Endoscopy 2019; 51(04): S85
DOI: 10.1055/s-0039-1681418
ESGE Days 2019 oral presentations
Friday, April 5, 2019 17:00 – 18:30: PEG Club B
Georg Thieme Verlag KG Stuttgart · New York

RADIOLOGICALLY VS ENDOSCOPICALLY-PLACED GASTROSTOMY FEEDING TUBES: AN AUDIT OF CURRENT PRACTICE AND CLINICAL OUTCOMES IN A LARGE, MULTI-SITE UK NHS TRUST

S Pannick
1   Charing Cross Hospital, London, United Kingdom
,
L Hicks
1   Charing Cross Hospital, London, United Kingdom
,
J Kim
1   Charing Cross Hospital, London, United Kingdom
,
Z Velji
1   Charing Cross Hospital, London, United Kingdom
,
K Colucci
1   Charing Cross Hospital, London, United Kingdom
,
A Wright
1   Charing Cross Hospital, London, United Kingdom
,
W Howson
1   Charing Cross Hospital, London, United Kingdom
› Author Affiliations
Further Information

Publication History

Publication Date:
18 March 2019 (online)

 

Aims:

The optimum method for gastrostomy tube placement is unclear. A Cochrane review found insufficient evidence to promote either percutaneous endoscopic gastrostomy (PEG) or radiologically inserted gastrostomy (RIG). Here, we audit practice and clinical outcomes for gastrostomies across our three acute hospitals.

Methods:

We searched the electronic medical record for patients undergoing their first attempted PEG or RIG insertion between 01/01/17 and 31/12/17. Indications, procedure details and 30-day complications were identified retrospectively. Summary statistics, group comparisons (chi-squared test) and multivariate logistic regression were calculated in Stata 14.2.

Results:

155 patients were identified; most had RIGs (85.2%). The median age was 64, and 5.8% had dementia; median pre-procedure CRP was 11.6 mg/l. The most common indications were unspecified dysphagia (45.2%), head or neck cancer (36.8%), and stroke (12.3%). 40.2% RIG patients had no documented contraindication to PEG. Patients seen by a nutrition specialist were significantly more likely to have a PEG (23.2% vs. 1.7%, p < 0.001).

Tab. 1:

30-day complications after gastrostomy insertion

Complication

Total cohort % (n)

PEG % (n)

RIG % (n)

P value (chi-square test)

Aspiration pneumonia

22.6% (35/155)

17.4% (4/23)

23.5% (31/132)

0.52

Tube blockage or displacement

10.3% (16/155)

4.4% (1/23)

11.4% (15/132)

0.31

Procedure-related readmission

5.2% (8/155)

4.4% (1/23)

5.3% (7/132)

0.85

Death

5.2% (8/155)

4.4% (1/23)

5.3% (7/132)

0.85

30-day complication rates are shown in Table 1. Peri-procedural hypoxia was more common with PEG (13.0% vs. 1.5%, p = 0.004). In a multivariate model accounting for age and CRP, tube type was not significantly associated with any complication. Higher CRP was associated with an increased risk of post-procedure bleeding (OR 1.04, p = 0.02).

Conclusions:

In this cohort, 30-day complications were very common. Peri-procedural hypoxia was more common with PEG, but aspiration pneumonia and tube displacement may have been more common with RIG. A randomised trial would better establish benchmark quality metrics, to be implemented by nutrition teams for optimal tube selection.