baseline parameters are reported as ± standard deviation (SD), ranges, and percentages. A two-tailed p-value below 0.05 was considered statistically significant.
Results
Demographic data. Of the 134 questionnaires, 124 were considered for demographic information and were analyzed. The sex ratio was 2.5:1, including 88 male (71%) and 36 female (29%) participants (Figure 2). The age of participants was 65.5±9.4 years, with a range of 0.1 to 31 years post treatment. The mean time after treatment was 6.5±7.2 years. At total of 54/124 HNC patients (44%) were laryngectomized. A detailed overview of patient demographic data is shown in Table II.
Study population pyramid displaying a sex-ratio 2.5 (male): 1 (female) with an increased cancer incidence between 60 and 70 years.
Patient demographics (N=126).
Weight trajectories. A total of 110 participants described their weight course from baseline to the potential endpoints: weight after surgery, (C)RT, and at convalescence.
The mean weight loss after the operation was 7% (p=0.0017), after (C)RT 11% (p<0.001), with a return to baseline weight after convalescence. In addition, the subgroup analysis shows a stratification in weight loss when comparing baseline weight groups: 40-79 kilogram (kg), 80-99 kg, and ≥100 kg (Figure 3).
Weight trajectories measured at baseline weight, after surgery (1), after chemoradiotherapy (2) and current weight (3). Stratified into three baseline weight subgroups: up to 80 kg, from 80 to 99 kg and from 100 kg; the latter experienced significantly more and permanent weight loss as detailed in Table III.
One-way ANOVA analysis showed significantly higher weight loss in people with a baseline weight ≥100 kg. Participants of this subgroup averaged a loss of 9% (p=0.0201) after surgery, and 22% (p<0.0001) after (C)RT, with a permanent deficit of 12% (p=0.0041).
The subgroups 40-79 kg and 80-99 kg showed similar results. At the time of the survey, these subgroups had recovered to baseline weight. The detailed results are displayed in Table III.
Nutrition impact symptoms (NIS). A total of 112 participants, the majority, reported NIS. The comparison between NIS during therapy and NIS at the time of the survey is displayed (Figure 4). More than half of the participants experienced a loss of taste after treatment (N=62; 55%), dry mouth (N=63; 56%), or dysphagia (N=64; 57%). The smallest share was allocated to dental problems (N=37; 33%). At the time of survey, loss of taste (N=24; 21%), dry mouth (N=47; 42%), dysphagia (N=48; 43%), and dental problems (N=24; 21%) remained.
Incidence of nutrition impact symptom(s) compared between time after treatment and at the time of survey with sex split.
The recovery period lasted up to a year after treatment (2.2±1.1). Moreover, some participants (N=15; 13%) reported a readjustment period longer than two years. The occurrence of NIS was stratified according to the treatment applied (Table IV). Concurrent CRT caused significantly more frequent loss of taste (N=26; 74%) (p=0.0461), and dysphagia (N=27; 77%) (p=0.0334). Moreover, participants experienced a higher percentage of long-lasting dysphagia than the average (57% vs. 38%). However, this was not statistically significant (p=0.050). For the subgroup with surgery as a singular treatment, the analysis showed no dental problems (N=0; 0%) (p<0.0031). Furthermore, they experienced less xerostomia (N=3; 16%) (p=0.0011) and loss of taste (N=2; 11%) (p=0.003) during treatment, as well as less permanent xerostomia (N=2; 11%) (p=0.0088).
Nutrition impact symptom subgroup split (N=112).
Nutrition. The modality of surgery had a negative impact on food intake (2.7±1.2 LS) with a tendency to last for months. The food intake was significantly more compromised when (C)RT was applied (2.4±1.1 LS) (p=0.0393).
Overall, the preparation of food in their domestic setting was rated as “good” (4.2±1.0 LS). In detail, nutrition was considered “important” (3.9±0.8 LS) and digestion functioned well (3.8±1.0 LS). Most respondents reported cooking alone at home (N=59; 53%). The same share reported being cooked for by a partner, and 29 by a family member (N=29; 26%). Furthermore, occasional difficulties with seasoning and digesting food (3.2±1.2 LS) were reported.
Quality of life support. In general, participants reported a good mood (3.9±0.7 LS) and ample support from their relations (4.3±0.7 LS). In addition, public reactions to their condition were predominantly positive (3.6±1.3 LS). The averaged LS value regarding the avoidance of restaurants indicated a low degree of social impairment (3.7±1.4 LS). Acceptance within the family scored a significantly higher mean (4.3±1.0 LS) (p=0.0005) compared to acceptance in public.
Counseling. Overall, 110 participants received nutritional counseling during therapy, and 108 after therapy. Throughout therapy, doctors were the most important contact in the healthcare system (N=39; 35%), followed by nutritionist/dietician (N=29; 26%) and nursing staff (N=28; 25%). Similar rates were given for consultation after treatment. The most notable fraction received support from doctors (N=34; 31%), nursing staff (N=29; 26%), and nutritionists/dieticians (N=29; 26%).
When asked which counseling has helped them most after therapy, the majority opted for physicians (N=37; 34%), followed by swallowing therapists (N=25; 23%), and nutritional counselors (N=23; 21%). Some participants wished for more support from doctors, dentists, and dietary specialists (N=39; 35%), followed by social services (N=21; 19%), and self-help networks (N=22; 20%).
Supplements. Only (N=7; 6%) used protein powder. Slightly higher rates were recorded for oral supplements (N=22; 20%), percutaneous endoscopic gastrostomy catheters (N=25; 23%), gastric tubes and parenteral nutrition (N=37; 34%). One third (N=36; 33%) received no nutritional support. Overall, there was no association between the intervention and the weight loss experienced. The average duration of supplementary feeding was several months (2.8±0.9 LS). The average satisfaction with supplementary feeding was moderate to good (3.4±0.9 LS).
Swallowing therapy. The majority (N=75; 68%) never received swallowing therapy. A minority (N=10; 9%) were not interested in swallowing therapy. Some (N=33; 30%) received swallowing therapy and slightly fewer (N=28; 25%) wished for more swallowing therapy. Only five participants (4%) received several series of swallowing treatments.
Concerns of the relatives. Of the 134 respondents, 100 had relatives report their concerns on four subjects. One-third of non-patient respondents (N=33; 33%) reported adjustments in cooking behavior, in particular concerning spices and consistency. A minority of respondents (N=7; 7%) experienced hardships with nutritional support and nutritional supplements. 16 respondents reported (N=16; 16%) social isolation of the patient, giving reasons such as social exclusion and NIS impairments that limited social interactions. A majority (N=54; 54%) reported fears and worries about the disease. The main concerns were tumor recurrence, permanent symptoms after treatment, and anxieties over the future.
Dental care. The average satisfaction with dental care was good (4.0±1.0 LS). A total of (N=47; 43%) felt their dentist advised them well, and (N=34; 31%) stated the dentist was familiar with the issue. However, only (N=17; 15%) were actively supported, while (N=16; 14%) had to address problems themselves. One-fourth (N=30; 27%) stated that dentures and dental restoration were their biggest expense.
Discussion
The incidence of HNC in Germany has gradually increased over the last 20 years (37). Despite recent advances in treatment and medication, such as checkpoint inhibitors and reduced radiation fields, frailty and mortality remain high (31, 33, 38-41). The mechanisms and risk factors for this are still the subject of current research. However, the patients’ perspective on nutrition and malnutrition remain unexamined. Our objective was to investigate those critical gaps in understanding with a survey directed at HNC survivors in which they could report on their experiences. The survey enquired about weight and treatment regimens and how they impacted nutritional intake. Furthermore, we focused on sociocultural implications relevant to the aftermath of HNC. The results of the survey show an elevated risk of malnutrition when baseline weight exceeds 100 kg, an elevated level of NIS in extensive treatments, and social retreat following surgery and (C) RT.
We found that the weight course was significantly impacted by therapy, baseline weight, and NIS. The weight loss averaged between 5.0-10.0 kg after surgery. In contrast to our findings van den Berg et al. described a loss as low as 1.5 kg for HNC patients in a prospective approach (42). However, further research is needed to obtain a more detailed stratification and validate weight loss after HNC surgery.
For the second endpoint, (C)RT, loss was seen in all subgroups had an average drop ranging from 9-22%. In comparison, Farhangfar et al. reported a mean weight loss of 3.9±8.0% for six months after (C)RT, which is similar to the mean weight loss of 3.8% reported by Mallick et al. after radiotherapy (6, 43). Moreover, a multivariant analysis by Simon et al. found a higher risk of malnutrition for patients with a normal body mass index (BMI 18.5–24.9) compared to patients with an abnormal BMI, which contrasts with the patient-reported figures (44). Hence, comparability is limited when other metrics, such as BMI, are used to determine malnutrition.
Lastly, our findings indicate that a habitual weight greater than 100 kg is associated with a permanent weight loss and malnutrition. To our knowledge, there is no referable publication. Therefore, overweight patients seem to be under recognized as malnourished.
Nutrition is vital to patients and various publications suggest a superior prognostic value for survival and complications. Compared to other entities, HNC patients are at a high risk of complications affecting quality of life (QoL), social eating, and social contact (18, 45). In particular, dysphagia induced by radiotherapy leads to severe weight loss, even in lower-staged laryngeal tumors (45). Stewardship of nutritional intake with intensified nutritional interventions can help decrease patient weight loss and therefore reduce risk of complications (40, 46). A qualitative analysis would be a vital addition to our findings, along with further research on how mortality is affected. Thus, validation and extended data can bring insight into weight trajectory risk groups and the connection between weight loss and NIS.
NIS refers to the symptoms experienced by patients which cause malnutrition and impair QoL and quality/frequency of social interaction. German guidelines recommend combined (C)RT for all stages of larynx carcinomas (47). However, HNC describes a heterogenic group of cancers that impair food intake, airway, and communication. Therefore, recommendations vary in other countries and for other HNC cases, usually a case-by-case decision (48). As a result, therapy may vary depending on staging and the form of treatment resulting in a stratified risk for NIS (49).
We identified an increased risk of NIS with concurrent CRT and a gradual increase of NIS from surgery to (C)RT. In addition, loss of taste and dysphagia was significantly more pronounced with CRT displayed in elevated odds ratio (OR). In concordance, surgery as sole treatment had the lowest OR, indicating this gradual increase. The occurrence of NIS is closely connected to weight trajectories. A recent prospective study displayed NIS trajectories and baseline levels (50). NIS impairing oral food intake was associated with weight loss greater than 10%, and NIS gradually decreased until the one-year follow-up.
Furthermore, a systemic review showed that among NIS, oropharyngeal dysphagia is most likely to compromise oral intake (51). Our findings showed that 68% of patients never received any swallowing therapy, and only 4% had multiple treatments. The highest OR (2.140) for dysphagia with (C)RT is consistent with findings from Eisbruch et al. (17, 52, 53). Hence the gap between the low number of interventions and the high incidence of NIS indicates great therapeutic potential.
Beyond the scope of this survey, aspiration and reflux impact patients and are related to dysphagia and problems of the stomatognathic system. With aspiration rates up to (90%) after treatment, there is no natural restitutio ad integrum despite less invasive therapies and innovations in radiation (46, 54).
Efficient management of NIS and a well-maintained dental situation are vital to maintaining nutritional intake for patients. Replacing lost teeth is usually expensive but is key to digestion and communication. Various studies related worse outcomes, higher mortality, and higher staged tumors to low-income (HNC) patients (55-57). Only 15% reported active support and 27% claimed dentures as their biggest expense. The findings suggest a gap in financial aid and should encourage dentists to proactive support cancer survivors.
A targeted therapy, including unilateral irradiation of the neck and excellent communication between the involved departments are key to attenuate malnutrition. Especially the timing of surgery and (C)RT has proven essential to reduce treatment toxicity and NIS (58). Moreover, our results show the high significance of longitudinal patient care by experienced personnel. The resulting therapeutic potential could positively impact mortality, frailty, and the ability to speak, breathe, eat, and swallow (59, 60).
Despite gaps in support and continuing NIS, nutrition and intake were important and overall perceived as “good”. The results reflect undiminished life quality and resilience of HNC survivors despite the hardships of cancer and treatment. However, one-fifth of respondents declared that they avoided restaurants after treatment, revealing social retreat. Furthermore, there were distinct differences between women and men when it comes to cooking. Whereas three-fourths of the women reported cooking alone, only one-fourth of the men did so. Thus, our results show a lack of inclusion in HNC patients’ experiences in Germany. Recent findings by Roick et al. complemented our research, addressing the decline of social support throughout the first year of HNC survivorship (61). In addition, a study conducted by Danker et al. identified a 40% withdrawal rate from conversations. Moreover, 50% were embarrassed, and 87% felt stigmatized because of their tracheostomy (62). The impact and extent of the mentioned disparities deserve further research, possibly by continuous PRO surveys to facilitate a feedback loop for interventions and needs (63).
Almost 100% of respondents received counselling during treatment and two-thirds received nutritional interventions. However, that seems insufficient to avoid critical weight loss. Hence, more proactive counselling and interventions are essential beyond the acute treatment phase (64). Figures from Erickson et al. confirm the gap between need and nutritional support and thus emphasize the immense therapeutic potential of counselling (65). Other studies even suggest implementing dietary counselling before treatment (66).
The figures for dietary counselling were balanced with dietitians and physicians scoring around 30%, respectively. Another study by Maschke et al. found dietitians to be the primary source of consultation (39%), followed by physicians (10%) (67). Those reverse figures could be a result of study design and recruitment but deserve further investigation.
Multilateral professional support is important to their recovery, with an emphasis on the role of the physician as a consultant. The findings emphasize that equilibrium is not attained with the completion of treatment. Keeping malnutrition at bay is a challenge that continues throughout an HNC patient’s life and therefore requires continuous support.
There are limitations to our findings that should be considered. Data collection from patient memories may be biased in both directions. On one side, remembrance of treatment and adaptation to NIS may appear less severe than the initial experience. Conversely, results of weight progression and the NIS may be exaggerated as they are also an inventory of losses in a self-assessment. The extent and direction of the bias can hardly be determined. However, regarding weight progression, the results resemble an overview rather than just a snapshot at a specific time. Moreover, the study design facilitated a broad range of patient years with coherent results for the self-help group participants and outpatients. Therefore, even though we cannot completely exclude a potential bias, a meaningful impact unlikely. The questionnaire is suitable for repetitive testing and can improve outcomes if performed regularly (68).
Furthermore, the impact of COVID-19-related lockdowns that were in place for the duration of our study deserves evaluation. This created a potential limitation as it resulted in a change of methods. The recruitment of more than 50% of participants from a single outpatient consultation center potentially limits the validity of the data for the national comparison.
In addition, the shutdown of self-help groups and daily interaction impacted participants’ lives with yet unknown extent and direction. Worldwide, postponement of surgery and other cancer treatments resulted in delays with unpredictable outcomes. Results were upstaging, increased mortality, morbidity, the likelihood of reoccurrence, and an increase in the psychological burden (69-71).
Lastly, the instrument or method of analysis may be oversimplified. The modified Likert Scales are simple and are ideal for capturing an intuitive response. However, the limited range of emotions connected to an item may indicate inaccuracy in this method. Overall, we consider the inaccuracy of the simplification to be negligible. To address the limitations, conducting a longitudinal, multicenter study could help affirm and expand our findings.
Conclusion
Higher habitual weight indicates higher risk for malnutrition. Extended treatment results in higher NIS rates. On the other hand, the relatively low intervention rates for NIS indicate therapeutic potential and deserve further research to optimize weight loss beyond therapy. PRO questionnaires help evaluate how treatment and disease affect someone’s life as well as provide individualized counseling and nutritional stewardship.
Footnotes
Authors’ Contributions
AC, JB, and JH initiated and designed the study. SW and JB conducted the patient recruitment. AC, JB, and JH performed collection, analysis, and interpretation of data. AC performed the statistical analysis. AC wrote the manuscript, SW, JB, and JH critically revised it. All Authors read and approved the final manuscript.
Conflicts of Interest
The Authors declare that they have no competing interests.
- Received December 30, 2022.
- Revision received January 26, 2023.
- Accepted January 31, 2023.
- Copyright © 2023 International Institute of Anticancer Research (Dr. George J. Delinasios), All rights reserved.
References
In this issue
Jump to section
Related Articles
Cited By...
- No citing articles found.