Enteral nutritional support by gastrostomy tube in children with cancer

Enteral nutritional support by gastrostomy tube in children with

cancer

Victor M. Aquino, MD, Carla B. Smyrl, RD, LD, Russell Hagg, BS, Kim M, McHard, MSN, RN, Laurel Prestridge, MD, a n d Eric S. Sandier, MD From the Department of Pediatrics, University of Texas Southwestern Medical Center at Dallas, and the Center for Cancer and Blood Disorders, Children's Medical Center of Dallas We examined the use of gastrostomy tubes in malnourished children with cancer as part of our ongoing efforts to improve their supportive care. Patients were examined on the basis of percentage of weight loss and percentage of desirable body weight. Twenty-five patients underwent gastrostomy tube placement followed by aggressive enteral nutritional support. Gastrostomy tubes were placed at a mean of 3,5 months (range, 0.3 to 8 months) after diagnosis; mean weight loss had been 10.1% (range, to 2 I%) of desirable body weight. There were no immediate postoperative complications. Gastrostomy tube feedings were well tolerated by all patients. All children gained or maintained weight, and 60% of the severely malnourished children returned to a desirable body weight after an average of 4.9 months (range, I to 13 months). Weight gain averaged 12.9% (range, to 45.4%) of desirable body weight. The most common complications were 38 episodes of inflammation at the gastrostomy tube site during periods of severe neutropenia, which were treated successfully with topically or orally administered antibiotics, and 13 episodes of cellulitis, which required intravenously administered antibiotics. The infection rate was 1.58 episodes per 1000 days of use compared with a rate of 5.0 per 1000 days previously reported with total parenteral nutrition. The monthly costs of gastrostomy tube nutrition support were 9% of those associated with use of total parenteral nutrition. Gastrostomy tube use in children with cancer is a safe, effective, and cost-effective method of reversing malnutrition. Further investigation with larger numbers of patients is warranted. (J PEDIATR 1995; 127:58-62)

Protein-energy malnutrition is commonly seen in children with cancer, both at the start and during continuation of intensive chemotherapy treatment programs. 1-s The causes of PEM include decreased intake, increased energy requirements, maiabsorption, changes in taste, nausea and vomiting, Supported by National Institutes of Health T32 Training Grant CA09640, the Children's Cancer Fund of Dallas, and Weekend to Wipe Out Cancer, Dallas, Tex. Submitted for publication Nov. 11, 1994; accepted Feb. 8, 1995. Reprint requests: Eric S. Sandier, MD, Department of Pediatrics, University of Texas Southwestern Medical Center at Dallas, 5323 Harry Hines Blvd., Dallas, TX 75235-9063. Copyright © 1995 by Mosby-Year Book, Inc. 0022-3476/95/$3.00 + 0 9120164039

58

mucositis, mechanical gastrointestinal problems, and environmental and psychologic factors, t, 3, 8 Malnourished children with cancer may have greater delays in chemotherapy,7, 9, to increased toxic effects of chemotherapy, 1°-14 and GT PEM TPN

Gastrostomytube Proteinenergy malnutrition Total parenteral nutrition

I

[

I

impaired cellular immunity as measured by skin test antigens and T-cell subset analysis, t5 Total parenteral nutrition and enteral feeding through nasogastric tube have been the most common methods of nutritional support in this population. Both are effective in

The Journal of Pediatrics Volume 127, Number 1

reversing malnutrition, restoring immune competence, and reducing delays in chemotherapy. 14 However, several conceres continue to be raised about the potential harm resulting from the use of nutritional support in patients with cancer. Increased tumor growth has been seen in some animal models16,16a, 17but has not been shown in any clinical trials.6,16 Another major concern has been the complications associated with TPN, especially catheter-related bacteremia. 18'18~ 19 The American College of Physicians has concluded that TPN is not effective and may even be harmful in malnourished patients with cancer.18 Poor compliance, psychologic distress, and difficulties in maintaining tube placement in younger children are frequently seen with the use of nasogastric tubes. 4 Enteral feedings through a gastrostomy tube have been effective in reversing the malnutrition seen in both tumorbearing mice 2° and children with head and neck rhabdomyosarcoma. 21 However, GT feedings have not been well studied in severely malnourished children with cancer. Theoretically, GT use could avoid some of the problems associated with TPN, including (1) the high cost of TPN solutions, (2) the need to monitor blood chemistry values during TPN use, and (3) the increased risks of bacteremia and septicemia related to central venous catheters. Therefore we evaluated the safety and efficacy of GT feedings in severely malnourished children with cancer. METHODS Patient selection. Between October 1990 and September 1994, all children with newly diagnosed cancer at the Children's Medical Center of Dallas were examined by a multidisciplinary nutrition team at the time of diagnosis and then on a monthly basis. Severity of malnutrition was evaluated by assessment of the patient's weight as a percentage of desirable body weight and by the percentage of weight loss. Patients were designated as severely malnourished if weight loss was more than 10% of the baseline weight, or if the patient's weight was less than 90% of the 50th percentile for the child's height on a standardized growth scale. 22 If no weight gain occurred despite several weeks of a high-protein, high-calorie diet and commercially available nutritional supplements, the patient was then examined for initiation of nutrition support. Enteral nutrition was the preferred route of supplementation. However, TPN was used for patients with a nonfunctional gastrointestinal tract or as an attempt to improve nutritional risk before surgica] placement of a GT. Patients with head and neck rhabdomyosarcoma who were scheduled to receive radiation and intensive chemotherapy received a GT immediately in anticipation of severe mucositis. Gastrostomy tube placement and use. The route of GT placement was selected by the primary care physician in consultation with a pediatric surgeon, pediatric gastroenter-

Aquino et aL

59

ologist, or both. Gastrostomy tubes were placed surgically through a Stamm gastrostomy with the use of a de Pezzer catheter 23 or by percutaneous endoscopy12327 Button GT placement was considered approximately 2 months after the initial gastrostomy. The button is a simple, skin-level, nonrefluxing device designed for long-term enteral feedings. For feeding, a small cap at the skin level is opened and an adapter is inserted to attach the tubing to a pump. 25 After tube placement an individualized program of nutritional support was implemented to reverse the malnutrition. Oral calorie intakes were estimated and tube feeding schedules were designed to meet the recommended dally allowance for calories. Commercially available polymeric enteral formulas (PediaSure, Ensure Plus, or TwoCal HN) were infused for 8 to 16 hours to allow freedom from the pump during waking hours. No extra vitamin or mineral supplements were added to the formulas. In addition to tube feedings, the nutritional care plan included individual counseling to encourage oral intake. Once calorie counts showed positive oral caloric intake with adequate weight maintenance, the patient was weaned from GT feeds. The total days of GT use were calculated for each patient. The GT was removed when chemotherapy was completed and the patient was maintaining adequate weight by oral intake, or if the patient had multiple infections requiring hospitalization for intravenously administered antibiotic therapy. Gastrostomy tube care. All patients and their parents were instructed on GT care by the nutrition team's nurse practitioner. For 2 weeks after surgery the dressing was changed dally and half-strength hydrogen peroxide was used to clean the skin surrounding the GT. Thereafter the site was cleaned with chlorhexadine (pHisoHex) soap. For patients with a button GT, the parents were advised to clean the area with soap dally and rotate the button 90 degrees. If redness appeared around the tube, parents were instructed to change to half-strength hydrogen peroxide followed by application of povidone-iodine (Betadine) and mupirocin (Bactroban) twice daily to the site. If the inflammation worsened, the patients were advised to have the site examined by a physician, who would then prescribe orally or intravenously administered antibiotics as needed. Cost analysis. Seven children receiving GT feedings and seven receiving TPN from the same home health care company were selected for analysis of monthly charges. Factors included in the analysis of monthly charges include charges for laboratory monitoring, nutrient support, and supplies. These charges were analyzed on the basis of total charges per calorie of nutritional support delivered. RESULTS Patient characteristics. Forty-six children with cancer received intensive outpatient nutritional support from Octo-

60

Aquino et al.

The Journal of Pediatrics July 1995

Table I. Characteristics of 25 children with cancer who received enteral feedings via gastrostomy tube

Table II. Results of enteral feedings through GT in 25 children with cancer

Characteristic

Age (yr): mean (range) Sex (No.) Male Female Diagnosis (No.) Rhabdomyosarcoma Osteosarcoma Acute lymphocyticleukemia Ewing sarcoma Brain tumor Wilms tumor Neuroblastoma Reasons for GT placement Severe malnutrition Head and neck tumor Bilateral vocal cord paralysis

10.2 (2-21) 9 16 7 5 5 3 3 1 1 20 4 1

ber 1990 to September 1994 (Table I). Twenty-five children, 2 to 17 years of age, had GTs placed after a mean body weight loss of 10.1% (range, up to 21%). Twenty patients had GTs placed because of severe malnutrition that developed during therapy for the malignancy. Five additional patients underwent gastrostomy before receiving any antineoplastic therapy; four had rhabdomyosarcoma of the head and neck and had a GT placed in anticipation of severe mucositis, and one patient had bilateral vocal cord paralysis and dysfunctional swallowing after resection of a large ependymorea. Twenty-one severely malnourished children received alternate nutritional support; 16 received TPN, and five received enteral feedings through a nasogastric tube. Indications for use of home parenteral nutrition included parental refusal of nasogastric or gastrostomy tube placement (seven), chemical pancreatitis (two), presence of a large abdominal tumor (one), severe Fanconi syndrome induced by ifosfamide (one), and malfunctioning gastrointestinal tract (five). Five patients, including one child awaiting a solid organ transplant, received nutritional support through a nasogastric tube because of the expected short duration of necessary nutritional support. Insertion of GT. Eighteen patients had a tube placed surgically and seven by percutaneous endoscopy at an average of 3.5 months (range, 0.3 to 8 months) after diagnosis. All patients tolerated GT placement well, and no postoperative complications were observed. No difference in outcome was seen between the two placement routes. Enteral feedings were initiated 24 to 48 hours after GT placement. Of the 25 • patients, 21 had a gastrostomy button inserted at an average of 8.7 weeks (range, 3.8 to 14.0 weeks) after gastrostomy. Four patients refused the gastrostomy button. No surgical

Mean ± SE (range)

Time from diagnosis to GT placement (mo) Duration of GT use (too) %Desh'able body weight at diagnosis %Desirable body weight at GT placement %Desirable body weight at GT removal %Desirable body weight gained

3.5 ± 0.4 (0.3-8.0) 10.5 ± 1.0 (1.0-22.0) 96.2 ± 2.2 (76-137) 87.0 - 1.9 (77-124) 98.9 +- 2.1 (82-130) 12.9 ± 2.3 (0-45.4)

complications resulted from gastrostomy button placement. Results of enteral feedings through GT. The efficacy of GT use is summarized in Table II. For all 25 patients, GTs were in place for a mean of 10.5 months (range, 1 to 22 months); these patients gained an average of 12.9% (range, up to 45.4%) of their desirable body weight. No patient lost weight during GT use. In the 20 children who received GTs because of malnutrition, 12 (60%) reached more than 95% of desirable body weight after a mean of 4.9 months (range, 1 to 13 months). All eight patients who did not reach at least 95% desirable body weight had weight gain averaging 7.2% of their desirable body weight. Among these eight patients, two children died while receiving therapy, two had the GT removed because of infection, and four had the GT removed at the completion of chemotherapy. The five patients who had GTs placed before the initiation of chemotherapy or radiation, or both, maintained or gained weight during therapy. Of the 25 children who received GT, two patients still have them in place. Eighteen patients have had the GT removed, four because of recurrent episodes of cellulitis or GT site breakdown, and 13 after adequate weight gain or because chemotherapy was completed, or both. One patient had poor tolerance of enteral feedings. Five patients died with the GT in place, four of relapse of their disease and one of hemorrhagic pancreatitis. Nutritional support charges. Monthly charges calculated on the basis of charges per kilocalorie showed that GT feedings were administered at 9% of the cost of TPN (p <0.001, Student t test; Table III). Complications. Complications resulting from prolonged placement of a GT were common but usually not serious. Gastrostomy site inflammation, defined as redness around the tube site, was seen on 38 occasions in 14 patients and treated at home with either topical or oral administration of antibiotics. A local infection requiring hospitalization and intravenous antibiotic therapy occurred 13 times in nine pa-

The Journal of Pediatrics Volume 127, Number 1

Aquino et al.

61

Table Ill. Charge analysis for GT feeding versus parenteral nutrition Route GT

No. of patients Age (yr) Weight (kg) Kcal/day Kcal/mo Nutrients (monthly charge) Supplies (monthly charge) Laboratory"studies* Total monthly charges Cost/kcalt

TPN

7 7.1 (3-15) 21.3 (9-45) 1474 (1080-1920) 44,220 (32,400-57,600) 348 (276-408) 665 (582-844) 84 1098 (952-1255) 0.03

7 9,2 (2.5-14) 27.7 (10-45) 1426 (830-2670) 42,789 (30,000-80,100) 11,714 (5100-16,223) 3155 (1293-4873) 312 15,181 (1567-21,408) 0.35

Values (exceptthose for age, number of laboratorystudies, and cost per kilocalorie)are expressedas mean (range). *Routinelaboratorystudiesfor GT includedmonthlymeasurementsof albumin,prealbumin,and transferrin;routinelaboratorystudiesfor TPN includedweekly chemistry panel analysisand twice-monthlyprealbumin/albuminratio determination. ?p <0.001 (Student t test).

tients. Blood culture results were negative in all but one child, in whom Escherichia coli was isolated from the blood but not from the gastrostomy site during an episode of severe neutropenia. All episodes of cellulitis occurred during periods of severe neutropenia (absolute neutrophil court, <250 x 106 cells/L [< 250 cells/ram3]). A variety of organisms were recovered from surface cultures of the tube site: Staphylococcus epidermidis (13), Staphylococcus aureus (four), Candida albicans (11), Enterococcus (four), Corynebacterium (five), c~-hemolytic streptococcus (five), and Candida krusei (one). Many of these isolates were thought to represent contamination from skin flora and were not found in other cultures. During a total of 8236 days of GT use, we observed an incidence of 1.58 severe infection episodes per 1000 days of GT use. Gastrostomy tube occlusion occurred in two patients and required replacement of the tubing. Transient diarrhea occurred in four patients. All episodes of diarrhea resolved after the rate of feedings was decreased, followed by readvancement at a slower rate.

DISCUSSION Severe cachexia is commonly seen in children and adolescents with cancer, both at diagnosis and during treatment. Several studies have suggested that children with PEM who are given nutritional supplementation either orally, enterally, or with TPN have fewer treatment delays and have improvement in immune status. 5, 8, 13, 14 Use of TPN can be effective in reversing the malnutrition associated with cancer and its treatment but may result in multiple complications. All children tolerated GT placement well. Theoretical risks of gastrostomy placement, including poor wound healing, skin breakdown, and bleeding around the gastrostomy site, were not seen.

As previously reported,2t early intervention with enteral feeding through gastrostomy was effective in preventing the severe malnutrition commonly associated with the treatment of head and neck rhahdomyosarcoma. All patients with severe malnutrition showed reversal of PEM and good weight gain. Although 40% did not reach ideal body weight, four of these eight patients had inadequate trials. These data are similar to the previously reported observations of children receiving TPN, in which TPN failed to reverse malnutrition in 20% to 36% of malnourished children with cancer.lO, 12,28, 29 Nutritional supplementation through a GT offers significant cost savings over TPN because of the lower costs of enteral formulas and because much less frequent monitoring of blood chemistry values is required. However, the difference in overall charges must take into account the charges associated with GT placement. These charges (approximately $5000) are similar to the charges associated with placement of a central venous line. However, most patients already have a central venous fine, and this additional charge would not be encountered in those patients receiving TPN. Although complications of enteral feedings through GT were frequent, most were not life threatening. The majority were due to local inflammation around the GT site during periods of neutropenia and were treated successfully with topically or orally administered antibiotics. Hospitalization for intravenously administered antibiotics was occasionally required, but bacteremia was not seen. The rate of infection, 1.58 per 1000 days of gastrostomy use, compares favorably with the rate of 5.0 infections per 1000 days reported for TPN through a central venous Catheter. 19 All episodes of cellulitis were also associated with fever and neutropenia, and it is unclear whether these patients would have needed admission had they not had the GT. However, four patients required

62

Aquino et al.

removal of the GT because of recurrent episodes of cellulitis. W e are currently experimenting with the use of different combinations of topically applied antibacterial and antifungal creams to reduce the infection rate. The case of hemorrhagic pancreatitis was thought to be related to chemotherapy and not to GT placement or feedings. Other serious and life-threatening complications associated with parenteral nutrition, such as subclavian thrombosis and pulmonary emboli, 18 are not a concern with GT use. W e conclude that enteral feedings through a GT are effective in maintaining weight when placed before delivery of antineoplastic therapy to children with head and neck tumors, and in reversing malnutrition in children with weight loss during an intensive chemotherapy regimen. Furthermore, enteral nutritional support is less costly than parenteral nutrition support. Whether enteral feedings through a GT are superior to TPN and whether other subgroups of patients would benefit from GT placement at diagnosis remain to be determined in future clinical trials. REFERENCES 1. Coates TD, Richard KA, Grosfeld JL, Weetman RM. Nutritional support of children with neoplastic diseases. Surg Clin North Am 1986;66:1197-212. 2. van Eys J. Nutrition in the treatment of cancer in children. J Am Coil Nutr 1984;3:159-68. 3. Mauer AM, Burgess JB, Donaldson SS, et al. Special nutritional needs of children with malignancy: a review. JPEN J Parenter Enteral Nutr 1990;14:315-24. 4. Rickard KA, Grosfeld JL, Coates TD, Weetman R, Baehner RL. Advances in nutrition care of children with neoplastic diseases: a review of treatment, research, and application. J Am Diet Assoc 1986;86:1666-76. 5. Rickard KA, Detamore CM, Coates TD. Effect of nutritional staging on treatment delays and outcome in stage IV neuroblastoma. Cancer 1983;52:58%98. 6. Rickard KA, Grosfeld JL, Kirksey A, Ballantine TV, Baehner RL. Reversal of protein-energy malnutrition in children during treatment of advanced neoplastic disease. Ann Surg 1979; 190:771-81. 7. Carter P, Carr D, van Eys J, et al. Nutritional parameters in children with cancer. J Am Diet Assoc 1983;82:616-22. 8. Souba J. Nutritional care of the individual with cancer. Nutr Clin Pract 1988;3:175-82. 9. Ghavimi F, Shils ME, Scott BF, Brown M, TamaroffM. Comparison of morbidity in children requiting abdominal radiation and chemotherapy, with and without total parenteral nutrition. J PEDIATR1982;101:530-7. 10. Hays DM, Merritt RJ, White L, Ashley J, Siegel SE. Effect of total parenteral nutrition on marrow recovery during induction therapy for acute nonlymphocytic leukemia in childhood. Med Pediatr Oncol 1983;11:134-40. 11. Balducci L, Little DD, Glover NG, Hardy CS, Steinberg MH. Granulocyte reserve in cancer and malnutrition. Ann Intern Med 1983;98:610-11. 12. van Eys J, Copeland EM, Cangir A, et al. A clinical trial ofhyperalimentation in children with metastatic malignancies. Med Pediatr Oncol 1980;8:63-73.

The Journal of Pediatrics July 1995

13. Rickard KA, Loghmani ES, Grosfeld JL, et al. Short- and long-term effectiveness of enteral and parenteral nutrition in reversing or preventing protein-energy malnutrition in advanced neuroblastoma: a prospective randomized study. Cancer 1985;56:2881-97. 14. Obama M, Cangir A, van Eys J. Nutritional status and an thracychne cardiotoxicity in children. South Med J 1983; 76: 577-8. 15. Hughes WT. Malnutrition. In: Patrick CC, ed. Infections in immunocompromised infants and children. New York: Churchill Livingstone, 1992:329-33. 16. Cameron IL, Pavlat WA. Stimulation of growth of a transplantable hepatoma in rats by parenteral nutrition. J Natl Cancer Inst 1976;56:597-602. 16a. Steiger E, Oram-Smith J, Miller E, Kuo L, Vars HM. Effect of mutation on tumor growth and tolerance of chemotherapy. J Surg Res 1975;78:455-66. 17. Daly JM, Copeland EM, Dudrick SJ. Effects of intravenous nutrition on tumor growth and host immunocompetence in malnourished animals. Surgery 1978;84:655-8. 18. American College of Physicians. Parenteral nutrition in patients receiving cancer chemotherapy. Ann Intern Med 1989; 110: 734-5. 18a. Christensen ML, Hancock ML, Gattuso L, et al. Parenteral nutrition associated with increased infection rate in children with cancer. Cancer 1993;72:2732-5. 19. Flynn PM. Vascular access device infections. In: Patrick CC, ed. Infections in immunocompromised infants and children. New York: Churchill Livingstone, 1992:289-98. 20. Stallion A, Foley-Nelson T, Chance WT, Zhang F, Fischer JE. Parenteral vs. enteral nutrition in tumor beating rats. JPEN J Parenter Enteral Nutr 1994;18:148-53. 21. Michael CA, Moertel CL, Schwenk WF, Telander RL, Smithson WA. Nutrition via gastrostomy in children undergoing therapy for rhabdomyosarcoma of the head and neck [Abstract]. Pediatr Res 1989;25:154A. 22. Hamill PV, Drizd TA, Johnson CL, Reed RB, Roche AF, Moore WM. Physical growth: National Center for Health Statistics percentiles. Am J Cfin Nutr 1979;32:607-29. 23. Ganderer MW, Ponsky JL, Izant RJ Jr. Gastrostomy without laparotomy: a percutaneous endoscopic technique. J Pediatr Surg 1980;15:872-5. 24. Stellato TA, Gauderer MW. Percutaneous endoscopic gastrostomy in the cancer patient. Am Surg 1988;54:419-22. 25. Ganderer MW. Gastrostomy techniques mad devices. Surg Clin North Am 1992;72:1285-98. 26. Stiegmann GV, Goff JS, Silas D, Pearlman N, Sun J, Norton L. Endoscopic versus operative gastrostomy: final results of a prospective randomized trial. Gastrointest Endosc 1990; 36:1-5. 27. George J, Crawford D, Lewis T, Shephard R, Ward M. Percutaneous endoscopic gastrostomy: a two-year experience. Med J Aust 1990;152:17-9. 28. Rickard KA, Foland BB, Detamore CM, et al. Effectiveness of central parenteral nutrition versus peripheral parenteral nutrition plus enteral nutrition in reversing protein-energy malnutrition in children with advanced neuroblastoma and Wilms' tumor: a prospective randomized study. Am J Clin Nutr 1983; 38:445-56. 29. Donaldson SS, Wesley MN, Ghavimi F, et al. A prospective randomized clinical trial of total parenteral nutrition in children with cancer. Med Pediatr Oncol 1982;10:129-39.