Stressed Out: Evaluating the Need for

Stress Ulcer Prophylaxis in the ICU

Josh Arnold, PharmD
PGY1 Pharmacy Resident

Pharmacy Grand Rounds

November 8, 2016

©2016 MFMER | slide-1

Objectives
• Identify the significance and potential
mechanisms of stress ulcer development in
critically ill patients
• Discuss evidence for the use of stress ulcer
prophylaxis in specific patient populations
• Describe the implications and potential adverse
effects associated with the use of stress ulcer
prophylaxis

©2016 MFMER | slide-2

Significance Implications
and Populations and adverse
mechanisms effects

©2016 MFMER | slide-3

Epidemiology
• Incidence
• Stress ulcers are present in the majority of
critically ill patients
• Within 24 hours of ICU admission, >75% of patients
demonstrate evidence of mucosal damage
• Approximately 5-25% will have bleeding
• ~1-4% will have clinically significant bleeding

• Complications
• Severe ulceration and bleeding can:
• Lengthen ICU stay by up to 8 days
• Increase mortality as much as 4-fold

Cook DJ, et al. Crit Care. 2001;5:368 - 75

ICU: intensive care unit
Fennerty MB. Crit Care Med. 2002;30(6 Suppl):S351-5.
Laine L, et al. Gastroenterology. 2008;135(1):41-60.
©2016 MFMER | slide-4
Which of the following is a potential
mechanism of stress ulcer development in
the critically ill patient?
A. Increased gastrointestinal motility
B. Decreased splanchnic blood flow
C. Increased bicarbonate secretion
D. Decreased duration of mechanical ventilation

©2016 MFMER | slide-5

 Increased Hypovolemia
Catecholamines
Decreased Cardiac Output
Proinflammatory
Cytokine Release
Increased
Vasoconstriction

Splanchnic Hypoperfusion

Reduced Reduced Decreased Acid

HCO3 Mucosal GI Back-
Secretion Blood Flow Motility Diffusion

Mucosal Vulnerability
Stollman N, et al. J Crit Care. 2005;20(1):35-45.
©2016 MFMER | slide-6
Additional factors

Mechanical • Promotes RAAS activity and catecholamine release

ventilation • High PEEP decrease venous return and reduces preload

• Opiates and sedatives can decrease gut motility and impair

Medications venous return

• Systemic hemodynamic changes due to vasopressor therapy
• Glucocorticoids and NSAIDS

Coagulopathy • Impaired ability to terminate active bleeding

PEEP: positive end-expiratory pressure

RAAS: renin-angiotensin-aldosterone system
Coagulopathy: platelets <50,000; INR >1.5; PTT
>2x control value
Buendgens L, et al. World J Crit Care Med. 2016;5(1):57-64.
Stollman N, et al. J Crit Care. 2005;20(1):35-45.
©2016 MFMER | slide-7
Significance Implications
and Populations and adverse
mechanisms effects

©2016 MFMER | slide-8

In what populations do you recommend
the use of stress ulcer prophylaxis?
A. Patients meeting guideline recommendations
B. Patients I believe are “high-risk”
C. All ICU patients
D. I ask a crystal ball

©2016 MFMER | slide-9

Cook et al. (1994)
• Prospective, multicenter, cohort study
• Primary endpoint: clinically important GI bleeding
• 2252 patients, 674 received SUP
• 71.8% H2RA; 0.3% omeprazole
• 33 patients met primary endpoint (1.5%)
Independent Risk Factors Odds Ratio Significance
Mechanical ventilation >48 hrs 15.6 p < 0.001
Coagulopathy 4.3 p < 0.001

• 1.6% patients diagnosed with sepsis

• 48.5% primary diagnosis of cardiovascular surgery
• Mortality 9.7%
SUP: stress ulcer prophylaxis
GI: gastrointestinal Cook DJ, et al. N Engl J Med. 1994;330(6):377-81.
©2016 MFMER | slide-10
Original Guidelines - AJHP 1999
• Recommended SUP use in “high risk” patients
• With coagulopathies
• Requiring mechanical ventilation for >48 hours
• With history of GI ulceration or bleed within 1 year
• Special populations
• Patients with ≥2 of the following

Occult
ICU stay >1 bleeding High-dose
Sepsis
week lasting ≥6 steroids
days

SUP: stress ulcer prophylaxis

GI: gastrointestinal
ASHP Board of Directors. Am J Health Syst Pharm. 1999;56(4):347-79.
©2016 MFMER | slide-11
Original Guidelines
• Number needed to treat
• Low risk patients: >900
• High risk patients: 30
• Limitations
• Little data on use of PPI therapy
• Low strength of evidence
• Outdated

GCS: glasgow coma score

BSA: body surface area ASHP Board of Directors. Am J Health Syst Pharm. 1999;56(4):347-79.
PPI: proton-pump inhibitor Cash BD. Crit Care Med. 2002;30(6 Suppl):S373-8.
©2016 MFMER | slide-12
What has changed? The 2000s
Mechanical ventilation >48 hrs Coagulopathy

Acute renal insufficiency Acute hepatic failure

Sepsis syndrome High-dose corticosteroids

History of gastrointestinal Thermal injury involving >35%

bleeding BSA

Severe head or spinal cord injury Low intragastric pH

Anticoagulation Hypotension

Major surgery (>4 hrs) Acute lung injury

Renal replacement therapy Number of comorbidities

BSA: body surface area

Quenot JP, et al. Curr Opin Crit Care. 2009;15(2):139-43.
©2016 MFMER | slide-13
 Do all our patients really
need stress ulcer prophylaxis?

©2016 MFMER | slide-14

Krag et al. (2015)
• Prospective, observational, international, 7-day
cohort study
• Primary outcome: clinically important
gastrointestinal bleeding during ICU stay
• Secondary outcomes: overt GI bleeding in ICU; 90
day mortality
• Total of 1,034 patients, with 27 patients
developing clinically important GI bleed

GI: gastrointestinal
Krag M, et al. Intensive Care Med. 2015;41(5):833-45.
©2016 MFMER | slide-15
Krag et al. (2015)
• 73% of patients received acid suppressant
therapy during ICU stay
• 59% of patients with clinically important GI
bleeding received SUP prior to bleed
• 90-day mortality
• Overall: 26.2%
• Without clinically important GI bleed: 25.4%
• With clinically important GI bleed: 55.6%
Independent Risk Factors OR (95% CI) Adj* OR (95% CI)
Overt GI bleeding 1.70 (0.70-4.10) 1.17 (0.43-3.21)
Clinically important GI bleeding 3.72 (1.72-8.04) 1.70 (0.68-4.28)
SUP: stress ulcer prophylaxis
GI: gastrointestinal
*adjusted for: age, gender, comorbidites, SOFA Krag M, et al. Intensive Care Med. 2015;41(5):833-45.
score, mechanical ventilation, coagulopathy, etc ©2016 MFMER | slide-16
When do patients bleed?
9
clinically important bleeds

8
Number of Patients with

7
6
5
4
3
2
1
0
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21
Duration of ICU stay (days)

Krag M, et al. Intensive Care Med. 2015;41(5):833-45.

©2016 MFMER | slide-17
What risks exist?

Independent Baseline Adj Odds Ratio (95% CI)

Risk Factors for clinically important GI bleed
≥3 co-existing diseases 8.9 (2.7-28.8)
Co-existing liver disease 7.6 (3.3-17.6)
Renal replacement therapy 6.9 (2.7-17.5)
Co-existing coagulopathy 5.2 (2.3-11.8)
Acute coagulopathy 4.2 (1.7-10.2)
Use of SUP on ICU day 1 3.6 (1.3-10.2)
Higher organ failure score 1.4 (1.2-1.5)
Mechanical ventilation on ICU day 1 1.3 (0.6-3.1)

Coagulopathy: platelets <50,000 mm and/or INR >1.5

either during current hospitalization or on ICU admission
Krag M, et al. Intensive Care Med. 2015;41(5):833-45.
©2016 MFMER | slide-18
Summary
• Found the incidence of clinically important GI
bleeding was low at 2.6%
• Surprising results:
• Mechanical ventilation was not associated with
increased risk of GI bleeding
• The use of stress ulcer prophylaxis on ICU day 1
was associated with an increased risk of GI bleeding
• Influx of new data poses the question –
Is stress ulcer prophylaxis
benefitting our patients?

©2016 MFMER | slide-19

POP-UP 2016
Primary Objective
Evaluate if prophylactic PPI administration is overtly beneficial or harmful

Design
Single-center, prospective, randomized, double-blind, parallel-group study

Inclusion Criteria
Anticipate mechanical ventilation >24 hours and receive enteral nutrition within 48 hrs

Exclusion Criteria
Use of acid-suppressive therapy prior to admission; admission with GI bleeding, history of PUD,
receiving >100mg daily of prednisolone, recent GI or cardiac surgery

PPI; proton pump inhibitor

PUD: peptic ulcer disease Selvanderan SP, et al. Crit Care Med. 2016;44(10):1842-50.
©2016 MFMER | slide-20
Baseline characteristics
Placebo Pantoprazole
Characteristics
(n=108) (n=106)
Age (yr) 52 (±17) 52 (±18)
APACHE III score 66 (±28) 66 (±26)
End-stage renal failure 0 (0%) 1 (0.9%)
Non-operative 68 (63%) 70 (66%)
Post-operative 31 (29%) 27 (25%)
Placebo Pantoprazole
ICU diagnosis
(n=108) (n=106)
Trauma 32 (30%) 31 (29%)
Neurologic 26 (24%) 35 (33%)
Respiratory/CV/Sepsis 41 (38%) 31 (29%)

Unless specified, data are mean (SD) Selvanderan SP, et al. Crit Care Med. 2016;44(10):1842-50.
©2016 MFMER | slide-21
Outcomes
Placebo Pantoprazole
Major Outcomes
(n=108) (n=106)
Clinically significant GI bleeding 0 (0%) 0 (0%)
Infective ventilator-associated complication 1 (0.9%) 2 (1.9%)
Positive C. difficile infection 0 (0%) 1 (0.9%)

Placebo Pantoprazole
Additional Outcomes Significance
(n=108) (n=106)
Length of stay (days) in the ICU 7 (4-14) 6 (3-11) p = 0.16
Ventilator-free days 21 (4-25) 21 (0-25) p = 0.69
90-day mortality 25 (23.1%) 30 (28.3%) p = 0.33

Selvanderan SP, et al. Crit Care Med. 2016;44(10):1842-50.

©2016 MFMER | slide-22
Strengths
• Prospective, randomized, double-blind trial
• Compared standard of care to placebo
• Only included mechanically ventilated patients
• Low risk of selection bias

Therefore, stress ulcer prophylaxis offers

no benefit compared to placebo…
right?

©2016 MFMER | slide-23

Limitations
• Insufficient sample size
• Low incidence of GI bleed
• Excluded 87% of patients recruited
• Unique characteristics of included patients
• Early implementation of other therapies
• Shorter duration of mechanical ventilation

©2016 MFMER | slide-24

Significance Implications
and Populations and adverse
mechanisms effects

©2016 MFMER | slide-25

Incorrect use of stress ulcer prophylaxis
• Rafinazari et al. (2016)

Design Results Conclusion

• Prospective, single • 160 (80%) • 38.5% of patients did

center received SUP not receive appropriate
SUP on ICU admission
• 200 randomly • 44.4% did not have
selected ICU patients an indication* • 81.2% were
over 9 months with continued on
stay >72 hrs • 6.3% did not receive inappropriate SUP
SUP although upon transfer from ICU
indicated

Rafinazari N, et al. J Res Pharm Pract. 2016;5(3):186-92.

*per ASHP guidelines Buendgens L, et al. World J Crit Care Med. 2016;5(1):57-64.
SUP: stress ulcer prophylaxis
©2016 MFMER | slide-26
Continued use after ICU stay
• Wohlt et al. (2007)
394 patients met eligibility criteria

357 patients prescribed SUP

80% continued SUP on transfer from ICU

60% of SUP continuation was inappropriate

24.4% discharged with inappropriate prescription

Unnecessary cost of $13,973

SUP: stress ulcer prophylaxis Wohlt, et al. Ann Pharmacother. 2007;41(10):1611-6.

©2016 MFMER | slide-27
Which of the following are potential adverse
effects of acid-suppression therapy?
A. Drug interactions
B. Electrolyte abnormalities
C. Clostridium difficile infection
D. Nosocomial pneumonia
E. All of the above

©2016 MFMER | slide-28

Clostridium difficile risk

n Population Findings
Community, PPIs
Meta-analyses1-3 ~800,000
general & ICU increase risk
Intubated >24
OR (CI): 1.29
MacLaren et al. 35,312 hours and SUP
(1.04-1.64)
>48 hours
PPIs may have
Faleck et al. 18,134 ICU ≥3 days
protective effect

1. Kwok CS et al. Am J Gastroenterol 2012;107(7):1011-1019.

2. Janarthanan S et al. Am J Gastroenterol 2012;107(7):1001-1010.
3. Arriola V et al. Infect Control Hosp Epidemiol 2016 Sep 28; Epub ahead of print.
MacLaren R et al. JAMA Intern Med 2014;174(4):564-574.
PPI: proton pump inhibitor Faleck DM et al. Am J Gastroenterol 30 Aug 2016; Advance online publication.
©2016 MFMER | slide-29
Pneumonia risk

n Population Findings
No difference in
Alhazzani et al. 1,100 ICU
PPI vs. H2RA
Intubated >24
PPI increased
MacLaren et al. 35,312 hours and SUP
risk vs. H2RA
>48 hours
No difference in
Alshamsi et al. 1,571 ICU
PPI vs. H2RA

Alhazzani W et al. Crit Care Med. 2013;41(3):693-705.

PPI: proton pump inhibitor MacLaren R et al. JAMA Intern Med 2014;174(4):564-574.
H2RA: H2 receptor antagonist Alshamsi F et al. Crit Care. 2016;20:120.
©2016 MFMER | slide-30
Future Direction
• Further larger RCT are needed evaluating:
• Specific risk factors for development of stress-
related mucosal damage
• Use of SUP compared to placebo in preventing
clinically important GI bleed
• SUP-ICU trial
• ASHP is currently updating their guidelines, with
plans for release in Spring 2017

RCT: randomized control trials

SUP: stress ulcer prophylaxis Møller MH: SUP-ICU. Available at
GI: gastrointestinal https://www.clinicaltrials.gov/ct2/show/NCT02467621.
©2016 MFMER | slide-31
Recommendations
• Optimizing other therapies to prevent mucosal
damage is highest priority
• Short-term use of SUP has a role in preventing
clinically important gastrointestinal bleeding in
high-risk populations
Acute Duration of ICU stay Chronic
Coagulopathy Septic shock
Mechanical ventilation Continuation of
Thermal injury therapy
Renal replacement >35% BSA
therapy

©2016 MFMER | slide-32

Stressed Out: Evaluating the Need for
Stress Ulcer Prophylaxis in the ICU
Josh Arnold, PharmD
PGY1 Pharmacy Resident

Pharmacy Grand Rounds

November 8, 2016

©2016 MFMER | slide-33

PPI vs H2RA
• The use of PPI for SUP is associated with significantly
lower rate of clinically important GI bleeding compared
to H2RA
• Meta-analysis n
Risk reduction Risk reduction
(bleeding) (mortality)
RR = 0.36 RR = 1.01
Alhazzani et al 1720
(95% CI, 0.19-0.67) (95% CI, 0.83-1.24)
Pongprasobchai OR = 0.42
569 n/a
et al (95% CI, 0.20-0.91)
RR = 0.39 RR = 1.05
Alshamsi et al 2117
(95% CI, 0.21-0.71) (95% CI, 0.87-1.27)

PPI: proton pump inhibitor Pongprasobchai et al. J Med Assoc Thai. 2009;92:632–637.
SUP: stress ulcer prophylaxis Alshamsi F et al. Crit Care. 2016;20:120.
GI: gastrointestinal Alhazzani W et al. Crit Care Med. 2013;41(3):693-705.
H2RA: histamine 2 receptor antagonist ©2016 MFMER | slide-34
PPI vs H2RA
120

100
% of time gastric pH >4

80

60 Omeprazole
Ranitidine
40

20

0
Day 1 Day 2 Day 3

Fennerty MB. Crit Care Med. 2002;30(6 Suppl):S351-5.

©2016 MFMER | slide-35
Stressed Out: Evaluating the Need for
Stress Ulcer Prophylaxis in the ICU
Josh Arnold, PharmD
PGY1 Pharmacy Resident

Pharmacy Grand Rounds

November 8, 2016

©2016 MFMER | slide-36