Вы находитесь на странице: 1из 3

Indian J Gastroenterol (NovemberDecember 2013) 32(6):366368

DOI 10.1007/s12664-013-0348-8


A comparative observational study of early versus delayed

feeding after percutaneous endoscopic gastrostomy
Manoj A. Vyawahare & Mrunal Shirodkar &
Amit Gharat & Prachi Patil & Shaesta Mehta &
K. M. Mohandas

Received: 29 June 2012 / Accepted: 6 July 2013 / Published online: 17 August 2013
# Indian Society of Gastroenterology 2013

Abstract from the next day. All patients were evaluated on days 1, 2, 7,
Background Percutaneous endoscopic gastrostomy (PEG) is and 30.
the procedure of choice for long-term enteral feeding. Results There were 55 patients (47 males) in group I and 54
Objective This prospective observational study was carried patients (38 males) in group II who were matched for age
out to compare the safety of commencing feeding 3 h com- (mean age 46.1 and 46.1 years, respectively). Complications
pared to 1624 h after PEG tube placement. included PEG site infection (1), peristomal leak (3), and
Methods One hundred and ten patients with oropharyngeal displacement of the PEG tube (1) in group I. PEG site
malignancies who had consented for PEG were enrolled. infection was seen in five patients in group II. All the
Trial-specific consent and IRB approval were not obtained complications were managed conservatively on an ambula-
because at the time when this study was done, this was not tory basis.
mandatory for observational studies which involved stan- Conclusion Initiation of tube feeding 3 h after an uncompli-
dard procedures. Alternate patients were started on early cated PEG was safe, well tolerated, and helped to reduce the
feeding within 3 h (group I) or after overnight observation hospital stay.
of 16 to 24 h (group II). Five hundred milliliters of Ringer's
lactate was infused over 4 h initially, followed 2 h later by Keywords Complications . Nutrition . Oropharyngeal
200 mL of formula feed. The patients were advised to take cancer
bolus feeds of 200 mL every 2 h and oral feeds ad libitum


M. A. Vyawahare (*) Percutaneous endoscopic gastrostomy (PEG) is a time-tested

Department of Medical Gastroenterology and Hepatology, procedure for long-term enteral alimentation [1, 2]. It is
Meditrina Institute of Medical Sciences, Ramdaspeth,
currently performed in patients suffering from neurological
Nagpur 440 010, India
e-mail: drmanojvyawahare@gmail.com disorders, neoplasms of the head, neck, and esophagus, and
those with head injury [3]. Two meta-analyses have con-
M. Shirodkar firmed the safety of early feeding after PEG placement [4,
Abbott Nutrition India, Mumbai, India
5]. These studies have shown that feeding after PEG can be
A. Gharat started early within hours, thus making the procedure a day
Department of Gastroenterology, Topiwala National Medical care procedure without any significant increase in the
College and B Y L Nair Hospital, Mumbai Central, procedure-related morbidity or mortality and thereby reduc-
Mumbai 400 008, India
ing the healthcare costs [611]. In spite of the evidence, there
P. Patil : S. Mehta is a tendency to keep the patients fasting until the next
Division of Digestive Diseases and Clinical Nutrition, Tata morning [3, 12, 13]. One of the views has been that Indian
Memorial Centre, Parel, Mumbai 400 012, India patients are more malnourished and therefore more likely to
develop complications. We carried out this prospective ob-
K. M. Mohandas
Department of Digestive Diseases, Tata Medical Center, servational study to compare the safety of starting early
Kolkata 700 156, India feeding after PEG tube placement in Indian circumstances.
Indian J Gastroenterol (NovemberDecember 2013) 32(6):366368 367

Methods and Materials Table 1 Clinical features of the 109 patients

Characteristics Group I Group II

This prospective observational study was conducted at the
Division of Digestive Diseases and Clinical Nutrition, Tata Patients 55 55*
Memorial Centre, Mumbai. One hundred and ten consecu- Mean age (years) 46.1 46.1
tive patients with head and neck cancers who had consented Male/female 47:8 38:16
to undergo PEG from 1 May 2006 to 31 September 2006 Sites of cancer
were included. Trial-specific consent and Institutional Re- Tongue 21 18
view Board (IRB) approval were not obtained because at the Buccal mucosa 13 21
time when this study was done, it was not mandatory to get Alveolus 15 09
IRB approval for observational studies which involved stan- Others 6 6
dard procedures. All the PEGs were performed using the pull Subjective global assessment
technique [1], using 22 Fr or 24 Fr commercial PEG tubes A 15 10
(Cook Endoscopy, Winston Salem, NC27105, USA). Alter- B 32 39
nate patients undergoing PEG were started on early feeding C 8 5
after 3 h or delayed feeding after overnight observation. The Serum albumin, mean (g/dL) 4.2 4.0
PEG was abandoned in one patient who developed stridor Major complications 0 0
after the endoscopy. The PEG procedure was uncomplicated 30-day mortality 0 0
in all the other 109 patients. Thus, 55 patients received Minor complications 5 5
feeding 3 h after the PEG placement (group I) and 54 patients
received feeding 1624 h after the PEG (group II). All *One patient did not undergo percutaneous endoscopic gastrostomy
patients were evaluated by the physician who performed due to stridor
the PEG before starting the tube feedings. The PEG feedings
were initially started with 500 mL of Ringer's lactate infused Discussion
over 4 h, followed 2 h later by 200 mL of commercial
formula feeds for the first day. Thereafter, they were advised Many studies have examined the use of early PEG feedings
bolus feeds of 200 mL every 2 h and oral feeds ad libitum and indicate that early PEG feeding is safe and well tolerated
next day. The patients were evaluated on days 1, 2, 7, and 30 by adult and pediatric patients [611]. Bechtold et al.
for major and minor complications. The vital signs, abdom- performed a meta-analysis of six randomized control trials
inal distension, tenderness, position of the tube, peristomal (n=467) that compared the early (4 h) vs. delayed or next-
leakage, and wound sites were evaluated at each visit by un- day feeding after placement of a PEG [4]. No statistically
blinded investigators. significant differences were noted between early and delayed
or next-day feedings for patient complications or death in
72 h. A statistically significant increase in gastric residual
volumes during day 1 was noted which did not affect overall
Results complications. The authors concluded that early feeding 4 h
after PEG placement may represent a safe alternative to
The baseline characteristics of the patients in the two groups delayed or next-day feedings [4]. Szary et al. performed
are summarized in Table 1. While the two groups were another meta-analysis of five studies (n=355) and concluded
similar with regard to mean age, there were more male that early tube feeding 3 h after PEG placement had no
patients in group I. The sites of cancer in the study subjects significant differences to delayed or next-day feeding in
were similar in both groups. The extent of malnutrition respect to complications, death in 72 h, or number of
determined by using the subjective global assessment score significant gastric residual volumes at day 1 [5].
revealed that vast majority of our patients had moderate to Despite the level I evidence now available, it remains
severe nutritional risk in both arms. The mean serum albu- common practice to delay post-PEG feedings. A survey of
min was similar in both groups. gastroenterologists in north-eastern United States in 1998
There were no major complications requiring re- revealed that although 82 % of specialists were aware of
hospitalization or intervention and no mortality in this study. the recent literature showing early feedings to be safe, only
Infection at the PEG stoma was seen in one patient in group I 39 % initiated feedings prior to 8 h and 11 % initiated
and five patients in group II. Peristomal leak was seen in three feedings prior to 3 h. The remaining 61 % of gastroenterol-
patients and displacement of PEG tube was seen in one patient ogists surveyed chose to delay feedings from 9 h to more
in group I. There was no statistically significant difference in than 24 h [12]. A subsequent web-based review of practices
the overall complication rates in the two groups. revealed that only 9 % of gastroenterologists initiate PEG
368 Indian J Gastroenterol (NovemberDecember 2013) 32(6):366368

tube feedings 3 h despite 41 % being aware of the recent conclude that initiation of early feeding 3 h after an uncompli-
literature regarding PEG feeding initiation times [13]. De- cated PEG is safe, well tolerated, and can shorten hospital stay.
layed feeding was practiced by endoscopists in Canada, with
median time to initiation of feeding for inpatients being
approximately 48 h with a range of 24 to 48 h [3].
Based on the meta-analyses, current guidelines recom- References
mend the initiation of PEG feedings within a few hours of
the procedure thereby enabling an early discharge [14]. At 1. Gauderer MWL, Ponsky JL, Izant RJ Jr. Gastrostomy without
the time while we were undertaking this study, multiple laparotomy: a percutaneous technique. J Pediatr Surg. 1980;15:8725.
2. Gomes CA Jr, Lustosa SA, Matos D, Andriolo RB, Waisberg DR,
surveys had indicated that many endoscopists still hold on
Waisberg J. Percutaneous endoscopic gastrostomy versus nasogas-
to delayed feeding until the next morning. This delay is tric tube feeding for adults with swallowing disturbances. Cochrane
probably an extrapolation of the surgical gastrostomy prac- Database Syst Rev. 2010;11, CD008096.
tices. Delaying the feeding is also attributed to keeping the 3. Pruthi D, Duerksen DR, Singh H. The practice of gastrostomy tube
placement across a Canadian regional health authority. Am J
gastric residual volumes low in order to minimize aspiration
Gastroenterol. 2010;105:154150.
and reduce the risk of peritoneal leakage and peritonitis. 4. Bechtold ML, Matteson ML, Choudhary A, Puli SR, Jiang PP, Roy
Recent trials show that increased gastric residuals correlate PK. Early versus delayed feeding after placement of a percutaneous
poorly with risk of aspiration [15]. Because the PEG stoma endoscopic gastrostomy: a meta-analysis. Am J Gastroenterol.
tract will take a week or more to epithelize and mature, 2008;103:291924.
5. Szary NM, Arif M, Matteson ML, Chaudhary A, Puli SR, Bechtold
delaying the feeds by a day would not have much role in ML. Enteral feeding within three hours after percutaneous endoscop-
reducing leaks. Meta-analysis of several surgical studies show ic gastrostomy placement: a meta-analysis. J Clin Gastroenterol.
that early postoperative nutrition is associated with significant 2011;45:e348.
reductions in total complications compared with traditional 6. Chumley DL, Batch WJ, Hoberman LJ, et al. Same day PEG
feeding, is it safe? Results of randomized prospective study. Am J
postoperative feeding practices and does not negatively affect Gastroenterol. 1993;88:1589.
outcomes such as mortality, anastomotic dehiscence, resump- 7. Choudhry U, Barde CJ, Markert R, Gopalswamy N. Percuta-
tion of bowel function, or hospital length of stay [16]. Yarze neous endoscopic gastrostomy: a randomized prospective com-
et al. studied the risk for hydroperitoneum/peristomal leak parison of early and delayed feeding. Gastrointest Endosc.
with a water-soluble contrast agent within 3 h post endoscopic 8. Werlin S, Glicklich M, Cohen R. Early feeding after percutaneous
gastrostomy button placement and none of the patients endoscopic gastrostomy is safe in children. Gastrointest Endosc.
showed evidence of leakage [17]. So, there is no rationale 1994;40:6923.
for delaying the feeds by 24 h. 9. McCarter TL, Condon SC, Aguilar RC, Gibson DJ, Chen YK. Ran-
domized prospective trial of early versus delayed feeding after percu-
Our results concur with these recommendations and we taneous endoscopic gastrostomy placement. Am J Gastroenterol.
changed our practices and started early feeding in all those 1998;93:41921.
who have had an uncomplicated PEG procedure. We were 10. Schulte-Bockholt A, Sabin M, Rosenstock U, et al. Immediate
able to initiate PEG feedings 3 h after the procedure without versus next day PEG feeding: a randomized prospective study in
ICU/intermediate care patients. Gastroenterology. 1998;114:A907.
any major detrimental effects. All of our patients had oro- 11. Stein J, Schulte-Bockholt A, Sabin M, Keymling M. A randomized
pharyngeal malignancies that are generally considered as prospective trial of immediate vs. next-day feeding after percutane-
high risk patients. Over two-thirds of our patients had mod- ous endoscopic gastrostomy in intensive care patients. Intensive
erate to severe malnutrition using the subjective global as- Care Med. 2002;28:165660.
12. Srinivasan R, Fisher RS. Early initiation of post-PEG feeding: do
sessment. There was no major morbidity like peritonitis, published recommendations affect clinical practice? Dig Dis Sci.
intraabdominal hemorrhages requiring re-admission or re- 2000;45:20658.
intervention, and no 30-day mortality. The frequencies of 13. Tauseef A, Vu LE, Sharma T, Shrinivasan N, Tierney WM, Rizvi
minor complications were similar in both groups and were SM. Post-PEG feeding time: a web-based national survey of gas-
troenterologists. Gastrointest Endosc. 2009;69:AB179.
managed on outpatient basis. This enabled us to discharge 14. Loser C, Aschl G, Hebuterne X, et al. ESPEN guidelines on
patients early after starting early PEG feeds. Health care artificial enteral nutrition- percutaneous endoscopic gastrostomy
costs are very relevant in India as in the West, as majority (PEG). Clin Nutr. 2005;24:84861.
of these are borne by the patients themselves. This also 15. McClave SA, Lukan JK, Stefater JA, et al. Poor validity of residual
volumes as a marker for risk of aspiration in critically ill patients.
helped to make beds to those on the admission waiting list. Crit Care Med. 2005;33:32430.
Our study had some limitations with regards to external 16. Osland E, Yunus RM, Khan S, Memon MA. Early versus tradition-
validity. It was not a randomized controlled trial and hence al postoperative feeding in patients undergoing resectional gastro-
was prone to bias and confounding. The statistical power of intestinal surgery: a meta-analysis. JPEN J Parenter Enteral Nutr.
this study was low due to a convenience sample size of 110 17. Yarze JC, Herlihy KJ, Fritz HP, et al. Prospective trial evaluating
subjects. Nonetheless, our study replicated and validated the early initiation of feeding in patients with newly placed one-step
results of the meta-analysis in a routine practice scenario. We button gastrostomy devices. Dig Dis Sci. 2001;46:8548.