A Randomized Study of Early Nasogastric versus Nasojejunal Feeding in Severe Acute Pancreatitis

1
A Randomized Study of Early Nasogastric versus
Nasojejunal Feeding in Severe Acute Pancreatitis

• American Journal of Gastroenterology, 2005

2
Original question

• TPN traditionally mainstay of nutritional support
  for severe pancreatitis ? Suppression of
  pancreatic exocrine secretion by bowel rest
• Does early enteral feeding confer a benefit in
  acute pancreatitis? Are there any other options?

3
Background

• No studies have compared enteral or parenteral
  feeding vs no feeding in severe pancreatitis

4
Background

• At least 6 randomized trials have compared
  enteral vs. parenteral feeding
• -overall results favour enteral decreased
  infective complications, surgical intervention
  and LOS
• Possible disadvantages of nasojejunal feeding
  related to insertion of NJ tube ?more
  complications ?delay in introduction of feeding

5
Background

• Novel question
• Is nasogastric and nasojejunal feeding in severe
  pancreatitis comparable?
• Glasgow Royal Infirmary

6
Aims

• Is nasogastric feeding as safe and as effective
  as nasojejunal feeding in severe pancreatitis?
• Does NG feeding result in reactivation of
  pancreatitis in any way?
• Does NG feeding avoid some of the problems of NJ
  feeding?

7
Method

• 50 pts at Glasgow Royal Infirmary from Oct 1997
  to July 2000
• -clinical and biochemcal presentation of acute
  pancreatitis (AP) serum amylase 3x normal
• -severe pancreatitis Glasgow prog score 3,
  APACHE II 6, CRP gt 150

8
Method

• Randomization by computerised random no.
  generation
• No blinding of participants or investigators
• Parameters measured daily
• APACHE II, CRP, visual analogue scale for pain,
  total analgesic requirements

9
Method

• NG tubes 8Fr Flocare
• NJ tubes placed at endoscopy half 8Fr flocare,
  other half 7Fr nasobiliary catheter
• Feeds commenced on avg 72 hrs post onset of pain
  at 30mL/hr, increasing over 24-48hrs to 100mL/hr
• Low fat semi-elemental feed carbo 75,
• protein 16, fat 9

10
Objective to assess any difference between NG
and NJ feeds

• Primary outcome measures
• CRP, APACHE II, pain score, total analgesia,
  need to convert to parenteral feeding
• Secondary outcome measures
• length of stay in hospital and ICU, mortality
• Power calc - 20 difference in CRP

11
Aetiology of pancreatitis

• Total 50pts 1 excluded
• 49pts
• / \
• 27 in NG gp 22 in NJ

Aetiology 27 NG 22 NJ
Gallstone 16 16
Alcohol 6 6
Other 5 0
12
Comparing the two gps

• Demographics (age,sex) similar for both gps
• Time to commence feeds median 72hrs and time to
  full rate of feeds (an additional 36 hrs) similar
  both gps
• 36 of 49pts tolerated a rate of 75 of target
  within 48hrs (19/27 in NG) and (17/22 in NJ)

13
Results

• Complications
• one pt had cardioresp arrest inserting NJ
• bloating - 1 NJ
• diarrhoea 3 NG and 1 NJ
• Death
• 12 of 49pts (24.5)
• 5/27 in NG gp vs. 7/22 in NJ gp
• gallstones 9, alcohol 3

14
Comparison of APACHE II scores similar for both
gps
15
No statistical difference with levels of CRP
between the two gps
16
Similar pattern of pain as measured by VAS and
analgesic requirements
17
Conclusions

• No difference in outcome for all end-points
  between those receiving NG and NJ feeds
• Nasogastric feeding in patients with severe
  pancreatitis is simpler and cheaper

18
Discussion

• Study design non blinded randomised prospective
  trial
• - small study 49 pts
• - largest study looking at enteral feeding in
  severe pancreatitis
• - non-blinded esp. to assessors observer bias
• Achieved aims in showing no difference between
  two gps of patients

19
Discussion

• Applicability
• -advantage of NG feeds in minimising delay but
  time to commencement of feeds 72hrs
• Timing and composition of feeds may be of equal
  importance ?addition of probiotics
• Novel question larger studies needed

20
The End