The Effect of Nasogastric Feeding after Surgery in Patients with Head and Neck Cancers – Retrospective Single Center Experience

Abstract

This study aimed to assess clinical factors that could predict the need for nasogastric feeding after surgery in patients with head and neck cancer (HNC) and evaluate the effect of tube feeding on selected laboratory parameters.

This single-center retrospective study included 153 patients who underwent surgery for HNC. Data on patient and tumor characteristics were collected, along with laboratory measurements. Logistic regression was used to identify the predictors of the need for nasogastric feeding. Laboratory parameters were compared between patients who required nasogastric feeding vs those who did not.

Nasogastric feeding was required in 90 patients (59%). Significant predictors of nasogastric feeding in HNC patients after surgery, which were revealed by univariate regression analysis, included low body mass index (odds ratio [OR] = 0.84), squamous cell carcinoma histology (OR = 8.05), T2 tumor stage (OR = 2.27), red blood cell count (M/µL) (OR = 0.44), hemoglobin levels (g/dL) (OR = 0.80), and mean corpuscular volume (fL) (OR = 1.10). Multivariate analysis showed that low BMI (OR = 0.87) and red blood cell count (M/µL) (OR = 0.32) were prognostic factors for nasogastric feeding. A significant percentage increase in white blood cell count from admission to discharge was noted in patients who required nasogastric feeding vs those who did not (p = 0.003).

Determining factors that predict the need for nasogastric feeding in HNC patients after surgery may support more personalized treatment planning to optimize clinical outcomes.

1. Introduction

Head and neck cancers (HNCs) constitute a diverse group of malignancies that develop in the head or neck region. They are the seventh most common type of cancer worldwide, with 1.1 million new diagnoses each year (1, 2). Approximately 90% of HNCs are squamous cell carcinomas (SCCs), which arise from the epithelial lining of the oral cavity, pharynx, and larynx (3). It is estimated that the incidence of HNC will increase by 30% by 2030 (2). The 5-year survival rate is about 50%, depending on the stage and location of the tumor (4). Tobacco smoking and alcohol consumption, either separately or in combination, are major risk factors for HNC – together they account for 72% of cases (2, 4).

Patients with HNC are at high risk of malnutrition, defined as a reduction in body weight by more than 10% over 6 months and a low body mass index (BMI < 20 kg/m²) (5). Unintentional weight loss was reported in 60% of patients at the time of diagnosis and in 86% after treatment (6). Malnutrition in HNC patients, leading to poor quality of life, increased morbidity, and worse survival rates, is a multifactorial syndrome (7). The location and characteristics of the tumor can impair the swallowing function, leading to insufficient food intake (7). Cancer cachexia, associated with metabolic abnormalities leading to reduced appetite and involuntary muscle and adipose tissue loss, occurs in up to 20.2% of HNC patients before treatment and in 41% of those after chemoradiotherapy (5, 6). Moreover, malnutrition can be caused by poor dietary habits and unhealthy lifestyle at baseline, including excessive alcohol and tobacco use (6, 7). Standard treatment options for HNC (surgery, radiotherapy, and chemotherapy) further worsen the nutritional status of patients due to severe dysphagia (6, 8, 9).

Surgery is commonly used in the treatment of early-stage oral cavity cancers and, to a lesser extent, of pharyngeal and laryngeal cancers (3). Surgical approach is associated with a cure rate of over 80% in patients with small primary cancer and no clinical nodal involvement, but surgical patients may develop dysphagia (3). Surgical tumor resection often leads to altered anatomy and impaired swallowing and chewing function, affecting the oral intake of nutrients (6). Moreover, postoperative complications may occur, such as seroma, hematoma, abscess, fistula, wound infection or dehiscence, and chyle leak, resulting in inflammation and edema compressing or infiltrating the structures involved with swallowing (10). If a patient after recent tumor resection consumes less than 60% of daily calorie intake for more than 7 days, enteral feeding should be considered (8). Enteral nutritional support may reduce the frequency of complications after surgery, including infection. Moreover, it may accelerate wound healing, thus improving the quality of life and reducing hospitalization time (6, 8).

Both nasogastric and gastrostomy feeding tubes are effective for enteral nutrition delivery in HNC patients when oral intake is not adequate or safe (9, 11). Nasogastric feeding, which is the standard technique, is generally recommended for a short period of up to 4 wk to avoid patient discomfort and potential risks of more prolonged use, such as lesions to the nasal wing, chronic sinusitis, gastroesophageal reflux, and aspiration pneumonia (11, 12). Percutaneous endoscopic gastrostomy is rather used for long-term enteral feeding (5). There is an ongoing debate in the literature on whether a prophylactic or reactive approach to enteral feeding is more effective in patients with HNC (13, 14). A nasogastric tube is more common in the reactive setting (14). Patients who are fed with a nasogastric tube are more likely to have it removed earlier than those who are fed by percutaneous endoscopic gastrostomy (15).

Numerous patients treated for HNC require nutritional support, which is often delivered via a nasogastric tube. This study aimed to identify clinical factors that predict the need for nasogastric feeding after surgery in HNC patients and evaluate the effect of tube feeding on selected laboratory parameters.

2. Methods

2.1. Patients and Setting

In this study, we performed a retrospective chart review of patients with HNC to identify predictors of the need for nasogastric feeding after surgery. We enrolled patients who underwent surgery for HNC between 2007 and 2017 in the Department of Cranio-Maxillofacial Surgery of Military Institute of Medicine – National Research Institute in Warsaw, Poland. We included all consecutive patients who had data on clinical nutrition available. During the study, a decision on nasogastric feeding was based on the Subjective Global Assessment, including 5 components of medical history (dietary intake, weight change, gastrointestinal symptoms, functional capacity, metabolic requirement) and 3 components of a brief physical examination (signs of subcutaneous fat and muscle wasting, malnutrition-related edema, and ascites). Also, patients with weight loss of > 10% in the prior three months before surgery were qualified for nasogastric feeding. We have used Harris-Benedict formula to estimate an individual’s basal metabolic rate.

A nasogastric tube was inserted for at least a week after surgery through which enteral nutrition was administered. The diet was determined individually for each patient and considered the appropriate caloric and macro- and micronutrient amounts. Daily requirements were determined according to a standard protocol on the basis of the results of laboratory tests, BMI, and weight loss in the last 3 months. All patients met the estimated calorie requirements.

An ethics approval was not required because of the retrospective study design.

2.2. Data Collection

In 2022 we collected data on patient and tumor characteristics, including sex, age, BMI, tumor site, histopathological tumor type and grade, tumor stage according to the TNM (tumor, node, metastasis) staging system, and type of lymphadenectomy. The histopathological tumor type was classified as SCC or non-SCC. Tumor site was classified as the tongue or the floor of the mouth (most common) vs. other sites. The following laboratory parameters were measured: white blood cell count (WBC), red blood cell count (RBC), hemoglobin, mean corpuscular volume (MCV), hematocrit, serum albumin, and total serum protein.

2.3. Statistical Analysis

Baseline data were assessed descriptively and compared between patients who required and did not require nasogastric feeding. Univariate and multivariate logistic regression was used to identify the predictors of the need for nasogastric feeding. We ran univariate regression models for the variables available at baseline; the multivariate model included all significant variables. For all logistic regression models, patients with unmeasurable tumor and nodal stages (Tx, Nx) were excluded. Linear regression was used to compare the percentage change in laboratory parameters from admission to discharge between patients who received and those who did not receive nasogastric feeding. The comparisons between groups were adjusted for age, sex, BMI, length of hospital stay, and type of lymphadenectomy. For all outcomes assessed in the linear regression models, we estimated marginal means with 95% confidence intervals in both groups. Laboratory values at discharge were analyzed descriptively and shown in Table 3. A P value of less than 0.05 was considered significant. All calculations were completed in the R software (v. 4.2.2).

3. Results

3.1. Characteristics of Study Patients

Of the 153 patients included in the study, 90 (58.8%) required nasogastric feeding. All patients had dysphagia with severity dependent on tumor location and size. The mean age of patients, proportion of women, and tumor site did not differ significantly between patients who required and those who did not require nasogastric feeding (Table 1). Patients fed by a nasogastric tube were characterized by significantly longer hospital stay, lower BMI, and a higher frequency of SCC than the remaining patients (Table 1). In addition, they had a significantly higher rate of G2 and T2 tumors and less often underwent selective lymph node dissection (Table 1). On admission, patients who required nasogastric feeding had significantly worse RBC parameters (RBC, hemoglobin, hematocrit, MCV) and significantly lower albumin levels (Table 1). WBC and total serum protein levels at baseline did not differ significantly between groups (Table 1).

Table 1. Clinical and demographic characteristics of patients depending on the need for nasogastric feeding.

Parameter	No nasogastric feeding (n = 63)	Nasogastric feeding (n = 90)	p-value^1
Sex, n (%)			0.811^2
Female	26 (42)	36 (40)
Male	36 (58)	54 (60)
Age (years)	65 (58, 74)	67 (60, 78)	0.355^3
Hospital stay (days)	4 (3, 8)	10 (7, 15)	<0.001^3
BMI	27.6 (24.5, 31.2)	23.0 (21.2, 26.2)	<0.001^3
Histopathology, n (%)			<0.001^2
SCC	40 (63)	84 (93)
Other	23 (37)	6 (6.7)
Tumor grade, n (%)			0.047^4
G1	4 (10)	6 (6.9)
G2	29 (72)	77 (89)
G3	7 (18)	4 (4.6)
Tumor site, n (%)			0.873^2
Tongue/floor of the mouth	23 (37)	34 (38)
Other	40 (63)	56 (62)
T stage, n (%)			<0.001^4
T1	20 (36)	27 (30)
T2	17 (31)	52 (58)
T3	8 (15)	4 (4.4)
T4	3 (5.5)	7 (7.8)
Tx	7 (13)	0 (0)
pN stage, n (%)			<0.001^2
N0	0 (0)	44 (49)
N1	9 (16)	25 (28)
N2	12 (21)	16 (18)
Nx	35 (62)	4 (4.5)
M stage, n (%)			<0.001^4
M0	30 (54)	2 (2.9)
M1	1 (1.8)	0 (0)
Mx	25 (45)	68 (97)
Lymphadenectomy, n (%)			<0.001^4
RND	3 (12)	5 (5.9)
MRND	7 (27)	55 (65)
SND	16 (62)	11 (13)
Other	0 (0)	14 (16)
RBC (M/µL)	4.67 (4.29, 5.04)	4.26 (3.88, 4.65)	<0.001^3
WBC (K/µL)	7.37 (5.46, 9.47)	7.75 (5.82, 9.38)	0.466^3
Hemoglobin (g/dL)	14.30 (13.20, 14.90)	13.35 (12.20, 14.38)	0.004^3
Hematocrit (%)	42.0 (39.0, 45.0)	40.0 (36.3, 43.8)	0.022^3
MCV (fL)	90.0 (87.0, 92.0)	94.0 (90.0, 96.8)	<0.001^3
Serum albumin (g/dL)	4.60 (4.12, 4.90)	4.30 (3.92, 4.50)	0.004^3
Total protein (g/dL)	7.10 (7.00, 7.50)	7.30 (6.82, 7.77)	0.850^3

Data are presented as median (interquartile range) unless otherwise indicated.

¹Significant differences at a P value <0.05.

²Pearson’s Chi-squared test.

³Wilcoxon rank sum test.

⁴Fisher’s exact test.

Abbreviations: BMI, body mass index; MCV, mean corpuscular volume; RBC, red blood cell count; MRND, modified radical neck dissection; RND, radical neck dissection; SND, selective neck dissection; WBC, white blood cell count.

3.2. Predictors of the Need for Nasogastric Feeding

Univariate regression analysis revealed that low BMI (odds ratio [OR] = 0.84), SCC histology (OR = 8.05), T2 tumor stage (OR = 2.27), RBC (OR = 0.44), hemoglobin levels (OR = 0.80), and MCV (OR = 1.10) were significant predictors of the need for nasogastric feeding. Multivariate analysis showed that low BMI (OR = 0.87) and RBC (OR = 0.32) were prognostic factors for nasogastric feeding. Detailed data are presented in Table 2.

Table 2. Logistic regression predicting the need for nasogastric feeding.

	Univariable logistic regression				Multivariable logistic regression
Parameter	n	OR	95% CI	P-value^1	OR	95% CI	P-value^1
Sex	152
Female		—	—		—	—
Male		1.08	0.56, 2.09	0.811	1.63	0.62, 4.34	0.318
Age (years)	153	1.02	0.99, 1.04	0.179	1.01	0.97, 1.04	0.714
BMI	151	0.84	0.77, 0.90	<0.001	0.87	0.79, 0.96	0.005
Histopathology	153
Other		—	—		—	—
SCC		8.05	3.21, 23.2	<0.001	3.25	0.96, 11.6	0.060
Tumor grade	127
G1		—	—
G2		1.77	0.43, 6.65	0.402
G3		0.38	0.06, 2.15	0.283
Tumor site	153
Other		—	—
Tongue/floor of the mouth		1.06	0.54, 2.07	0.873
T stage	138
T1		—	—		—	—
T2		2.27	1.03, 5.08	0.044	2.47	0.94, 6.66	0.068
T3		0.37	0.09, 1.35	0.144	0.20	0.03, 1.05	0.067
T4		1.73	0.42, 8.79	0.466	1.12	0.19, 9.50	0.908
RBC (M/µL)	143	0.44	0.24, 0.77	0.007	0.32	0.13, 0.74	0.010
WBC (K/µL)	143	1.05	0.92, 1.20	0.466
Hemoglobin (g/dL)	143	0.80	0.64, 0.98	0.037
Hematocrit (%)	143	0.95	0.88, 1.01	0.134
MCV (fL)	143	1.10	1.03, 1.18	0.006
Serum albumin (g/dL)	120	0.57	0.22, 1.27	0.208
Total protein (g/dL)	123	0.84	0.43, 1.55	0.579

¹Significant differences at a P value <0.05.

Abbreviations: CI, confidence interval; OR, odds ratio; others, see Table 1.

3.3. The Effect of Tube Feeding on Hematological and Biochemical Parameters

In a linear regression analysis, the significant percentage increase in WBC from admission to discharge in patients requiring nasogastric feeding vs. the remaining patients was observed (p = 0.003) (Fig. 1). RBC, hemoglobin, and hematocrit levels decreased significantly from admission to discharge in both groups, but there were no significant differences between groups (Fig. 1). No changes in MCV were observed in any of the groups (Fig. 1). Serum albumin and total protein levels decreased significantly from admission to discharge, without significant differences between groups (Table 3). Differences in blood parameters at discharge depending on the need for nasogastric feeding are presented in the Table 3.

Figure 1. Percent change in blood parameters from admission to discharge. Point estimates show marginal means from linear regression models, and error bars show 95% confidence intervals. P-values are for comparisons between patients who required vs. those who did not require nasogastric feeding. All comparisons were adjusted for age, sex, body mass index, length of hospital stay, and type of lymphadenectomy. Significant changes in each group were considered significant when error bars did not cross the reference line of 0.

Table 3. Differences in blood parameters at discharge depending on the need for nasogastric feeding.

Parameter	No nasogastric feeding (n = 63)	Nasogastric feeding (n = 90)	P-value^1
RBC (M/µL)	3.54 (3.11, 4.15)	3.31 (2.91, 3.91)	0.070
WBC (K/µL)	8.5 (6.3, 10.9)	11.4 (9.0, 15.4)	<0.001
Hemoglobin (g/dL)	10.85 (10.00, 12.40)	10.34 (9.32, 11.90)	0.107
Hematocrit (%)	32.0 (29.0, 36.8)	31.3 (27.7, 36.0)	0.483
MCV (fL)	89.0 (86.2, 94.0)	94.0 (90.0, 98.0)	<0.001
Serum albumin (g/dL)	3.70 (3.30, 4.10)	3.20 (2.60, 3.60)	<0.001
Total protein (g/dL)	6.15 (5.75, 6.47)	5.20 (4.50, 5.90)	<0.001

Data are presented as median (interquartile range).

¹Significant differences at a P value <0.05; Wilcoxon rank sum test.

Abbreviations: see Table 1.

4. Discussion

In the current study, we assessed the results of tube feeding and identified the predictive factors for nasogastric feeding in HNC patients after surgery based on patient and tumor characteristics as well as changes in blood parameters from admission to discharge.

Cancer patients have a high risk of malnutrition due to disease-related cachexia and anti-cancer therapy. Cancer-related cachexia results in decreased appetite and weight loss. Pro-inflammatory processes often lead to insulin resistance, and loss of body fat and muscle mass. The nutritional intervention aimed at the prevention of further loss of body weight should be integrated part of the management of patients with HNC, especially after surgery (8, 16, 17). Tumor location in patients with HNC is related to difficulties in food intake. Also, all surgical procedures even small, cause severe pain and dysphagia which altogether cause the inability to intake food by oral route (6).

Most research to date has focused on clinical predictors for tube feeding in patients receiving radiotherapy and/or chemotherapy (18–23). This is because the risk of malnutrition increases with multimodality treatment toxicity, resulting in significant weight loss and increased need for enteral feeding (15).

To our knowledge, this is the first study examining blood parameters as potential predictors of the need for nasogastric feeding after surgery in patients with HNC. A similar study by Plonowska et al. (24) evaluated nasogastric feeding after transoral robotic surgery for oropharyngeal SCC, along with demographic, clinical, histopathologic, and treatment risk factors. Patients with oropharyngeal cancer required nasogastric feeding less often than patients with HNC in our study (23.2% vs. 58.8%). This may be due to the application of minimally-invasive surgical treatment by Plonowska et al. (24). According to the investigators, increased tumor size and concurrent bilateral neck dissection were significant predictors of nasogastric feeding. In contrast to our study, T2 tumor stage was not significantly associated with enteral feeding (24).

In a retrospective study by Mangar et al. (25), conducted on 160 patients referred for radical radiotherapy for HNC (of whom 61 after surgery), 31% of patients required enteral feeding (including two-thirds before radiotherapy). Similar to our study, low BMI and RBC were found to be significantly associated with the need for enteral nutrition (25). Other prognostic factors included performance status, advanced cancer, weight loss before treatment, age, and smoking.

Wermker et al. (26) identified clinical parameters to calculate risks as the reduced nutritional state against the possible complications of prophylactic feeding tube placement in surgically treated patients, such as BMI, the size and localization of the tumor, lymph node affection, resection of the root of the tongue or the oropharyngeal region, and performance of a neck dissection. Predictors of the need for long-term postoperative tube feeding, as reported by various investigators, included stage III/IV cancers, flap reconstruction, pharyngectomy, current tracheotomy, tumor size and location, heavy alcohol use, preoperative weight loss, and chemotherapy (27–29). More factors associated with severe dysphagia often necessitating the use of a feeding tube were described in a critical review by Koyfman and Adelstein (14). However, their data were mainly related to patients undergoing radiotherapy or chemotherapy.

In our study, patients who required nasogastric feeding had worse blood-related parameters (RBC, hemoglobin, hematocrit) at baseline than those without such a requirement. Nevertheless, enteral nutrition led to an improvement of these parameters, and no significant differences were observed between groups at discharge. A significant improvement in hemoglobin and hematocrit levels after enteral feeding was also reported by Kakkar et al. (30). In our study, a percent decrease in serum albumin and total protein from admission to discharge was comparable between groups. It is possible that the difference would have been more significant if patients did not receive enteral nutrition (30, 31). On the other hand, there was a significant increase in WBC in patients fed via a nasogastric tube. This may be explained by inflammatory response to nasogastric tube placement (32).

The main limitation of our study is its retrospective design. Moreover, the number of subjects corresponding to other tumor sites than the tongue or the floor of the mouth (most common) was too small to analyze separately. In our study, prophylactic enteral feeding was not used in all patients but only in those who had worse blood parameters after surgery. This may have influenced the results of between-group comparisons.

Furthermore, conducting long-term follow-up studies could provide insights into the impact of nasogastric feeding on patient outcomes and determine blood parameters as potential predictors of the need for tube feeding after surgery in HNC patients.

5. Conclusions

Nutritional assessment in HNC patients is important for decision-making about treatment. In our study, we identified simple predictors of the need for nasogastric feeding in patients after surgery, such as low BMI, SCC histology, T2 tumor stage, RBC, hemoglobin levels, MCV, and serum albumin. Our findings may support more personalized treatment planning and allow physicians to inform their patients in advance about the possible need for feeding tube placement during therapy. Moreover, our results may facilitate the identification of high-risk patients in whom prophylactic enteral nutrition should be considered already at the time of their initial surgery to optimize quality-of-life and clinical outcomes.

Author Contributions

Conceptualization, A.Ch., D.J.; methodology, A.Ch., K.Z.; software, A.Ch. K.Z.; validation, A.Ch., D.J.; formal analysis, A.Ch.; investigation, A.Ch.; resources, A.Ch.; data curation, A.Ch., K.Z. P.G.; writing—original draft preparation, A.Ch.; writing—review and editing, A.Ch., D.J., K.Z., J.N., P.G.; visualization, A.Ch., J.N., P.G.; supervision, A.Ch; D.J.; project administration, A.Ch.; funding acquisition, A.Ch. All authors have read and agreed to the published version of the manuscript.

Informed Consent Statement

Due to the retrospective nature of the study, informed consent of the patients was not required because the study analyzed anonymous clinical data.

Institutional Review Board Statement

The study was conducted in accordance with the Declaration of Helsinki. An ethics approval was not required due to the retrospective nature of the study.

Acknowledgments

Editorial assistance was provided by Proper Medical Writing, Warsaw, Poland.

Funding

This research received no external funding.

Conflicts of Interest

The authors declare no conflict of interest.

Data Availability Statement

The data that support the findings of this study are available from the corresponding author, upon reasonable request.