Stroke: gastrostomy tube feeding decreased death and improved nutritional status in dysphagia.

Clinical bottom line (level 1b)

  1. Patients with acute dysphagic stroke who were given gastrostomy tube feeding were less likely to die than those given nasogastric tube feeding (NNT = 2 at 6 weeks) .
  2. Patients given gastrostomy tube feeding were more likely to have improvement in weight than those given nasogastric tube feeding (NNT = 2 at 1 weeks) .
Norton et al: British Medical Journal 1996; 312: 13-16
Expires November 2002

The study

Unblinded ?concealed randomised trial with intention-to-treat
Setting: one university hospital and one district general hospital, UK

30 patients (aged mean 78 years, 63% female) acute cerebrovascular accident with persisting dysphagia for eight or more days

Excluded if
  • previous history of gastrointestinal disease whichwould preclude siting a gastrostomy tube
  • unfit for upper gastrointestinal endoscopy and intravenous sedation


    • Control Group: (n = 14, 14 analysed): enteral nutrional support via a fine bore nasogastric tube
    Experimental Group: (n = 16, 16 analysed): enteral nutritional support via a gastrostomy tube
    Sedation was induced by using 5-10 mg diazepam and a prophylactic antibiotic (cefuroxime 750 mg intravenously) was administered at the same time. Rate of delivery of feed was 50 ml per hour for the first 24 hours gradually increaseing to an average of 100 ml per hour.
    100% followed for 6 weeks

    The evidence

    Outcome Time to outcome CEREERRRR
    (95% CI)    		ARR
    (95% CI)    		NNT
    (95% CI)
    mortality 6 weeks 8
    (57.1%)    		 2
    (12.5%)    		 78.0%
    (14.0% to 94.0%)    		 44.6%
    (14.1% to 75.2%)    		 2
    (1 to 7)
    improvement in weight 1 weeks 1
    (12.5%)    		 10
    (76.9%)    		 74%
    (26% to 91%)    		 64.4%
    (32.0% to 96.8%)    		 2
    (1 to 3)
    improvement in haemoglobin (g/L) 1 weeks 2
    (20.0%)    		 4
    (26.7%)    		 8.00%
    (-42% to 41%)    		 6.67%
    (-26.7% to 40.1%)    		 15
    (NNT = 2 to infinity;
    NNH = 4 to infinity)
    improvement in albumin (g/L) 1 weeks 1
    (10.0%)    		 9
    (60.0%)    		 56%
    (15% to 77%)    		 50.0%
    (19.0% to 90.0%)    		 2
    (1 to 5)
    improvement in mid-arm circumference (cm) 1 weeks 0
    (0.00%)    		 6
    (46.2%)    		 46%
    (11% to 67%)    		 46.2%
    (19.1% to 73.3%)    		 2
    (1 to 5)

  • There was only 70% follow-up for measurement of weight, haemoglobin, albumin and mid-arm circumference.

    • Citation

    1. Norton B, Homer-Ward M, Donnelly MT, et al: A randomised prospective comparison of percutaneous endoscopic gastrostomy and nasogastric tube feeding after acute dysphagic stroke. British Medical Journal 1996; 312: 13-16
    Contributor: Clare Wotton and Musab Hayatli, November 1999
    Reviewer:

    Clinical Question.
    Patient acute dysphagic stroke
    Intervention or Exposure percutaneous endoscopic gastrostomy
    Comparison nasogastric tube feeding
    Outcome mortality, treatment failure and nutritional status