Needle Catheter jejunostomy: A controlled, prospective, randomized trial in patients with gynecologic malignancy

Nick M. Spirtos, MD, and Samuel C. Ballon, MD
Stanford, California

Back to WCC Published Papers

Sixty patients with gynecologic cancer entered a prospective, randomized study of immediate postoperative feeding. Thirty-three women in the study group received an elemental diet (Vivonex HN) delivered through a needle catheter jejunostomy. Twenty-seven patients in the control group were given standard 5% dextrose and electrolyte solutions. Patients in both groups were stratified according to nutritional status as determined by anthropometric evaluation and levels of serum albumin, total protein, and transferrin. These parameters also were measured at intervals throughout the study. Only one catheter-related complication occurred. Patients in the study group received significantly more calories (p = 0.01) and were better able to maintain serum levels of transferrin (p = 0.05) than those in the control group. An elemental diet administered through the needle catheter jejunostomy effectively maintains postoperative nutrition and is associated with few complications. (Am J Obstet Gynecol 1988; 158: 1285-90)

Key words: Immediate postoperative nutrition, elemental diet, needle catheter jejunostomy

In the 15 years since Delaney et al. 1 described a simplified technique of insertion of a needle catheter jejunostomy tube, there has been renewed interest in this method to provide early postoperative nutritional support. Efficacy in patients requiring esophageal, gastric, or intestinal resection has been reported. 2 Ballon 3 reported the use of needle catheter jejunostomy in patients with gynecologic malignancies undergoing radical pelvic and abdominal operations. Unlike intravenous hyperalimentation, few complications are associated with needle catheter jejunostomy use, no sterilization of formula is necessary, nursing care is minimal, and it is relatively inexpensive. 3,5

Much of the ascribed benefit of enteral feedings has been garnered from uncontrolled studies. 3,5-11 Moreover, of the controlled studies, only Smith et al. 12 randomized patients with cancer. Thus we undertook a prospective, randomized study comparing enteral feedings of Vivonex HN (Norwich Eaton Pharmaceuticals Inc., Norwich, N.Y.) via needle catheter jejunostomy to intravenous dextrose (5%) in lactated Ringer's solution.

Material and methods

Sixty patients with gynecologic malignancies were prospectively randomized into two treatment groups. A computer-generated random number table determined assignment into one of two groups: needle catheter jejunostomy (study patients) or intravenous 5% dextrose (control patients). The study patients had a needle catheter jejunostomy placed at the time of operation in a manner previously described. 3 Postoperatively, the patients in this group were given Vivonex HN, with the infusion started in the recovery room. The concentration and volume were increased according to the schedule in Table I. The control patients randomized to receive intravenous fluids were given 5% dextrose in lactated Ringer's solution with electrolyte supplementation as determined by individual need. All patients continued on their assigned regimen until adequate oral nutritional support was achieved postoperatively.

Table I. Defined diet via needle catheter jejunostomy

OR, Operating Room

Every patient underwent a physical examination. Preoperative anthropometric measurements included usual body weight (in kilograms), percent usual body weight, and triceps skin fold thickness, Karnofsky performance grading was done on each patient. Preoperative tests included a complete blood count with differential and an SMA-18.

Patients were defined as being nutritionally malnourished and classed as "high risk" if the initial serum albumin level was <3.2 gm/dl or if there was an involuntary wight loss >9.9 kg. On days 1,4,7, and the day of discharge, a blood count with differential was measured and SMA-18 tests were performed. Daily levels of serum transferrin, serum glucose, total protein, albumin, sodium, potassium, chloride, and carbon dioxide were determined as were urine glucose and acetone levels. Percent change in body weight was noted as were changes in serum protein, albumin, and transferrin levels, total lymphocyte count, and triceps skin fold thickness.

Statistical analyses were done on total calories as well as on differences between pre-and postoperative measurements for body weight loss (in percent); Karnofsky performance score; total protein, albumin, and transferrin levels; and triceps skin fold thickness. An unbalanced analysis of variance procedure was used with these dependent variables and the class variables: risk group (high versus low) and feeding method (intravenous versus needle catheter jejunostomy) and the interaction between these two.

Table II. Operations

Table III. Preoperative nutritional status

Data are mean ± SD.

Results

The operative procedures performed on the 33 study and 27 control patients were similar (Table II). The preoperative nutritional status was not significantly different between the two groups (Table III). Eight of 33 study patients and seven of 27 control patients had preexisting malnutrition. For the purposes of statistical analysis these patients were classified as high risk.

The patients with a needle catheter jejunostomy received substantially more calories per day than the control group (1342 versus 491 calories; p = 0.01). This appeared to have little objective effect on the difference between the pre- and postoperative nutritional status regardless of risk assignment. The only variables that showed a significant effect of feeding treatment (at p = 0.05) were the total lymphocyte count and serum transferrin level. Both of these were barely significant, with the treatment group differences just exceeding the minimum statistically significant difference. For total lymphocyte count, the control group had a postoperative count 10 cells/ *?* greater than their preoperative count, whereas that in the study group dropped by 454 cells/*?*. The minimum significant difference as computed by a Bonferroni multiple means comparison test was 438 cells/*?*. The study group had postoperative transferrin levels 1 mg/dl higher than preoperative values, whereas those in the control group dropped by 29 mg/dl. The minimum significant difference was 25 mg/dl.

Table IV. Minor complications

Table V. Calories delivered via needle catheter jejunostomy in the postoperative patient

*Mead Johnson & Co., Evansville, Ind.
†Norwich Eaton Pharmaceuticals, Norwich, NY

The hospital stay in the group with a needle catheter jejunostomy was 9.63 +/- 2.8 days and was not significantly different from that in the control group (10.3 +/- 1.77 days). Complications (nausea, vomiting, diarrhea, infection and ileus) were similar in each group (Table IV). There was one catheter complication in the study group. A similar number of patients in each group were treated with postoperative antibiotics. No patients developed peritonitis and two patients in each group developed wound infections.

Comment

It is well established that although gastric emptying is slowed in the immediate postoperative period, the function of the small intestine remains intact. A logical extension of this has been the development of postoperative nutritional support through needle catheter jejunostomy feedings. Our study confirms the findings in other recent reports: that needle catheter jejunostomy is an effective method of supplying calories in the immediate postoperative period (Table V). Complications most often associated with jejunostomy feedings are nausea, vomiting, and abdominal bloating (Table IV). the mainstay of treatment in these cases is the discontinuation of the feedings until the patient becomes asymptomatic. Restarting the infusion at this point rarely results in patient intolerance severe enough to warrant catheter removal. Catheter-related complications have decreased since most surgeons have modified the insertion technique to include suturing of the small bowel serosa to the anterior abdominal wall. 14 Other infrequent but serious complications associated with needle catheter jejunostomy include peritonitis, small bowel ischemia, and obstruction.

Objective evidence supporting the assertion that postoperative nutritional support through needle catheter jejunostomy benefits the patient with cancer is scarce. Recent reports addressing this issue are discordant. In 1979 Yeung et al., 2 in a nonrandomized, controlled study, demonstrated that patients receiving Flexical through a needle catheter jejunostomy better maintained serum protein levels (p < 0.05) and total body weight as compared with control patients. However, there was no difference in hospital stay between the two groups. A year later Hoover et al. 13 confirmed in a randomized, controlled study that postoperative enteral nutrition was associated with significantly improved serum protein preservation (p < 0.001) as well as positive nitrogen balance. However, in a controlled, randomized study, Smith et. al. 12were not only unable to confirm the findings of Hoover and Yeung, but they found that patients nourished via needle catheter jejunostomy had a significantly longer hospital stay (p < 0.001). Only the patients in the study of Smith et. al. had cancer.

Our findings are much more consistent with those of Hoover et al. 13 and Yeung et al. 2 In our study the hospital stay in both groups approximated 10 days. Patients receiving needle catheter jejunostomy feedings (Vivonex HN) were better able to maintain adequate levels of serum proteins than the control patients (lactated Ringer's solution 5%). Measurement of serum transferrin levels bears this out. Not surprisingly, only serum transferrin (half-life = 4 days) and not serum albumin (half-life = 14 days) was indicative of the protein-sparing effects associated with needle catheter jejunostomy feedings.

Whether changes in serum proteins are significant is a more difficult issue to resolve. Some investigators have shown these proteins to be insensitive indicators of nutritional status. 15 Others, including Mullen et al. 16 and Makinski and Ruggiero, 17 have found that decreased total albumin and transferrin levels are associated with increased postoperative complications. Three theories that could explain a lack of change in serum protein levels in patients receiveing nutritional support through needle catheter jejunostomy are: (1)The serum albumin and protein levels are not elevated becaue of an increase in the extracellular fluid compartment; (2) there is increased protein degradation in patients with cancer; and (3) the length of treatment was not sufficient to demonstrate or cause an increase in serum protein levels. Evaluating our data leads us to conclude that perhaps the length of treatment in most cases was not sufficiently long to affect serum protein and albumin levels, the half-lives of which exceed that of transferrin.

We did not find that needle catheter jejunostomy feedings are associated with prolonged hospitalization as reported by Smith et. al. 12 Our treatment and control groups had hospital stays of equal duration. This is an important finding because our study and that of Smith et al. are among the few controlled and randomized pertaining to patients with cancer. Smith et al. point out that part of the increased hospitalization associated with needle catheter jejunostomy may be because there were four patients in their study group who had stays >30 days, compared with only one such patient in the control groups. One of the four patients had a wound dehiscence that was ascribed to the needle catheter jejunostomy feedings and associated abdominal distention. Given our experience with rapidly reaccumulating ascites with resultant abdominal distention after operation for ovarian cancer, we find such an explantion for wound dehiscence unlikely. Based on our findings and these considerations, we find it unlikely that needle catheter jejunostomy is associated with prolonged hospitalization. Review of the data of Smith et al. also reveals that 11 of 25 patients with a needle catheter jejunostomy either had catheter complications or were unable to tolerate the feeding regimen. On the other hand, only 1 of 33 patients in our study had a catheter-related complication and all 9 patients who were symptomatic at one time or another during therapy ultimately tolerated feedings via needle catheter jejunostomy. Obvious differences that might explain such disparate findings include: (1)Smith et al. used 16-gauge catheters whereas we used 14-gauge catheters, and (2) they used a nonelemental diet (Isocal) whereas we used an elemental diet (Vivonex HN). Clearly our findings do not parallel those of Smith et al. Perhaps these differences in materials are responsible for the difference in our studies.

Besides having similar average hospital stays, our study and control groups had a similar number of postoperative complications. Anthropometric evaluation results and most serum protein measurements were not significantly different between the two groups. Only the serum transferrin level (as previously discussed) and the total lymphocyte count were found to differ between the two groups after treatment. A decreased total lymphocyte count, unless associated with energy, has not been previously linked to poor postoperative outcome.

Our study lends support to the assertion that most patients are not adversly affected by short periods of postoperative malnourishment. However, until we are able to accurately predict which patients will be unable to tolerate postoperative malnutrition, needle catheter jejustomy remains a valuable means to provide routine nutritional support during the early postoperative period.

In summary, calories can be delivered enterally with very few complications and high patient tolerance though a needle catheter jejunostomy. Protein preservation as reflected by changes in serum transferrin levels was significantly improved in the treatment group. The cost to the patient is less than comparable regimens given parenterally. Accordingly, needle catheter jejunostomy is a valuable means of providing postoperative nutritional support in the patient with gynecologic malignancies.

References

1. Delany HM, Carneval NJ, Garvey JW. Jejunostomy by a needle catheter technique. Surgery 1973;73:786.

2. Yeung CK, Young GA, Hackett AF, Hill GL. Fine needle catheter jejunostomy--An assessment of a new method of nutritional support after major gastrointestinal surgery. Br J Surg 1979;66:727.

3. Ballon SC. Effective early postoperative nutrition by defined formula diet via needle-catheter jejunostomy. Gynecol Oncol 1982;14:23.

4. Allardyce DB, Groves AC. A comparison of nutritional gains resulting from intravenous and enteral feeding. Surg Gynecol Obstet 1974;139:179.

5. Bower RH, Talamini MA, Sax HC, Hamilton F, Fischer JE. Postoperative enteral vs parenteral nutrition. Arch Surg 1986;121:1040.

6. Andrassy RJ, Dubois T, Page CP, Patterson RS, Paredes A. Early postoperative nutritional enhancement utilizing enteral branched-chain amino acids by way of a needle catheter jejunostomy. Am J Surg 1985;150:720.

7. Burt ME, Stein TP, Brennan MF. A controlled, randomized trial evaluating the effects of enteral and parenteral nutrition on protein metabolism in cancer-bearing man. J Surg Res 1983;34:303.

8. Dunn EL, Moore EE, Jones T. Nutritional support of the critically ill patient. Surg Gynecol Obstet 1981; 153:45.

9. Feldtman RW, Archie JP Jr. A three year experience with needle catheter jejunostomy in a community hospital. Surg Gynecol Obstet 1984;159:23.

10. Hayashi JT, Wolfe BM, Calvert CC. Limited efficacy of early postoperative jejunal feeding. Am J Surg 1985;150:52.

11. Thompson JS, Burrough CA, Hodgson PE. Experience with needle catheter jejunostomy for postoperative nutritional support. Nebr Med J 1983 (Oct):326-30.

12.Smith RC, Hartemink RJ, Hollinshead JW, Gillett DJ. Fine bore jejunostomy feeding following major abdominal surgery: A controlled randomized clinical trial. Br J Surg 1985;72:458.

13. Hoover HC Jr. Ryan JA, Anderson EJ, Fischer JE. Nutritional benefits of immediate postoperative jejunal feeding of an elemental diet. Am J Surg 1980;139:153.

14. Delany HM, Carneval N, Garvey JW, Moss CM. Postoperative nutritional support using needle catheter feeding jejunostomy. Ann Surg 1977;186:165.

15. Forse RA, Shizgal HM. The assessment of malnutrition. Surgery 1980;88:17.

16. Mullen JL, Gerner MH, Buzby GP, Goodhart GL, Rosato EF. Implications of malnutrition in the surgical patient. Arch Surg 1979;114:121.

17. Kaminski MV Jr. Ruggiero RP. Nutritional reassessment: A guide to initiation and efficacy of enteral hyperalimentation. Int Surg 1979;64:33.

Editors' note: This manuscript was revised after these discussions were presented.

Discussion

Dr. Joseph E. Markee, Yakima Washington. In 1974 Bristain et al. 1 stated that up to 50% of surgical patients showed some evidence of nutritional deficiencies. Since then there has been little argument on what actually represents clinically significant malnutrition, and whether nutritional support is cost-effective or of benefit to the patient. 2,3 Although the study of Spirtos and Ballon does not address these issues directly, their prospective, randomized study of immediate postoperative feedings does provide useful information on the following questions: (1) Can needle catheter jejunostomy be safely used in gynecologic/oncology patients undergoing extensive surgery? and (2) can this procedure provide nutritional support in these patients?

First, is needle catheter jejunostomy a safe procedure? In this study, 60 patients were prospectively randomized into two treatment groups. A similar number of patients in each group received operative antibiotics. There were no occurrences of peritonitis and two patients in each group developed wound infections. The incidence of intestinal complications, specifically nausea, vomiting, diarrhea, infection, and ileus, was not higher in the study group than in the control group. Only 1 of 33 patients with a needle catheter jejunostomy developed a catheter complication. The length of hospital stay was the same in the two groups.

Second, can needle catheter jejunostomy provide nutritional support for these patients? When compared with a similar group of patients who received an intravenous solution of 5% dextrose in lactated Ringer's solution, patients managed with needle catheter jejunostomy received significantly more calories. In addition, the study group did have a slight but significantly increased transferrin level. Although there is still controversy regarding the significance of anthropometric immunologic, and biochemical markers in the nutritonal evaluation of surgical patients, there appears to be a correlation between changes in transferrin levels and changes in nitrogen balance.4,5

What is the role for this procedure in gynecologic surgery? In addition to malnutrition and extensive abdominopelvic surgery, other indications have included planned postoperative radiotherapy or chemotherapy. 6 Patients undergoing extended field radiotherapy who are treated with needle catheter jejunostomy after radical pelvic surgery have experienced fewer therapeutic interruptions for bone marrow suppression, nausea, or diarrhea when compared with a control group. 7

As described by Spirtos and Ballon, the use of an elemental diet delivered by small-caliber flexible tube can be a superior alternative to total parenteral nutrition in most patients as long as there is a functioning small bowel. 8

Should needle catheter jejunostomy be used in all patients postoperatively? The authors stated that this procedure remains a valuable means of providing routine nutritional support in the early postoperative period. However, they also raise the problem of how to predict which patients will be unable to tolerate postoperative malnutrition. In looking for other markers besides anthropometric, immunologic, or biochemical values, Allison9 has suggested that voluntary oral intake was directly related to the nutritional status of patients on admission. His group looked at the effect of nutritional support on elderly women admitted to the hospital with a fractured femur. A beneficial effect of supplemental nutrition on mortality and rehabilitation time (defined as the time between operation and independent mobility) was seen only in those patients with a nutritional deficit and insufficient oral intake on admission. 10

I would like to ask Dr. Ballon two questions: A high-risk group of nine patients with preexisting malnutrition, defined by a serum albumin level <3.2 gm/dl or a weight loss of at least 9.9 kg, was included in both the study and control population. Did these high-risk groups differ significantly in terms of postoperative complications, recovery, or biochemical determinations? Second, was there any assessment of the preoperative oral intake in the high-risk groups?

References

1. Bristain BR, Blackburn GL, Hallowell E, Heddle R. Protein status of general surgical patients. JAMA 1974;230:858-60.

2. Sutton G, Karran SJ. The diagnosis of malnutrition: nutritional parameters. Curr Opin Gastroenterol 1985;1:281-7.

3. Diongi R, Dominioni L, Jemos V, et al. Diagnosing malnutrition. Gut 1986;27 (suppl):5-8.

4. Silk DBA. Future of enteral nutrition. Gut 1986;27:116-22.

5. Fletcher JP, Little JM, Guest PK. A comparison of serum transferrin and serum prealbumin as nutritional parameters. J Parenter Enter Nutr 1987;11:144-7.

6. Ballon SC. Effective early postoperative nutrition by defined formula diet via needle-cather-jejunostomy. Gynecol Oncol 1982;14:23-32.

7. Girtanner RE. Preop and postop nutritional support. Contemp Obstet Gynecol 1985;25:153-73.

8. Bower RH, Talamini JA, Sax HC, Hamilton F, Fischer J. Postoperative enteral vs parenteral nutrition. Arch Surg 1986;121:1040-5.

9. Allison SP. Some psychological and physiological aspects of enteral nutrition. Gut 1986;27:18-24.

10. Bastow MD, Rawlings J, Allison SP. Benefits of supplementary tube feeding after fractured neck of femur: a randomized controlled trial. Br Med J 1983;287:1589-92.

Dr. Ballon (Closing). First of all, I would like to acknowledge Dr. Nick Spirtos, who was a fellow with us and is now on the faculty at Stanford, who analyzed all the data for this report.

There were in the high-risk patients no specific differences that we noticed as a subgroup compared with the total group of control and study patients. Unfortunately, we did not assess oral intake in either the high-risk or total groups of patients before their admission to the hospital, other than by simply weighing them and determining the measures of nutritional status as described.

I would like to make two comments. One is that it is very easy to show that one can increase calories, increase nitrogen, and improve serum protein levels in the immediate postoperative period by delivering these kinds of elemental diets. However, it is quite another thing to quantitate and objectively assess the significance of these nutritional gains. Regardless, based on the number of calories, amount of protein, and other measures of nutritional status, it is hard to argue against the use of this technique in patients who are malnourished. Specifically, it has a use in that subgroup of patients who ordinarily would not be candidates for total parenteral nutrition but might, in fact, benefit from some nutritional support.

Finally, until about 3 years ago we never undertook an operation specifically for the purpose of inserting a jejunostomy catheter. However, in patients with endstage ovarian carcinoma who come to the hospital with an obstructive intestine and bulky residual disease, who have failed first-, second-, and third-line therapy, and in whom an operation will be unsuccessful at correcting the intestinal obstruction, I think it makes sense (after a very quick workup) to get those patients to the operating room, make one or two small incisions, and insert a gastrostomy tube and a jejunostomy feeding catheter. I think this makes the difference between 2 or 3 months of real anguish in the hospital or the ability to get these patients home without an intravenous catheter or a tube in their nose, to be cared for by their friends and family in a far more comfortable and cost-effective setting.

Back to WCC Published Papers

Sponsoring Sites                  Terms of Use & Copyright ©2000 - 2007