F L U O R O S C O P I C A I D I N N A S O D U O D E N A L T U B E I N S E R T I O N S H O U L D I T B E T H E L A S T C H O I C E ? T ü m a y U m u r o ğ lu , M . D . / Z e y n e p E ti, M . D . A r z u T a k ı l , M . D . / F. Y ı l m a z G ö ğ ü ş , M . D . D e p a r t m e n t o f A n e s th e s io lo g y a n d R e a n im a tio n , S c h o o l o f M e d ic in e , M a r m a ra U n iv e rs ity , Is ta n b u l, T u rk e y . A B S T R A C T
O b je c tiv e : Th e aim of this prospective and randomized study w a s to com pare the methods used for the a sse ssm e n t of the nasoduodenal tube position in critically ill patients according to the s u c c e s s rate, time spent on su cce ssfu l placem ent and cost effectivity.
M e th o d s: Sixty critically ill patients for whom enteral feeding with nasoduodenal tube (ND) w as planned, w ere allocated randomly into 3 groups. Th e a sse ssm e n t of the location of the ND tube w a s done by auscultation of the loudest sound location over the right flank in group A, pH determination of the duodenal aspirate in group PH and fluoroscopic view in group F. The failure criteria w a s the placem ent of a ND tube in the stom ach in all groups, the necessity of having more than two flat abdominal radiographs in group A and PH , the duration of fluoroscopy for more than 10 m inutes in group F. Th e su c c e ss rate, the time of su cc e ssfu l placem ent and the total co st w ere d eterm ined and com pared statistically by using a n a ly sis of variance and F ish e r’s E x a c t T e st.
R e s u lt s : Th e s u c c e s s rate w a s significantly higher in group F (95% ). Th e time of successful placem ent w as significantly shorter in group F (75.15 ± 5.32 min ve rsu s 147.95 ± 77.05 min in group A and 177.75 ± 154.84 min in group PH ).
T h e re w a s no difference between groups regarding the total cost. A 25% false positive result w as found in the PH group i.e. the tube w as found to be in the stomach after the radiological evaluation although pH values were > 4.
C o n c lu s io n : For critically ill patients in whom enteral feeding is planned with ND, fluoroscopy should be preferred w henever possible for the placem ent of the ND tube because the su cc e ss rate is higher, the placem ent is quicker and it is more cost-effective than the conventional methods.
K e y W o r d s : Enteral nutrition, Enteral tube insertion, Fluoroscopy.
IN T R O D U C T IO N
Tube feeding should be considered when a patient with a functional gut cannot or will not eat, and a method of a c c e ss can be safely obtained. The w ays of a c c e ss are obtained by nasogastric or nasoenteric tubes. It w as shown that feeding beyond the pylorus with nasoenteric tubes is a sso cia te d with a sig nificant reduction in g astro e so p h ag e al regurgitation and a trend toward le ss m icroaspiration (1 ). H ow ever, nasoenteric tube placem ent g enerates some difficulties. Spontaneous transpyloric p assag e of
Correspondence to: Tumay Umuroglu, M.D. ■ Department o f Anesthesiology and Reanimation. Marmara University Hospital Aitunizade. 81190 Istanbul, Turkey,
e.mail address: trans6@hotmail.com
Tumay Umuroglu, el al
the standard feeding tubes after 24 hours is only 30% (2). To overcome these difficulties several manual techniques are developed for the e a se of insertion. Zaloga (3), H eiselm an (4) and Rujeles (5) developed some of these manual techniques which represent the easiest and least expensive way of gastrointestinal a c c e ss a s it is possible to insert them manually at the bedside in a blind manner. If blind placem ent is not possible, one may use radiologic assista n ce i.e. fluoroscopy for post-pyloric nasoenteric tube placement. The fluoroscopic method has good su c c e ss rates when compared to manual placem ent techniques (6-8) but se em s to be more expensive and time consuming than the other methods and requires transportation to the radiology unit. In this regard, it is reasonable to investigate the overall cost of
all these m ethods and accordingly, this
p rospective, random ized study has been
undertaken to com pare the methods used for the a ssessm en t of nasoduodenal tube position in critically ill patients.
M A TER IA LS A N D M E T H O D S
The study w as approved by our institutional ethic committee and each patient or blood relative gave written informed consent. 60 critically ill neurosurgical patients (operated for intracranial lesions or ane urysm s) aged between 20-70 years, A P A C H E sc o re s 6-24 and for whom administration of a nasoduodenal tube (Abbott Laboratories, 12 F , flexiflo, polyurethane, w a te r- activated lubricant, radiopaque) w as planned, were included in the study. Th e indications for enteral feeding were lack of gag reflex or G C S < 9. The presence of bowel sounds w as required. Patients with a history of diabetes m ellitus, hepatic or renal failure, spinal cord trauma, electrolyte im balance, receiving dopaminergic or sedative agents were excluded from the study. All oral feedings or H2 receptor antagonists were stopped at least 6 hours before tube placement to ensure the presence of gastric acidity (9).
In each patient the distance from the xiphoid to earlobe to nose w as m easured and marked on the tube. Patients were placed in a supine position with their heads elevated to 30° and a prelubricated, polyurethane nasoduodenal tube (10 F) with a stylet w as inserted through the selected nostril and advanced into the stomach
till the premarked distance w a s reached. All procedures were performed by a specialized anesthesiologist. The patients w ere allocated randomly into 3 groups. In the first group (group A) the tube w as attached to a three-way stopcock and a 50 mL syringe. The tube w as advanced 15- 20 cm past the xiphoid-earlobe-nose mark while insufflating 50 mL air into the stom ach. A s the tube advanced, the change in the location of air bubbles w as auscultated through the right flank. When the loudest sound at the possible location of the duodenum over the right flank w a s heard, the tube w as left in place and taped. In the second group (group PH ) the tube w as again advanced 15-20 cm past the xiphoid-earlobe- nose mark until transpyloric sensation w as felt as Rujeles (5) described, suction w a s applied to obtain a sam ple of the intestinal contents to be
determined for its pH value with blood gas
analysis system (IRM A S L , Diam etric M edical, U SA ). If the pH value w a s > 4, the tube w a s left in place and fixed, otherwise it w a s withdrawn 15 cm and a second attempt w a s m ade. In the third group (group F) the tube w a s inserted under fluoroscopic view and when the tube entered the first part of the duodenum, it w a s fixed. After the procedure, all patients w ere given 10 mg metoclopramide i.v. to promote gastric motility and extrem e right lateral decubitus position with the head in 0° elevation w a s given during 30 minutes to facilitate the entrance of the tube to post pyloric sp ace. The place of all tubes w as confirmed with a flat abdominal radiograph. In all groups the study continued until the tube w as seen at least in the first part of the duodenum and the total time spent w as recorded.
Failure criterias w ere a s follows: Placem ent of the tube in the stom ach in all groups, the necessity of having more than 2 abdominal radiographs in group A and P H , inability to obtain an intestinal content during suction or to obtain a pH value < 4 during two subsequent attempts in group PH and duration of fluoroscopy more than 10 minutes in group F.
All flat abdominal radiographs w ere evaluated by a physician who w a s unaw are of the procedure used. The place of the tube w as recorded.
Th e su c c e ss rate, the total cost and the time of su c c e ssfu l p lacem en t in all groups w ere recorded.
Fluoroscopic aid in nasoduodenai tube insertion
Th e time of s u c c e s s fu l p lacem ent w a s determined a s the time spent from the beginning of the ND tube insertion until the decision made about the correct placem ent of the ND tube with flat abdominal radiograph in all groups.
R e su lts w ere e x p re s se d a s m ean ± S D . Significance of differences w a s determined by using One w ay A N O V A , Tukey-Kram er Multiple C o m p ariso n s test w a s used to determ ine p airw ise co m parison s and binary data w ere analyzed with F is h e r’s E x a c t T e st. P < 0.05 w as considered significant.
RESULTS
The dem ographic ch aracteristics of patients did not differ significantly between groups (Table I).
No com plications such a s pulmonary intubation, laryngeal sp a sm , uncontrolled bleeding in nose or throat or vomiting occurred in any group during the procedure.
In the fluoroscopy group, no com plications such a s d esatu ration or hem o d ynam ic instability occured in any patient during transportation.
Th e su c c e s s rate w a s significantly higher in the fluoroscopy group (95% ) com pared to others (Fig. 1).
Th e time of su cc e ssfu l placem ent of the tube w as significantly shorter in the fluoroscopy group (Table II).
T h e re w a s no d ifferen ce betw een groups regarding the total cost (Table III).
The cost of the methods w as a s follows: pH m easurem ent 7.5 $ , flat abdominal radiograph 10.6 $ and fluoroscopy 27.5 $. The total cost w as
calcu late d accord ing to the num ber of
radiographs needed.
W e failed to obtain intestinal contents from the tip of the ND tube in 2 (10% ) patients and in 5 (25% ) patients radiological a sse ssm e n t showed that the ND tube w a s placed in the stomach although m easured pH va lu e s w ere higher than 4.
Table I: Demographic characteristics of patients (mean±SD)
Group A Group PH Group F
N 20 20 20
Age (year) 53.55 ± 17.28 56.10 ± 15.10 48.25 ± 17.70
Gender (M/F) 13/7 12/8 12/8
Weight (kg) 69.60 ± 9.40 67.15 ± 9.70 71.00 ±11.60
Height (cm) 169.30 ±12.30 169.30 ± 13.10 170.90 ± 12.40
A: auscultation; F: fluoroscopy; PH: pH value
95%
1001
an
RCIV-A P H F
F i g . l : S uccess rate
(*p<0.05 com pared to group F)
A: auscultation; F; fluoroscopy; PH; pH value
T a b l e II: T im e to s u c c e s s fu l p la c e m e n t
Group A Group PH Group F
Time (min) 147.95 ± 77.05' 177.75 ± 154.84" 75.15 ±5.32
A: auscultation; F: fluoroscopy; PH: pH value
('p<0.05 and *p<0.01 compared to fluoroscopy group)
T a b l e III: T o ta l c o s t
Group A Group PH Group F
Cost ($) 27.06 ± 20.50 43.15 ±35.00 38.10 ± 0.00
A: auscultation; F: fluoroscopy ; PH: pH value (p>0.05)
D IS C U S S IO N
According to the results of our study, in patients with decreased gag reflex or decreased G C S , fluoroscopic placem ent alone se e m s to be more su ccessfu l and quicker than auscultation and pH m easurem ent methods.
Many bedside insertion tech n iq ue s for the p assag e of nasoenteric tubes beyond the pylorus
Tümay Umuroglu, et al
in critically ill patients are available. One of them is a "blind” approach with auscultation technique (hearing the progression of the loudest sound locations from left to the right abdomen). The su cc e ss rate of this technique is reported to be low in many studies (10,11) a s well as in our
study (6 0 % ). Furtherm ore if inadvertent
placem ent into the low er portion of the esophagus takes place, the sounds may mislead the clinicians (12).
pH guided technique developed by Zaloga, Heiselm an and R u jeles (3-5) is another method for the insertion of the nasoduodenal tube. Many
studies enrolled to e valu ate its efficacy
concluded that pH m easurem ent technique is successful only if it is used in the combination with another technique. For exam ple, Metheny et al (13) stated that pH test strip used in combination with a bilirubin test strip improved the ability to differentiate between gastric and intestinal tube placem ent. The sam e authors, in another study, concluded that pH m easurem ents in conjunction with determinations of trypsin and pepsin en zym e co ncen tratio n s in feeding aspirates help to predict the tube position (14). We used pH m easurem ent without a combination with another method and sim ilar to these studies our su c c e ss rate w as low (55% ). Th is method can give false positive results when the tube is misplaced into the lungs a s the high pH of sam ples such as 7.87 obtained from the tip of nasoduodenal tube can be misinterpreted (14). So, it is possible to feed the lungs. W e did not have this complication in our study but we had false positive results of about 25 % in PH group. H2 receptor antagonist agents increase gastric fluid pH and this may lead to misinterpretation of
the tube position. For this reason we
discontinued H2 receptor antagonists 6 hours before the procedure. So false positive results may not be due to the use of these agents but we did not know whether these patients had pyloric insufficiencies causing higher gastric pH values b ecau se of the regurgitation of intestinal contents. It is obvious that pH m easurem ent technique when used alone is not reliable in critically ill patients.
Fluoroscopical techniques have also been used and found su ccessful in placing nasoduodenal tubes in patients for whom aspiration risk is high (15). In a retrospective study the tube w as
fluoroscoplcally placed distal to the third portion of the duodenum in 8 6 .6 % (16). Our su c c e s s rate w as 95% in the fluoroscopy group and this difference in the incidence may be due to the target place of the tube as w e planned to place the tube in the first portion of the duodenum w hereas in the study of G utierrez (16) it w as planned to be placed in the third portion. Also in the study of Huerta (17) it is ad vised that nasoenteric tubes should be placed with the guidance of fluoroscopy a s the caloric delivery is quicker when compared with the blind technique.
The most important point in all techniques is that proper placement of the tube must be absolutely verified before feeding begins. Th is verification is u sually done by taking a flat abdom inal radiograph. O ne exception for this is the fluoroscopy technique since the place of the nasoduodenal tube is already confirmed during the procedure.
Th e low su c c e ss rate of any technique lead s to increased radiation exposure for the patient himself as well a s for the other patients in the ward and especially for the ICU staff. B esid e this
untoward effect, m ultiple plain abdom inal
radiographs increase the cost of the procedure and consum e time. In our study we did not find any sig nificant d iffe re n ce s betw een groups regarding the total cost even though w e took an
additional abdom inal radiograph in the
fluoroscopy group a s the number of radiographs needed w a s higher in the pH and auscultation groups.
Another method, endoscopic nasoenteric tube p lacem ent h as been recen tly investig ated (18,19). Th is technique has som e limitations because the tube may migrate back into the stomach during the withdrawal of the scop e, thus it still requires the confirmation with a radiograph.
The use of electrom agnetic technique has also been investigated (20). In the study of K e arn s (20), the s u c c e ss rate w a s found to be high, but the tube w as accepted to be in place if it w a s just below the diaphragm . Th e y did not intend to place it in the duodenum.
In overall, the time of su cc e ssfu l placem ent of the tube w as longer than it w a s found in other resembling studies. Th is delay m ay be explained
Fluoroscopic aid in nasoduodenal tube insertion
by the difference in time spent for taking portable radiographs in different intensive care units.
Fluoroscopic placem ent of the nasoduodenal tubes is the preferred method when other attempts fail. According to the results of our study we recom m end using fluoroscopic placem ent of the nasoduodenal tubes in a specific group of critically ill p atien ts su ch a s patients with decreased gag reflex or d ecreased G C S in the first order, b e c a u se it is a time sa vin g , s u c c e s s fu l, and co st-effective method. Furtherm ore, these ad vantages combined with low com plication risk and le ss exp osu re to radiation a s there is no n ecessity for abdominal radiographs, le ave s its single disadvantage to be
the requirem ent of transportation of these
patients to the radiology unit.
REFERENCES
1. H eyland DK, D ro ver JW, M acD onald S, e t at. E ffe c t o f p o s tp y lo ric fe e d in g on g a s tro e s o p h a g e a l re g u rg ita tio n a n d p u lm o n a ry m ic ro a s p ira tio n : R esults o f a ra n d o m ize d co n tro lled trial. Crit Care Med 2 0 0 1 ; 2 9 :1 4 9 5 - 1 5 0 1 .
2. B engm ark S. Progress in p erio p erative enteral tu b e feeding. Clin H u tr 1 9 9 8 ; !7 : 1 4 5 -1 5 2 . 3. Zaloga GP. B edside m e th o d fo r placing sm all
b ow el feeding tubes in critically ill patients: A pro spective study. C hest 1 9 9 1 ; 1 0 0 :1 6 4 3 -
1646.
4. H eiselm an DE, Vidovich RR, Mildovich G, et al. ria s o ln te s tin a l tu b e p la c e m e n t with a p h sen sor feed in g tube. JPEbi 1 9 9 3 ; 17: 5 6 2 - 5 6 5 .
5. Rujeles S, G o m ez G, C adena E, et al. Sondas d e alim e n ta c ió n enteral: Validación de una técnica sencilla d e in tu b ació n transpilorica. Universitas M edica 1 9 9 3 ; 3 4 : 19-23.
6. G rant JP, Curtas MS, Kelvin EM. Radiologic p la c e m e n t o f n a s o je ju n a l feeding tubes with im m e d ia te feedings using a n o n e le m e n ta l diet. JPEbi 1 9 9 3 ; 7 :2 9 9 -3 0 3 .
7. P rager R, L a b o y V, Venus B. V alue o f ra d io lo g ic a s is ta n c e d u rin g tran sp ylo ric intubation. C rit Care M ed 1 9 8 6 ; 14: 1 5 1 -1 5 2 . 8. Pearce CB, D u n can HD. E n te ra l feeding,
blasogastric, n a s o je ju n a l, p e rc u ta n e o u s
endoscopic gastrostom y o r je ju n o s to m y : Its indications an d lim itations. Postgrad Med J 2 0 0 2 ; 78: 1 9 8 -2 0 4 .
9. M arik PE, Lorenza A. Effect o f tube feedings on the m e a s u re m e n t o f gastric intram ucosal pH. Crit Care M ed 1 9 9 6 ; 24:
1 4 9 8 -1 5 0 0 .
10. Welch SK, H an lo n MD, Waits M, et al. Com parison o f fo u r bedside indicators used to predict duodenal feeding tube p lacem en t with radiography. JPEbi 1 9 9 4 ; 18: 5 2 5 -5 3 0 . I I . bieumann MJ, M eyer CT, Dutton JL, et al. Hold
that x-ray: aspirate p H a n d auscultation prove en teral placem ent. J Clin Gastroenterol 1 995; 2 0 : 2 9 3 -2 9 5 .
12. M etheny HA, Spies M, Eisenberg P. Frequency o f n a s o e n te ra l tu b e d is p la c e m e n t and associated risk factors. Res biurs Health 1 986; 9: 2 4 1 -2 4 7 .
13. M e th e n y HA, S m ith L, S te w a rd BJ. D evelo pm ent o f a reliable and valid bedside test fo r bilirubin and its utility fo r im proving p rediction o f feeding tu b e location, blurs Res 2 0 0 0 ; 4 9 : 3 0 2 -3 0 9 .
14. M etheny HA, Steward BJ, Smith L, et al. pH an d concentrations o f pepsin and tripsin in feeding tu b e aspirates as predictors o f tube placem ent. JPEbi 1 9 9 7 ; 21 : 2 7 9 -2 8 5 .
15. Hillard AE, W addell JJ, M etzler AI f f et al. Fluoroscopically guid ed nasoenteric feeding tu b e p la c e m e n t versus bedside placem ent. South M ed J 19 9 5 ; 8 8 : 4 2 5 -4 2 8 .
16. G u tie rre z ED, B alfe DM. Fluoro scop ically guid ed nasoenteric feeding tu b e placem ent: results o f a I-y e a r study. Radiology 1991;
I 78: 7 5 9 -7 6 2 .
I 7. H uerta G, Puri VK. btasoenteric feeding tubes in critically ill patien ts (fluoroscopy versus blind), n utrition 2 0 0 0 ; 16: 2 6 4 -2 6 7 .
18. Kwauk ST, Miles D, Pinilla J, et al. A sim ple m e th o d fo r en d o s c o p ic p la c e m e n t o f a n a s o d u o d e n a l feeding tube. Surg Endosc
1 9 9 6 ; 10: 6 8 0 -6 8 3 .
19. Patrick PG, M arulendra S, Kirby DF, et al. Endoscopic nasogastric-jejunal feeding tube p la c e m e n t in critically ill patients. Gastrointest Endosc 1 9 9 7 ; 4 5 : 72-76.
2 0 . Kearns PJ. A c o n tro lle d co m p a ris o n o f traditional feeding tu b e verification m ethods to a b edside, e le c tro m a g n e tic technique. JPEbi 2 0 0 1 ; 2 5 : 2 1 0 -2 0 5 .
