EBQ:Nasogastric aspiration and lavage in emergency department patients with hematochezia or melena without hematemesis: Difference between revisions
ClaireLewis (talk | contribs) |
|||
(22 intermediate revisions by 3 users not shown) | |||
Line 13: | Line 13: | ||
| fulltexturl= http://onlinelibrary.wiley.com/doi/10.1111/j.1553-2712.2009.00609.x/full | | fulltexturl= http://onlinelibrary.wiley.com/doi/10.1111/j.1553-2712.2009.00609.x/full | ||
| pdfurl=http://onlinelibrary.wiley.com/doi/10.1111/j.1553-2712.2009.00609.x/pdf | | pdfurl=http://onlinelibrary.wiley.com/doi/10.1111/j.1553-2712.2009.00609.x/pdf | ||
| status=Under Review | |||
}} | }} | ||
Line 19: | Line 20: | ||
==Conclusion== | ==Conclusion== | ||
Nasogastric aspiration has a poor diagnostic performance, and cannot be relied upon to rule out an UGIB as the source of bleeding. | |||
==Major Points== | ==Major Points== | ||
*While NG aspiration is commonly used in order to ascertain the source of bleeding (UGIB vs LGIB), its diagnostic utility in patients with hematochezia or melena without hematemesis is not supported by the literature. | |||
*With included study sensitivities all below 90% and -LR's greater than 0.2, a negative NG aspirate or lavage should not be used to rule out UGIB, or change the clinical management of these patients. | |||
*As the test is rated among the most painful procedures done in the ED, not without complications (0.3%-0.8%), its utility should be considered prior to placement. | |||
*A coordinated plan between the ED & GI teams, as to whether the aspiration results will change EGD timing should be taken into consideration when a decision is made to place, or not place an NG. | |||
==Study Design== | ==Study Design== | ||
Systematic literature review. | Systematic literature review. | ||
*The authors identified 969 studies related to their question, 3 of which met inclusion criteria for the review. | |||
*All three studies were retrospective in design | |||
==Population== | ==Population== | ||
===Patient Demographics=== | ===Patient Demographics=== | ||
Varied, depending on the study. | |||
*Mean age: 66-73 | |||
*Sex: 62-66% male | |||
*One study (Cappell, 2005) only included patients also having an MI | |||
===Inclusion Criteria=== | ===Inclusion Criteria=== | ||
Cross sectional studies in which all patients presenting with melena or hematochezia underwent nasogastric aspiration (with or without lavage) followed by esophagogastroduodenal endoscopy (EGD). | Cross sectional studies in which all patients presenting with melena or hematochezia underwent nasogastric aspiration (with or without lavage) followed by esophagogastroduodenal endoscopy (EGD). | ||
Line 34: | Line 46: | ||
Studies that included patients with esophageal varices, hematemesis, or coffee ground emesis. | Studies that included patients with esophageal varices, hematemesis, or coffee ground emesis. | ||
==Interventions== | ==Interventions== | ||
As this is a review article, the authors performed no interventions. In each study, all patients underwent nasogastric aspiration and EGD. | |||
==Outcomes== | ==Outcomes== | ||
===Primary Outcome=== | ===Primary Outcome=== | ||
Operating characteristics of NG aspiration in diagnosing UGIB in patients with hematochezia or melena without hematemesis | |||
{| {{table}} | |||
| align="center" style="background:#f0f0f0;"|'''''' | |||
| align="center" style="background:#f0f0f0;"|'''Sensitivity''' | |||
| align="center" style="background:#f0f0f0;"|'''Specificity''' | |||
| align="center" style="background:#f0f0f0;"|'''PPV''' | |||
| align="center" style="background:#f0f0f0;"|'''NPV''' | |||
| align="center" style="background:#f0f0f0;"|'''LR+''' | |||
| align="center" style="background:#f0f0f0;"|'''LR-''' | |||
|- | |||
| Aljbereen 2004||68%||54%||41%||78%||1.44||0.61 | |||
|- | |||
| Capped 2005||84%||82%||93%||64%||4.74||0.2 | |||
|- | |||
| Witting 204||42%||91%||81%||61%||4.44||0.65 | |||
|} | |||
===Secondary Outcomes=== | ===Secondary Outcomes=== | ||
Complications from | Complications | ||
*Rate of 1.6% (reported in only one study) | |||
*Included gastric erosions resulting from suctioning, and epistaxis. | |||
=== | ===Other=== | ||
Overall rate of UGIB as source was 32% (Aljebreen), 50% (Witting), and 74% (Cappell) | |||
==Criticisms & Further Discussion== | ==Criticisms & Further Discussion== | ||
*All three papers reviewed are retrospective cross-sectional studies, only one of which blinded data abstractors to outcomes. | |||
*All studies used a different reference or 'gold' standard (active bleeding on EGD vs. stigmata of recent bleeding, etc.) | |||
*Question stills remains - does a positive NG make the procedure worthwhile? In select patients who are unstable and a positive NG may expedite an EGD there may be benefit. In those stable patients however , it is unlikely to change the management course. | |||
==Funding== | ==Funding== | ||
Unknown | |||
==Sources== | ==Sources== |
Latest revision as of 02:17, 14 July 2016
PubMed Full text PDF
Clinical Question
Is nasogastric lavage and aspiration in the ED an effective test to distinguish between an upper versus lower source of GI bleed in patients who have either melena or hematochezia without hematemesis?
Conclusion
Nasogastric aspiration has a poor diagnostic performance, and cannot be relied upon to rule out an UGIB as the source of bleeding.
Major Points
- While NG aspiration is commonly used in order to ascertain the source of bleeding (UGIB vs LGIB), its diagnostic utility in patients with hematochezia or melena without hematemesis is not supported by the literature.
- With included study sensitivities all below 90% and -LR's greater than 0.2, a negative NG aspirate or lavage should not be used to rule out UGIB, or change the clinical management of these patients.
- As the test is rated among the most painful procedures done in the ED, not without complications (0.3%-0.8%), its utility should be considered prior to placement.
- A coordinated plan between the ED & GI teams, as to whether the aspiration results will change EGD timing should be taken into consideration when a decision is made to place, or not place an NG.
Study Design
Systematic literature review.
- The authors identified 969 studies related to their question, 3 of which met inclusion criteria for the review.
- All three studies were retrospective in design
Population
Patient Demographics
Varied, depending on the study.
- Mean age: 66-73
- Sex: 62-66% male
- One study (Cappell, 2005) only included patients also having an MI
Inclusion Criteria
Cross sectional studies in which all patients presenting with melena or hematochezia underwent nasogastric aspiration (with or without lavage) followed by esophagogastroduodenal endoscopy (EGD).
Exclusion Criteria
Studies that included patients with esophageal varices, hematemesis, or coffee ground emesis.
Interventions
As this is a review article, the authors performed no interventions. In each study, all patients underwent nasogastric aspiration and EGD.
Outcomes
Primary Outcome
Operating characteristics of NG aspiration in diagnosing UGIB in patients with hematochezia or melena without hematemesis
' | Sensitivity | Specificity | PPV | NPV | LR+ | LR- |
Aljbereen 2004 | 68% | 54% | 41% | 78% | 1.44 | 0.61 |
Capped 2005 | 84% | 82% | 93% | 64% | 4.74 | 0.2 |
Witting 204 | 42% | 91% | 81% | 61% | 4.44 | 0.65 |
Secondary Outcomes
Complications
- Rate of 1.6% (reported in only one study)
- Included gastric erosions resulting from suctioning, and epistaxis.
Other
Overall rate of UGIB as source was 32% (Aljebreen), 50% (Witting), and 74% (Cappell)
Criticisms & Further Discussion
- All three papers reviewed are retrospective cross-sectional studies, only one of which blinded data abstractors to outcomes.
- All studies used a different reference or 'gold' standard (active bleeding on EGD vs. stigmata of recent bleeding, etc.)
- Question stills remains - does a positive NG make the procedure worthwhile? In select patients who are unstable and a positive NG may expedite an EGD there may be benefit. In those stable patients however , it is unlikely to change the management course.
Funding
Unknown