Decompression in canine GDV: orogastric intubation? gastric trocarisation?

Background

Gastric decompression is an essential part of the initial pre-operative stabilisation of dogs with GDV once aggressive intravenous fluid therapy is under way. There are two methods of achieving this: orogastric intubation and needle puncture (gastric trocarisation).

The prime purpose of decompression at this stage is to relieve gaseous distension; gastric lavage and more thorough gastric evacuation are reserved for the intraoperative period.

For many years now my approach has been to use orogastric intubation under judicious sedation (pure opioid + midazolam) and I have reserved gastric trocarisation for those cases in which:

  • The patient is intolerant of orogastric intubation despite appropriate judicious sedation
  • The patient is intolerant of orogastric intubation and deemed too unstable even for judicious sedation
  • A stomach tube cannot be passed

Very rarely I have performed needle decompression when a patient has been visibly deteriorating in front of me as it is much quicker to get the equipment you need and perform the procedure than it is to set up and perform orogastric intubation.

Of course orogastric intubation is not risk free. Many patients require sedation and there are risks of iatrogenic injury to the oesophagus or stomach if care is not taken, as well as a possible increased risk of aspiration pneumonia. For some clinicians needle decompression is the default technique as it is very quick and easy to perform, does not require sedation, and is effective in relieving gaseous distension. So what are some of theoretical downsides? Why do other clinicians reserve needle decompression as their second-line technique? Some of the concerns are that:

  • Needle puncture does involve blindly placing a needle through a gastric wall that may already be severely compromised; this may cause further compromise of the stomach wall and require more extensive surgical intervention during subsequent laparotomy such as gastrectomy or invagination.
  • There is potential risk of gastric contents leaking into the abdomen.
  • There is a risk of accidental injury to other organs/structures, especially the spleen, if care is not taken.

Now to be honest I stick to doing what I do and teaching what I teach because it has worked for me. Following the approach I do, in the environments in which I have been lucky enough to work survival rates for dogs with GDV are very high and to a degree we adopt a ‘if it ain’t broke don’t try to fix it’ mentality. But is there any evidence in the literature about any of this? Well I came across the paper below; according to these authors there are no other published veterinary studies investigating the risks of gastric trocarisation.

Paper

Goodrich ZJ, Powell LL, Hulting KJ. Assessment of two methods of gastric decompression for the initial management of gastric dilatation-volvulus.  J Sm Anim Prac 2013. 54:75–79.

Objectives

  • To report the effectiveness of orogastric tubing and gastric trocarisation
  • To report their complications
  • To investigate any differences in complication rate between the two methods

Design

  • Retrospective medical record review

Setting

  • University teaching hospital in USA
  • June 2001 – October 2009

Population

  • 116 dogs with GDV
  • Inclusion criteria:
    • GDV confirmed via a right lateral abdominal radiograph
    • Method of preoperative gastric decompression was noted in record
    • Definitive surgical correction was performed

Interventions

  • Orogastric intubation with sedation (typically benzodiazepine + opioid) or gastric trocarisation without sedation
    • Procedures described in paper seem standard
    • Laparotomy including partial gastrectomy or gastric invagination as indicated ± splenectomy

Outcome

“Comparisons were made to determine if statistically significant associations existed between the method of gastric decompression and gastric compromise treated with surgical intervention and overall survival.”

Some results

  • Decompression procedure in 116 dogs:
    • Orogastric tubing in 31 (27%)
    • Gastric trocarisation in 39 (34%)
    • Combination of both techniques in 46 (40%). “For those dogs undergoing both procedures, the decision of which procedure to perform first was not standardized and was based on clinician preference.”
    • Orogastric tubing:
      • Successful in 59 (76•6%) dogs, unsuccessful in 18 (23•4%) dogs
      • No complications during orogastric tubing during tube passage or subsequent gavage
      • No dogs that underwent orogastric tubing were diagnosed with or developed clinical signs of aspiration pneumonia or oesophageal perforation while hospitalised
      • Trocarisation:
        • Successful in 73 (86%) dogs, unsuccessful in 12 (14%) dogs
        • No evidence of gastric leakage into the abdomen noted at surgery in any dog undergoing gastric trocarisation
        • Splenic laceration occurred in 1 of 85 (1•2%) dogs in which gastric trocarisation was performed but splenectomy not required
        • Gastric perforation was not noted in any dog undergoing orogastric tubing or gastric trocarisation
        • Median length of hospitalisation:
          • Orogastric tubing 1 day (1-6 days)
          • Trocarisation 2 days (1-5 days)
          • Orogastric tubing + trocarisation 1 day (1-6 days)
          • One hundred eleven dogs (95•7%) survived to discharge, 5 (4•3%) dogs did not: orogastric tubing performed in two, gastric trocarisation in one, combination of orogastric tubing and gastric trocarisation in two.

“No significant difference was found between the method of gastric decompression and gastric compromise requiring surgical intervention (χ2=0•8, P=0•68) or survival to discharge (χ2=0•38, P=0•83).”

Authors' conclusions

In conclusion, this study found minimal complication rates and high success rates of two methods of gastric decompression in this population of dogs presenting with GDV.

Some comments about the study, some of which the authors mention

  • Having survival to discharge as one of the primary outcomes is good because this is a patient-centred outcome that is important clinically. However,  given the very low mortality rate it is extremely unlikely that the results would have shown an association between one decompression technique or the other and mortality. Moreover both techniques were used in a number of cases reducing the number available for direct comparison between the two techniques. Potentially an association could be demonstrated with a much larger sample size. However it is essential to remember that even then an association would not infer causality, i.e. it would be one thing to say that one or other of these techniques was shown to be associated with increased mortality, a whole other thing to say that the decompression technique caused the greater mortality. Other factors that may influence mortality would need to be accounted for.
  • States why cases that were identified in the initial database search were then excluded: “Cases were excluded from analysis if the method of preoperative gastric decompression was not noted in the medical record.”
  • Patient-centred primary outcomes relevant to clinical practice
  • Retrospective study with all the usual limitations that affords in terms of bias, accounting for confounders, poor recording of information etc. and invariably some information was missing for some of the 116 dogs; retrospective studies ideally act to generate a hypothesis which would then be tested in a prospective randomised controlled trial.
  • Single centre which may lead to doubt about external validity – extrapolation to other populations of dogs with GDV in other hospitals. That said, their techniques described seem somewhat ‘standard’. Survival rate reported here is comparable to that in other reports from similar university teaching hospital-type environments.
  • One of the stated objectives is to report ‘effectiveness’ of the two decompression techniques. There is then no further mention of this until the penultimate paragraph of the Discussion where it is alluded to in a roundabout way by mentioning that one could possibly try to assess improvement in haemodynamic status after decompression.
  • Not stated whether gastropexy was performed and if so, how.

My bottom line

So I guess after reading this paper I don’t think it will have any impact on my clinical practice. It is definitely good to have some published information on the subject; however the significance of this evidence given the nature and limitations of the study need to be borne in mind when considering any impact on clinical practice. On the face of it, based on this study, the reasons for which I and others do not use needle trocarisation as our default technique may seem even less valid…but conceptually I do not like the idea of creating a hole in an organ unless I feel I need to and I would like a lot more and better quality evidence before reconsidering. One other thing that needs to be borne in mind is that it is an implicit assumption in this study discussion that all cases subsequently undergo surgical intervention. This is in keeping with my recommendations but one would need to think even harder about performing gastric trocarisation if surgery was not to be performed.

That said, I will worry just a little less when I do need to perform needle decompression…not that I worried that much anyway. As I say, thankfully in the environments in which I have worked survival rates for dogs with GDV are very high regardless of decompression technique. I would imagine that in environments in which this is not the case, the decompression technique is likely to be a relatively trivial consideration with respect to explaining the high mortality which is more likely to relate to inadequate resuscitation with fluid therapy, choice of sedative and anaesthetic agents, intraoperative monitoring, surgical technique and post-operative care. However I would also stress that when sedating for orogastric intubation we advocate ‘judicious’ sedation with cardiorespiratory-sparing agents, typically using a pure opioid [usually already administered shortly after presentation] and a benzodiazepine.

Love to hear any thoughts/comments…