APPLIED NUTRITIONAL INVESTIGATION

Nasoenteric Feeding Tubes in Critically IllPatients (Fluoroscopy Versus Blind)

Guillermo Huerta, MD, and Vinod K. Puri, MDFrom the Department of Critical Care Services, Providence Hospital, Southfield,

Michigan, USA

Numerous complications have been encountered with small-bore nasoenteric feeding tubes, some poten-tially life threatening. Patients particularly at risk are those with anatomic abnormalities, debilitation, orneurologic impairment. Fluoroscopy has been reported to be a safe, efficacious modality for the placementof these tubes. Thirty critically ill patients were studied to assess caloric delivery, costs, and complicationsassociated with both fluoroscopically and blindly placed feeding tubes. All patients had either a trache-ostomy or an endotracheal tube. They were randomized to group A (fluoroscopy) or group B (blind).Caloric delivery was greater in group A patients on days 1 through 5, with statistically significantdifferences on days 1 through 4. The mean daily calories per patient over the study period was 1135696 and 6626 110 (mean6 SEM) in groups A and B, respectively (P , 0.01). Costs were similar inboth groups. The most frequent problems encountered were difficult insertion, tubes requiring replace-ment, and failure to intubate the duodenum. We conclude that critically ill patients intubated eitherendotracheally or with tracheostomy should have nasoenteric feeding tubes placed with the guidance offluoroscopy. Nutrition 2000;16:264–267. ©Elsevier Science Inc. 2000

Key words: nasoenteral feeding tubes, fluoroscopy, placement

INTRODUCTION

Nutritional support is an integral part of patient care in the inten-sive care unit. In the presence of a functional gastrointestinal tract,enteral nutrition is the preferred method of feeding.1

Nasoenteric feeding tubes are generally safe and effective.However, several complications such as nutrient pneumonitis,pneumothorax, nutrient peritonitis, and pulmonary hemorrhagehave been documented.2–4 Many of these problems have been dueto difficult tube insertion and uncertain placement. Patients atincreased risk are those with conditions such as tracheostomy,endotracheal intubation, cardiomegaly, neurologic impairmentsecondary to trauma, cerebrovascular accident, meningitis, severemalnutrition, congestive heart failure, and renal failure.1,5 Theseconditions are frequently encountered in critical care patients.

Prager et al. reported a high uncomplicated success rate innasoenteric tube placement with fluoroscopic guidance,6 and wehave frequently encountered problems with the blind insertion ofthese tubes in our Critical Care Center. Therefore, we prospec-tively studied the two methods in critically ill patients, with addi-tional observations about caloric delivery and costs.

MATERIALS, METHODS, AND PATIENT POPULATION

Entry into the study required that patients have either a tracheos-tomy or endotracheal tube, a functional gastrointestinal tract, anda high likelihood that their stay in the Critical Care Center would

be of sufficient duration to require enteral feeding for a minimumof 2 d to amaximum of 5 d.

Thirty critically ill but hemodynamically stable patients wereenrolled and randomized into two groups. Group A comprised 17patients (including 2 cross-overs) with a fluoroscopically placedenteral feeding tube (Flexiflo enteral feeding tube, 8 F, RossLaboratories, Columbus, OH, USA), and group B comprised 15patients with blind placement. Both groups were comparable as tosex, age, and clinical problems (Table I).

Group A patients were transported to the Special ProcedureRoom, where the feeding tubes were inserted under fluoroscopicguidance with the use of intravenous metoclopramide in 10-mgaliquots to a maximum of 20 mg. Postprocedure abdominal x-rayswere obtained in successful placements. Enteral feedings wereordered shortly thereafter.

Group B patients had their tubes inserted at the bedside. Ab-dominal x-rays were taken after lying the patient on the right sidefor 2 h. If the duodenum was reached, feeding was initiated;otherwise, a follow-up x-ray was obtained 10 h later, with feedinginstituted if duodenal intubation occurred. A third abdominal x-raywas ordered at 24 h after insertion if the previous two insertionsdid not show the feeding tube in the duodenum. At this point,whether or not duodenal intubation had occurred, feeding wasinitiated, provided that gastric placement was satisfactory. All tubemanipulations were based on the physician’s judgment.

Daily abdominal x-rays were obtained in all patients to docu-ment tube position, up to 5 d. Additional x-rays were ordered ifaspiration or tube dislodgement was suspected and after reposi-tioning or replacement. Feeding tubes were replaced by the orig-inal method of insertion during each study period.

Nutrition was titrated as follows: 50 cc/h of half-strength Isocal(Mead-Johnson, Evansville, IN, USA) on day 1 that was increasedto full strength on day 2, 75 cc/h on day 3, and 100 cc/h on day 4,provided that the rate or concentration was tolerated. Daily caloricintake was recorded, as were complications, tube repositioning, orreplacement, and x-rays were obtained. Cost comparisons werebased on patient charges.

This work was carried out at Mt. Carmel Mercy Hospital, Detroit, Mich-igan, during Dr. Huerta’s Fellowship in Critical Care Medicine.

Correspondence to: Vinod K. Puri, MD, Department of Critical CareServices, Providence Hospital, 16001 West Nine Mile Road, Southfield,MI 48037, USA.

Date accepted: Dec. 15, 1999.

Nutrition 16:264–267, 2000 0899-9007/00/$20.00©Elsevier Science Inc., 2000. Printed in the United States. All rights reserved. PII S0899-9007(99)00307-X

We and a staff radiologist independently interpreted all x-rays.Statistics are based on Student’st test and the chi-square methodof analysis.

RESULTS

Duodenal intubation was accomplished in 15 of 17 (88.2%) groupA patients versus 3 of 15 (20%) group B patients (P , 0.001).Feeding tubes reaching the duodenum remained there for theduration of the study period in both groups. The tubes in patientnumbers 4 and 9 of group B were noted in the duodenum at 10.25and 11 h after insertion, respectively. The second tube in patientnumber 13 was noted in the duodenum at 2.25 h. Fluoroscopicallyplaced tubes remained patent for 4.36 0.27 d, whereas thoseplaced blindly were functional for 3.16 0.4 d (P , 0.02).Feedings were initiated 2.26 4.8 h and 29.66 0.35 h afterinsertion in groups A and B, respectively (P , 0.01).

Caloric delivery for group A patients was greater on days 1through 5 (Fig. 1), with statistically significant differences on days1, 2, 3, and 4 (P , 0.01, 0.01, 0.05, and0.05, respectively). Themean daily caloric delivery per patient over the study period isshown in Figure 2. Group A patients received an average of11356 96 calories per day, and those in group B received 6626110 (P , 0.01).

Complications and problems that were encountered are listed inTable II. The most serious was aspiration in patient number 4 ingroup B. Tube replacement was required in 6 of 13 (46%) groupB patients and in none of the group A patients (P , 0.01). Thetrachea was intubated in 3 patients in both groups. Difficulty ofinsertion was encountered in 3 of 17 (18%) group A patients andin 6 of 15 (40%) group B patients. Although this problem wasencountered more than twice as often in group B, it was notstatistically significant. Patient number 4 in group A did notreceive enteral feeding because of a suspected gastric outlet ob-struction detected during fluoroscopy. In patient number 6, ingroup B multiple attempts to pass the feeding tube into the stom-ach were unsuccessful. With subsequent cross-over into group A(patient number 13), duodenal intubation was also unsuccessful,but enteral feeding was instituted because the tip of the tube laynear the region of the pylorus. Patient number 8 in group B wasextubated and placed on oral intake when, with removal of thetube, a knot was found near its distal end.

Cost comparisons are based on the patient charges. The averagecost per patient during the study period was similar for groups Aand B at $438.006 9 and $476.006 29, respectively. Nosignificant differences were noted.

DISCUSSION

Enteral alimentation is the preferred method of nutritional supportin patients with a functional gastrointestinal tract.3 General guide-lines for the insertion of nasoenteric feeding tubes have beenestablished for several years.7 Nevertheless, complications con-

FIG. 1. Comparison of caloric intake between two methods of tube place-ment. For 5 d, endoscopically placed tubes allowed significantly highercaloric intake.

FIG. 2. Mean daily caloric intake was significantly higher in patients withendoscopically placed tubes.

TABLE I.

CLINICAL CHARACTERISTICS OF PATIENTS IN GROUPS AAND B

Clinical characteristics Group A (fluoroscopy) Group B (blind)

Age (mean6 SEM) 456 6 566 5Age, range 21–78 18–84Sex (male/female) 13/4 9/6Neurologic (non-traumatic) (%) 1/17 (6) 3/15 (20)Closed head injury (%) 8/17 (47) 6/15 (40)Postoperative (%) 4/17 (24) 5/15 (7)Miscellaneous (%) 4/17 (24) 5/17 (33)ICU mortality (%) 4/17 (24) 3/15 (20)

ICU, intensive care unit.

NASOENTERIC FEEDING TUBES IN CRITICALLY ILL PATIENTS 265

tinue to be frequent.1,2–6 Ease of insertion cannot be relied uponfor proper placement. Postinsertion x-rays do not guarantee correctplacement. The use of fluoroscopy has recently been reported toenhance placement of these tubes, particularly with regard topostpyloric intubation.2

Our postpyloric intubation rate was far greater with fluoros-copy. This finding is consistent with that of Prager et al.6 andLevenson et al.8 published results comparing postpyloric intuba-tion rates of weighted versus non-weighted nasoenteric tubesplaced blindly. One hundred seventy-three patients were enrolled,but 58 did not complete the 3-d study period due to tube occlu-sions, dislodgment, and other problems. Their success rate was57% for weighted and 67% for non-weighted tubes. In our blind-placement group, duodenal intubation occurred only 20% of thetime, which is consistent with the finding of Prager et al. (15%).6

The severity of illness in our study population may account for thedifference in results from those of Levenson et al.8 Nevertheless,an 88.2% postpyloric rate with fluoroscopy appears greater thanthat achieved in the blind-placement groups in the study by Lev-enson et al.8

Caloric delivery was significantly greater in group A patients.We attributed this to earlier institution of feeding, fewer compli-cations, longer tube duration, and less frequent replacement. In-terruptions of feeding in both groups were due to diagnostictesting, e.g., nuclear scans, departmental x-rays, and surgicalinterventions.

Initial costs appeared higher in group A, but with repeated tubereplacement and manipulations in group B, costs were found to besimilar. Our study population was at increased risk of complica-tions due to the presence of endotracheal tubes or tracheostomiesin addition to their underlying diseases.

Feeding increases gastric organisms.9 Reports have implicatedgastric colonization as a source of nosocomial pneumonia inmechanically ventilated patients.10–12 The mechanism appears tobe retrograde colonization of the oropharynx, with subsequentpulmonary aspiration and infection.13–16It seems only logical that

steps to reduce this occurrence, such as postpyloric alimentation,may alter morbidity and mortality in the intensive care unit.Incidence of nosocomial pneumonia in a small randomized studywas 10.5% with gastric feeding versus none with jejunal tubefeeding.

In conclusion, we feel that whenever possible fluoroscopyshould be used to place nasoenteric tubes in critically ill patients,particularly those on mechanical ventilation. This method appearsto be safe, cost effective, well tolerated, and, at least initially,facilitates a greater caloric delivery. Zaloga has suggested othermethods for bedside insertion of postpyloric tubes.17

REFERENCES

1. Heymsfield SB, Bethel RA, Ansley JD, et al. Enteral hyperalimentation: analternative of central venous hyperalimentation. Ann Intern Med 1979;90:63

2. Bohnker BK, Arliman LE, Hoskins WJ. Narrow bore nasogastric feeding tubecomplication. A literature review. Nutr Clin Pract 1987;2:203

3. Balogh GJ, Adler SJ, Vandervoude J, et al. Pneumothorax as a complication offeeding tube placement. AJR 1983;141:1275

4. McDanal JT, Wheeler DM, Ebert J. A complication of nasogastric intubation:pulmonary hemorrhage. Anesthesiology 1983;59:356

5. Theodore AC, Frank JA, Ende J, et al. Errant placement of nasoenteric tubes: ahazard in obtunded patients. Chest 1984;86:931

6. Prager R, Laboy V, Venus B, et al. Value of fluoroscopic assistance duringtranspyloric intubation. Crit Care Med 1986;14:151

7. Gording AM Jr. Enteral nutritional support—guideline for feeding tube selectionand placement. Postgrad Med 1981;70:155

8. Levenson R, Turner WW, Dyson A, et al. Do weighted nasoenteric feeding tubesfacilitate duodenal intubation? JPEN 1988;12:135

9. Pingleton SK, Henthorn DR, Lui C. Enteral nutrition in patients receivingmechanical ventilation. Multiple sources of tracheal colonization include thestomach. Am J Med 1986;80:827

10. Montecalvo MA, Stiger KA, Farber HW, et al. Nutritional outcome and pneu-monia in critical care patients randomized to gastric versus jejunal tube feedings.Crit Care Med 1992;20:1377

TABLE II.

GROUP B (BLIND): PATIENT PROBLEMS AND OUTCOMES

Age/sex Clinical problems Outcome

61/F Restrictive/obstructive lung disease pneumonia, coagulopathy Survived59/M Subarachnoid hemorrhagic, alcoholic liver disease Survived18/M Closed head injury, hemorrhagic contusion fractured right femur and left humerus Survived28/F Outpatient therapeutic abortion complicated by cardiopulmonary arrest, disseminated by cardiopulmonary arrest,

disseminated intravascular coagulation, anoxic encephalopathy, adult respiratory distress syndromeSurvived

45/M Closed head injury, status post craniotomy for subdural hematoma Survived66/M Urosepsis, chronic obstructive pulmonary disease Survived63/M Closed head injury, flail chest, pulmonary disease Survived73/F Pulmonary fibrosis, septic shock–urosepsis Survived65/F Left fem-pop bypass, aspiration pneumonia, adult respiratory distress syndrome, abdominal sepsis, congestive

heart failureExpired

80/M Pneumonia, adult respiratory distress syndrome, chronic obstructive pulmonary disease, septic arthritis Survived65/F Closed head injury, multiple facial fractures, bilateral meningitis Survived84/M Aspiration pneumonia, adult respiratory distress syndrome, urosepsis Expired31/M Depressed skull fracture, basilar skull fracture, facial fractures, hemorrhagic contusion Survived47/M Skull fracture, subdural hematoma, status post craniotomy Survived60/F Bilateral chronic subdural hematoma status post craniotomy, chronic renal failure axillary vein thrombus,

congestive heart failureExpired

F, female; M, male.

NASOENTERIC FEEDING TUBES IN CRITICALLY ILL PATIENTS266

11. Norwood SH, Civetta JM. Evaluating sepsis in critically ill patients. Chest1987;92:137

12. Lipman TO. Nasopulmonary intubation with feeding tubes. Therapeutic misad-venture or accepted complication? Nutr Clin Pract 1987;2:45

13. Ciocon JO, Silverstone FA, Grover M, et al. Tube feedings in elderly patients.Indications, benefits, and complications. Arch Intern Med 1988;148:429

14. Celis R, Torret A, Gatell JM, et al. Nosocomial pneumonia. A multivariateanalysis of risk and prognosis. Chest 1988;93:316

15. Craven DE, Kunches LM, Kilensky V, et al. Risk factors for pneumonia andfatality in patients receiving continuous mechanical ventilation. Am Rev RespirDis 1986;133:792

16. Bartlett JG, O’Keefe P, Tally FP, et al. Bacteriology of hospital acquiredpneumonia. Arch Intern Med 1986;146:868

17. Zaloga GP. Bedside method for placing small bowel tubes in critically ill patients.A prospective study. Chest 1991;100:1643

ANNOUNCEMENT

NASOENTERIC FEEDING TUBES IN CRITICALLY ILL PATIENTS 267