Percutaneous endoscopic gastrostomy tube feeding in patients with head and neck cancer.

We performed a retrospective chart review of 55 patients who had been treated for head and neck cancer to evaluate the complication rate associated with percutaneous endoscopic gastrostomy (PEG) for nutritional support. We found that complications occurred in only 6 patients (10.9%); 3 of these patients (5.5%) had granulation around the site of the PEG tube, and 1 each experienced a tube malfunction, migration of the PEG tube, and leakage around the tube. No wound infection or other complication associated with PEG tubes was noted. Of the 6 complications, 3 occurred in 31 patients who underwent PEG tube placement at the time of tumor resection, and 3 occurred among 16 patients who received a PEG tube postoperatively. We conclude that placement of a PEG tube is a safe method of providing nonoral nutritional support for patients with head and neck cancer.

Introduction

Patients with head and neck cancer are often nutritionally depleted. Factors that contribute to this depletion include dysphagia, odynophagia, cachexia associated with cancer, and the morbidity associated with surgery, chemotherapy, and radiation therapy. These factors can contribute to a 10% or greater loss of baseline body weight during treatment. (1)

Traditionally, placement of a nasogastric tube has been the primary feeding modality for these patients. (2) Although nasogastric tubes are effective, they are associated with complications notably, laryngeal irritation and persistent gastroesophageal reflux.

The percutaneous endoscopic gastrostomy (PEG) tube is an alternative to the nasogastric tube. Its use was first described by Gauderer et al in 1980. (3) Reported complication rates during PEG have been highly variable, ranging from 5 to 40%. (4-12) In 2001, Mekhail et al concluded that the risks of PEG are too high in patients with head and neck cancer, and they called for further investigations to determine its role. (7)

In this article, we describe our study of the incidence and type of complications associated with PEG in patients who were treated for head and neck cancer at an urban teaching hospital.

Patients and methods

We performed a retrospective review of the charts of 55 patients--44 men and 11 women, aged 43 to 83 years (mean: 61[+ or -])--with squamous cell carcinoma of the head and neck who had undergone a PEG procedure between Jan. 1, 1995, and Aug. 30, 2001. All PEGs had been performed in the Department of Surgery at University Hospital, which is affiliated with the New Jersey Medical School in Newark.

In addition to recording specific complications, we noted other factors such as tumor location and stage, the time of PEG tube placement relative to the time of tumor resection, and the time of the onset of complications. All PEGs had been performed in the operating room with the patient under general anesthesia. The standard pull method of tube placement was used with one modification: the tube was inserted into the abdomen through a 3-cm incision rather than through a standard puncture opening.

Results

Lesion site. The most common site of the malignant lesion (20.0% of all cases) was the base of the tongue (table).

Type of complications. Complications occurred in only 6 patients (10.9%). The most common complication was the formation of granulation tissue around the PEG tube, which occurred in 3 patients (5.5%). One patient (1.8%) experienced a tube malfunction, another experienced migration of the tube, and 1 experienced leakage around the tube. No wound infection or other complication associated with PEG was noted.

Time of tube placement. Thirty-one patients underwent PEG tube placement at the time of tumor resection (i.e., intraoperatively). In all but 1 of these patients, the tube was inserted immediately preceding resection. Three of these patients experienced complications--2 developed granulation tissue and 1 experienced tube migration (the tube was subsequently readjusted).

Sixteen patients underwent tube placement postoperatively, and 3 experienced complications--1 had the tube malfunction, 1 developed granulation tissue, and 1 experienced leakage.

Of the remaining 8 patients, 1 underwent tube placement preoperatively and 7 were treated nonsurgically. None experienced a complication.

Time of onset. The onset of the 3 cases of granulation tissue occurred at 3, 8, and 73 weeks following PEG. The tube malfunction, the migration, and the leakage occurred at 4 weeks, 6 weeks, and 19 weeks, respectively. All but 1 of these complications occurred within 5 months of tube placement.

Tumor stage. Tumor stage had been recorded for 54 of the 55 patients. Thirty-four patients had a stage IV tumor, 10 had a stage III tumor, 4 had a stage II tumor, and 6 had a stage I tumor. The 3 cases of granulation tissue all occurred in patients with a stage IV tumor. The tube malfunction occurred in a patient with a stage III tumor, the migration occurred in a patient with a stage IV tumor, and the leakage occurred in a patient with a stage III tumor.

Length of hospital stay. The overall mean length of hospital stay was 18 days ([+ or -]16), including 14.6 days ([+ or -]12.3) for patients with a tongue-base tumor, 18.5 days ([+ or -]10.2) for patients with a glottic tumor, and 17.5 days ([+ or -]15) for patients with a tonsillar tumor.

Discussion

The optimal method of providing nonoral nutritional support for patients with head and neck cancer is unclear. Both long-term nasogastric tubes and PEG tubes have been used. PEG tubes offer potential advantages over nasogastric tubes in that their use can result in a decrease in the length of hospital stay (a 61% decrease among patients with tongue-base, laryngeal, and tonsillar lesions, according to one study (13), an increase in patient comfort, (14) improvement in nutritional status, (15) and a more favorable cosmetic outcome? (16)

On the other hand, the use of PEG tubes has been reported to possibly promote a longer period of nonoral feeding secondary to the deconditioning of the muscles of deglutition. (7) Also, concerns have been raised about premature removal, (5) wound infection at the site, (5,8-10) and other complications, such as leakage, aspiration, and peritonitis. Less common complications include tumor implantation, (11,12) gastric perforation, gastric bleeding, and gastrocolic fistula.

In 1999, Walton reported a major complication rate of 22.5% and a minor complication rate of 17.5%. (5) The primary major complication was premature tube removal and the primary minor complication was wound infection. In our study, the complication rate (10.9%) was relatively low, and none of these complications included premature tube removal or wound infection.

Wound infection has been identified as the most common complication by several investigators. (8-10) We believe that the lack of wound infections in our study may be attributable to the fact that the incision made in the abdominal skin of these patients was larger than usual. Bacteria are less likely to become trapped in larger incisions.

Twelve cases of implantation of tumor at the PEG site have been reported. (11,12) In these cases, tumor may have been seeded in the abdominal wall when the catheter was pulled through the aerodigestive tract.

We did not compare our patients to a matched group of patients with nasogastric tubes, but a comparison with the findings of others may be instructive. For example, the mean length of stay for patients with tongue-base lesions in our study was 14.6 days ([+ or -]12.3); Gibson and Wenig (13) reported that in patients with tongue-base tumors, the mean length of stay was 42 days for those who were fed through a nasogastric tube (and 26 days for those fed through a PEG tube). The corresponding figures for patients with tonsillar lesions were 17.5 ([+ or -]15), 48, and 19 days.

We conclude that the use of PEG tubes is a safe method of delivering nonoral nutritional support for patients with head and neck cancer and that the complication rate is acceptable.

Table. Distribution of primary tumor sites

Tumor site              n

Base of the tongue      11
Glottis                 8
Floor of the mouth      7
Supraglottis            4
Oropharynx              4
Tonsil                  4
Cheek                   3
Hypopharynx             3
Retromolar area         2
Subglottis              2
Border of the tongue    1
Hard palate             1
Mandibular gingiva      1
Maxillary sinus         1
Midface                 1
Nasopharynx             1
Pharynx                 1

References

(1.) Newman LA, Vieira F, Schwiezer V, et al. Eating and weight changes following chemoradiation therapy for advanced head and neck cancer. Arch Otolaryngol Head Neck Surg 1998;124:589-92.

(2.) Selz PA, Santos PM. Percutaneous endoscopic gastrostomy. A useful tool for otolaryngologist--head and neck surgeon. Arch Otolaryngol Head Neck Surg 1995; 121:1249-52.

(3.) Gauderer MW, Ponsky JL, Izant RJ, Jr. Gastrostomy without laparotomy: A percutaneous endoscopic technique. J Pediatr Surg 1980;15:872-5.

(4.) Gibson SE, Wenig BL, Watkins JL. Complications of percutaneous endoscopic gastrostomy in head and neck cancer patients. Ann Otol Rhinol Laryngol 1992;101:46-50.

(5.) Walton GM. Complications of percutaneous gastrostomy in patients with head and neck cancer--an analysis of 42 consecutive patients. Ann R Coil Surg Engl 1999;81:272-6.

(6.) Saunders JR. Jr., Brown MS, Hirata RM, Jaques DA. Percutaneous endoscopic gastrostomy in patients with head and neck malignancies. Am J Surg 1991;162:381-3.

(7.) Mekhail TM, Adelstein DJ, Rybicki LA, et al. Enteral nutrition during the treatment of head and neck carcinoma: Is a percutaneous endoscopic gastrostomy tube preferable to a nasogastric tube? Cancer 2001;91 : 1785-90.

(8.) Fox VL, Abel SD, Malas S, et al. Complications following percutaneous endoscopic gastrostomy and subsequent catheter replacement in children and young adults. Gastrointest Endosc 1997;45:64-71.

(9.) Gutt CN, Held S, Paolucci V, Encke A. Experiences with percutaneous endoscopic gastrostomy. World J Surg 1996;20:1006-8.

(10.) Cunliffe DR, Swanton C, White C, et al. Percutaneous endoscopic gastrostomy at the time of tumour resection in advanced oral cancer. Oral Oncol 2000;36:471-3.

(11.) Douglas JG, Koh W, Laramure GE. Metastasis to a percutaneous gastrostomy site from head and neck cancer: Radiobiologic considerations. Head Neck 2000;22:826-30.

(12.) Potochny JD, Sataloff DM, Spiegel JR, et al. Head and neck cancer implantation at the percutaneous endoscopic gastrostomy exit site. A ease report and a review. Surg Endosc 1998;12:1361-5.

(13.) Gibson S, Wenig BL. Percutaneous endoscopic gastrostomy in the management of head and neck carcinoma. Laryngoscope 1992; 102: 977-80.

(14.) Verhoef MJ, Van Rosendaal GM. Patient outcomes related to percutaneous endoscopic gastrostomy placement. J Clin Gastroenterol 2001;32:49-53.

(15.) Lees J. Nasogastric and percutaneous endoscopic gastrostomy feeding in head and neck cancer patients receiving radiotherapy treatment at a regional oncology unit: A two year study. Eur J Cancer Care (Engl) 1977;6:45-9.

(16.) Wicks C, Gimson A, Vlavianos P, et al. Assessment of the percutaneous endoscopic gastrostomy feeding tube as part of an integrated approach to enteral feeding. Gut 1992;33:613-16.

From the Division of Otolaryngology--Head and Neck Surgery (Dr. Baredes and Mr. Behin) and the Department of Surgery (Dr. Deitch), New Jersey Medical School, University of Medicine and Dentistry of New Jersey, Newark.

Reprint requests: Daniel Behin, Division of Otolaryngology--Head and Neck Surgery, New Jersey Medical School, 90 Bergen St., 8th Floor, Newark, NJ 07103. Phone: [phone]; tax: [phone]; e-mail: [e-mail]

Originally presented as a poster at the Eastern Section meeting of the Triological Society; Jan. 24-26, 2003; Boston.