MANAGEMENT OF ACUTE LOWER GASTROINTESTINAL BLEEDING

1
MANAGEMENT OF ACUTE LOWER GASTROINTESTINAL
BLEEDING

• Standard of Care Rounds
• November 11, 2004
• Mark J. Russo, MD
• Advisor Tracey D. Arnell, MD

2
OUTLINE

• Resuscitation
• Clinical Presentation
• Diagnostic Evaluation and
  Non-Surgical Treatment
• Surgical Management

3
RESUSCITATION

• Indications for transfusion
• Profuse bleeding
• Persistent hemodynamic instability despite
  crystalloid resuscitation
• Symptomatic anemia (CP, SOB, orthostasis with Hgb
  lt 10)
• AMI or unstable angina with Hgb lt 10

4
(No Transcript)
5
RESUSCITATION

• Blair SD et al, 1986
• transfused group
• rebleeding 9 pts
• longer clotting times in the transfused group.
• control group
• rebleeding 1 pt
• significantly (Plt0.001) more bleeding in the
  transfusion group
• no difference in mortality (2 vs 1)

6
RESUSCITATION

• Carson et al, 2002
• 300 severely anemic, postop pts who refused blood
  transfusion

7
RESUSCITATION

• Canadian Critical Care Trials Group
• restrictive strategy (Hgb 7.0-9.0)
• age lt 55yo
• less critically ill (APACHE scores lt 20)
• liberal strategy (Hgb 10.0-12.0)
• ischemic cardiac disease

8
RESUSCITATION

• Conclusion
• more blood is not always better
• most pts, esp age lt55 yo and less critically ill,
  keep Hgb 7-8
• pts with ischemic cardiac dz, esp AMI or unstable
  angina, keep Hgb lt 10

9
RESUSCITATION

• Platelets
• Maintain gt 50k in an active bleed
• recommended by multiple consensus statements
• no evidence to support recommendations
• Anti-Platelet Agents (APAs)
• Platelet transfusion no evidence to support

10
44124911
11
ACCESS
12
RESUSCIATION

• Summary
• 80 GI bleeds - even massive GI bleeds - will
  resolve spontaneously
• Get appropriate access 2 large bore IVs
• Hct
• Hgb gt7 - otherwise healthy, age lt 55, less
  critically ill
• Hct gt10 - ischemic cardiac dz, AMI, unstable
  angina

13
RESUSCIATION

• Summary
• Plt
• maintain gt 50k
• platelet transfusion may or may not correct
  platelet dysfunction 20 APA

14
CLINICAL PRESENTATION

• Knowing the site of bleeding is more important
  than knowing the cause
• Goals of Bedside Eval
• differentiate UGI from LGI bleed
• acute LGI bleeding 80 colon 20 SI
• r/o easily treatable causes of bleeding (i.e.,
  hemorrhoid)

15
CLINICAL PRESENTATION

• Hematemesis
• vomiting of blood clots or coffee grounds
• suggestive of an UGI source
• Hematochezia
• the passage of liquid blood or clots per rectum
• 3/4 colonic source 11 prox lig of Treitz
• Melena
• stools with altered blood that are black and
  tarry and have a distinctive odor
• suggestion of UGI source
• may cont for days p bleeding stops

16
CLINICAL PRESENTATION

• NGT Lavage - Witting et al, 2004
• () NGTL
• 23 of pts
• UGI bleeding (LR 11),
• (-) NGTL
• 72 of pts
• offers little information (LR 0.6)

17
CLINICAL PRESENTATION

• Anorectal dz - 10 of LGI bleeding
• Anoscopy - r/o treatable dz

18
CLINICAL PRESENTATION

• Summary
• Presentation correlates not with location but
  with the rate of transit
• Hematemesis almost always UGI
• Hematochezia 3/4 patient with have a LGI source
• Melena more likely UGI than LGI
• NGTL () highly suggestive of UGI

19
LOCALIZATION

• Knowing the site of bleeding is more important
  than knowing the cause
• Modalities
• Tagged RBC Scan

20
BLEEDING SCAN
21
TAGGED RBC SCAN

• Advantages
• Safe, noninvasive
• Readily available
• Detects slow bleeds
• at a rate of 0.1 to 0.5 m/min
• more sensitive than angiography
• 73 - 100

• Disadvantages
• only localizes bleeding to an area of the
  abdomen, not SB vs LB
• not therapeutic
• less specific than endoscopy and angiography

22
TAGGED RBC SCAN
23
TAGGED RBC SCAN

• Hunter JM et al, 1990
• incorrectly localized the bleed 58 of the time
• subgroup of 19 patients who underwent a surgical
  procedure based on () nuclear scan
• 8 (42) had incorrect surgical procedures

24
TAGGED RBC SCAN

• Feingold, 2004
• 13 of 21 (62) with period of instability had ()
  tagged RBC scans
• 6 of 29 (21) stable patients had )() scans
• pts requiring surgery were acccurately localized
• conclusion RBC scan useful if applied
  appropriately

25
TAGGED RBC SCAN

• Useful
• to confirm bleeding
• in planning angiography
• Not useful
• massive bleeds or critical illness

26
COLONOSCOPY

• Advantages
• potential for precise localization
• diagnostic success 51 - 90
• potential for rx
• therapeutic success 69 - 100
• ability to collect pathologic specimens

• Disadvantages
• requires a technically skilled endoscopist, not
  available at all centers
• various rates of rebleeding depending on source
• post bleed transit

27
COLONOSCOPY

• Diagnostic procedure of choice when bleeding has
  stopped
• Many reports of good localization of bleed
  (74-85) even with hematochezia

28
ANGIOGRAPHY

• Advantages
• anatomic localization is accurate
• specificity- 100
• sensitivity 47 (acute bleeding), 30 (recurrent
  hemorrhage)
• 41-86 bleeds localized
• therapeutic intervention

• Disadvantages
• requires active blooding gt 0.5 cc/min

29
ANGIOGRAPHY

• Order of vascular investigation
• superior mesenteric artery
• 50 to 80 of diverticular bleeds
• 100 bleeds from proven angiodysplasia
• inferior mesenteric artery
• celiac artery

30
VASOPRESSIN INFUSION

• Causes reliable arteriolar vasoconstriction and
  bowel contraction, resulting in decreased blood
  flow
• 36-100 will stop bleeding
• gt90 of patients with LGIB due to diverticular
  disease or angiodysplasia
• rebleed rate 26-71
• may be used to temporize bleed prior to surgical
  resection
• avoid in pts with cardiac dz

31
EMBOLIZATION

• definitive means of controlling hemorrhage
• stops bleeding 67-100
• rebleed 0-33
• LGI compared to UGI tract has weaker blood supply
• supplied by end arterties
• 5-21 post-embolic ischemia reported
• 0-40 required emergent lap for bleeding and/or
  ischemia

32
SUPERSELECTIVEEMBOLIZATION

• target artery - the vasa recta also within the
  marginal arteries and the distal intestinal
  arcades
• a number of primary embolic agents
• CPMC IR uses microcoils

33
(No Transcript)
34
SUPERSELECTIVEEMBOLIZATION

• Operator skill success catheter placement
• Funaki et al - 93
• Debarros at al - 100

35
SUPERSELECTIVEEMBOLIZATION

• Kuo et al 2003, 22 patients treated for LGI
  bleeding
• Immediate hemostasis - 100
• Rebleeding - 14
• Objective postembolization f/u
• 1 of 14 had minor ischemic changes
• 0 of 14 had major ischemic changes
• Meta-analysis of 144 cases of LGI
• minor ischemia - 21 (13 of 62)
• infarction - 0

36
SUPERSELECTIVE EMBOLIZATION

• (a) Angio via SMA in man with tics and acute LGI
  hemorrhage _at_ the hepatic flexure.
• (b) Superselective embolization with 2 microcoils
  in the vasa recta
• Kuo J Vasc Interv Radiol, 2003 14, 1503-1509

37
CAPSULE ENDOSCOPY

• Advantages
• Higher yield (50-70) for bleeding than
  enteroscopy (30)
• examination of the entire SB
• Disadvantages
• does not permit tissue sampling
• no therapeutic intervention
• risk of retention
• inexact localization

38
PROVOCATIVE BLEEDING STUDY

• Administration of heparin, thrombolytic agents
• When all other options are exhausted
• yield 30-70

39
INDICATIONS FOR SURGERY

• Bleeding refractory to other therapies
• Hemodynamic instability
• Re-bleeding after non-operative treatment, esp if
  localized

40
SURGICAL MANAGEMENT
41
SURGICAL MANAGEMENT

• In a hemodynamically stable patient additional
  options
• bowel inspection
• LB
• SB
• intraoperative upper GI endoscopy
• intraoperative lower GI endoscopy

42
CONCLUSIONS

• Resuscitation is the first step
• 80 of GI Bleeds resolve spontaneously
• Localization of bleed is the goal
• Directed Segmental resection is preferred
  surgical treatment
• Multiple modalities for dx
• center dependent

43
(No Transcript)
44
(No Transcript)