OBJECTIVE: To describe current opinions about stress-related mucosal disease (SRMD) prevention in Canadian pediatric intensive care units (PICUs).

METHODS: A 22-question survey covering several aspects of SRMD was sent to all identified PICU attendings in Canada.

RESULTS: Sixty-eight percent of identified attendings completed the questionnaire. Thirty-eight percent were based in Quebec, 31% in Alberta, and 31% from other provinces. Most attendings (78%) had worked in a PICU for 6 years or more. When asked about risk factors for prescribing SRMD prevention drugs (more than 1 answer was accepted), the most popular answers were prior history of gastric ulceration/bleeding (33 respondents), coagulopathy (28 respondents), and major neurologic insult (18 respondents). Almost half of the attendings (48%) mentioned that they prescribe SRMD prophylaxis directly upon PICU admission to more than 25% of their patients. Forty-nine percent of respondents subjectively estimated that clinically significant upper gastrointestinal bleeding (UGIB; defined as UGIB associated with either hypotension, transfusion within 24 hours of the event, or death) occurred in less than 1% of their patients. Fifty-seven respondents (93%) used ranitidine as first-line therapy (average dose: 4.1 mg/kg/day, mainly intravenously). As second-line therapy, 32 attendings (52%) used pantoprazole and 13 (21%) used omeprazole.

CONCLUSIONS: Despite the paucity of guidelines on SRMD prevention and the low reported incidence of clinically significant UGIB, SRMD prevention is frequently used in Canadian PICUs. Ranitidine is the first-line drug used by most attendings.

Stress-related mucosal disease (SRMD) is a comorbidity often seen in pediatric intensive care units (PICUs). Depending on the definition used, its prevalence in critically ill children ranges from 6% to 10%, with nearly 1% of cases being referred to as clinically significant upper gastrointestinal bleeding (UGIB), defined as UGIB associated with hypotension, need for transfusion in the first 24 hours after diagnosis, or death.1–3 Mortality in critically ill adult patients with SRMD has been reported to be 5 times higher than for those without it. Stress-related mucosal disease is known to have significant clinical and economic impacts. Gauvin et al4 found that SRMD in pediatric patients is associated with lower hemoglobin concentration, a higher rate of blood transfusions, prolonged mechanical ventilation, and prolonged PICU hospitalization, the latter factors known to be positively correlated with increased morbidity and mortality. They also showed that hospitalization fees were approximately 4 times higher for these patients.

Multiple aspects of SRMD have been addressed, albeit more so in adult than pediatric literature. Risk factors (respiratory failure, coagulopathy, and fasting, among others) have been studied in both populations.2,5 Different prophylactic medications have also been studied. Surface agents, H2-receptor antagonists, and proton pump inhibitors (PPIs) are among the most studied.6 While there are no widely accepted clinical guidelines currently available to help clinicians with SRMD prevention in children, there are meta-analyses and guidelines addressing adult prophylaxis.5,7–11 The majority of these documents recommend that H2-receptor antagonists be used to prevent stress ulcer in vulnerable adult ICU patients. Despite these guidelines, SRMD prophylaxis is not uniformly prescribed in adult ICUs. Daley et al12 studied SRMD prevention among American intensivists working in adult ICUs. Most participants believed that SRMD prevention was necessary for their patients, even though clinically significant UGIB was infrequent. The choice of intervention was diverse among these respondents. Most (64%) preferred H2-receptor antagonists but there was an increasing trend of PPI use despite evidence suggesting a relationship between PPI use and nosocomial ventilator-associated pneumonia.13 Currently, there are no data on the pediatric use of SRMD prevention in Canada. The purpose of this study was to assess current opinions among Canadian PICU attendings regarding SRMD prevention, risk factor assessment, and drug prescription.

The Centre Hospitalier Universitaire de Québec Ethics Board (C11-02-160) approved this descriptive study on SRMD prevention in Canadian PICUs. The study population was staff physicians working in 1 of the 16 Canadian PICUs. To identify our population, an email was sent to each Department of Pediatrics and PICU director in Canada, requesting contact information for their current staff. Available PICU Web sites were also browsed for similar information. The sole exclusion criterion was being already involved in the current study as a co-investigator.

An English survey was adapted from a previously published questionnaire.12 It was reviewed and critiqued by 2 pediatric intensive care fellows who were not otherwise involved in this study. Twenty-two questions were included in the final version. Twenty of them were multiple-choice questions. The questionnaire took about 10 minutes to complete. The addressed topics were the definition of SRMD, risk factors for SRMD prophylaxis, and choice of therapeutic intervention. Anonymous demographic data were also included (primary workplace, PICU characteristics, professional experience, and training). The questionnaire is available in the  Appendix.

As per recommendations to increase survey participation,14 the questionnaire, along with information about the study, was sent to the 90 identified attendings twice within a 1-month interval (April 29, 2011, and May 30, 2011). Respondents received questionnaires via both electronic and conventional mailing systems. The electronic version of the questionnaire could be completed and submitted via a Web browser. A blinded third person collated and anonymized the data before sending it to the investigators. After receiving completed questionnaires, data were compiled by using Excel 2007 (Microsoft, Redmond, WA). Confidentiality was ensured during the entire process by using non-identified mailing material.

Demographic Data

Of the 90 identified attendings eligible for our survey, 61 (68%) answered the questionnaire, 8 (13%) responding electronically. Demographic data are shown in Table 1. Forty-eight respondents (78%) had worked in PICU for 6 years or more. Most of the surveyed attendings had been trained and were working in Quebec or Ontario. Respondents represented 14 different institutions spread throughout Canada. The institution with the greatest number of respondents represented 20% of the overall providers surveyed.

Table 1.

Demographic Data of the Respondents

Demographic Data of the Respondents
Demographic Data of the Respondents

SRMD Definition and Epidemiology

Forty-nine percent of respondents subjectively estimated that clinically significant UGIB (defined in the questionnaire as UGIB associated with either hypotension, transfusion within 24 hours of the event, or death) occurred in less than 1% of their patients (Table 2). Respondents most frequently defined failure of their SRMD prevention as blood in gastric aspirate/nasogastric tube (34%), spontaneously externalized visible bleeding (28%), and documented ulcers or gastritis per esophagogastroduodenoscopy (21%).

Table 2.

Estimated Epidemiology of Clinically Significant UGIB in Canadian PICU

Estimated Epidemiology of Clinically Significant UGIB in Canadian PICU
Estimated Epidemiology of Clinically Significant UGIB in Canadian PICU

SRMD Prevention

Respondents were questioned regarding their SRMD prophylaxis assessment. As one of the main goals of this study, respondents were asked about the risk factors used to influence their decision on whether or not to prescribe SRMD prophylaxis. To this end, they could choose up to 3 of 15 risk factors commonly described in the literature, including “other.” Thirty-three respondents (54%) chose prior history of gastric ulceration/bleeding; 28 (46%), coagulopathy; and 18 (30%), major neurologic insults (Figure 1). Almost half of the attendings (48%) prescribed SRMD prophylaxis directly on PICU admission to more than 25% of their patients. The main reasons for discontinuing SRMD prevention were tolerating enteral feedings for 66% of attendings, resolution of all risk factors for 13%, and clinically improved status for 11% (Figure 2). Twenty-one percent of respondents monitored SRMD prevention efficacy, mostly with gastric pH measurement.

Figure 1.

Risk factors* used to prescribe stress-related mucosal disease prevention in pediatric intensive care units.

* Respondents could choose 3 different answers among the list provided in the questionnaire.

† Percentage data based on total number of respondents (n=61).

‡ Platelet count <50,000/mm3 and/or international normalized ratio >1.5 and/or partial thromboplastin time >2 times normal.

§ Glasgow Coma Score <8.

¶ Congenital or acquired (including secondary to medication such as corticosteroids).

Figure 1.

Risk factors* used to prescribe stress-related mucosal disease prevention in pediatric intensive care units.

* Respondents could choose 3 different answers among the list provided in the questionnaire.

† Percentage data based on total number of respondents (n=61).

‡ Platelet count <50,000/mm3 and/or international normalized ratio >1.5 and/or partial thromboplastin time >2 times normal.

§ Glasgow Coma Score <8.

¶ Congenital or acquired (including secondary to medication such as corticosteroids).

Close modal
Figure 2.

Reasons for discontinuing SRMD prophylaxis.

SMRD, stress-related mucosal disease.

Figure 2.

Reasons for discontinuing SRMD prophylaxis.

SMRD, stress-related mucosal disease.

Close modal

Choice of Medication

Ninety-three percent of respondents used ranitidine as first-line therapy (26% because of efficacy, 23% because of cost, and 16% because of ease of administration). As second-line therapy, PPIs were the most commonly used drugs; 52% of attendings used pantoprazole and 21%, omeprazole. First-line therapy was prescribed via intravenous route for 48 respondents (79%), with 5 using a continuous infusion. Mean administered dose of ranitidine was 4.1 mg/kg/day (ranging from 1 to 7 mg/kg/day). Ranitidine was mostly prescribed by using intermittent dosing: 5 respondents (9%) in 2 divided doses, 34 (60%) in 3 divided doses, and 9 (16%) in 4 divided doses. Seven percent of respondents reportedly had an SRMD prevention protocol in their PICU.

To our knowledge, this is the first study describing SRMD prevention among PICU attendings. Results from this study suggest that although SRMD prophylaxis is not prescribed to every patient in Canadian PICUs, it is a relatively common practice. These PICU attendings use several different factors to assess at-risk patients. Ranitidine was by far the most commonly prescribed drug; however, we believe there were no inciting factors favoring ranitidine over other H2-receptor antagonists, since there is no specific formulary for its use in Canada and SRMD prevention protocols were uncommonly used.

Canadian PICU attendings identified prior history of gastric ulceration/bleeding, coagulopathy, and major neurologic insult as major risk factors requiring SRMD prophylaxis. Previously, Lacroix et al1 prospectively studied patients admitted in a Canadian PICU in order to determine the frequency of UGIB (defined as an episode of hematemesis or if any amount of blood was seen in drainage from a nasogastric tube). They found 4 major risk factors for UGIB in critically ill pediatric patients: high Pediatric Risk of Mortality (PRISM) score,15 pneumonia, coagulopathy, and multiple trauma (including severe head trauma). PRISM is a scoring system used to predict survival in ICU patients, using neurologic findings, vital signs, and laboratory results collected during the first 24 hours after ICU admission. In another survey study,12 risk factors identified by ICU attendings were respiratory failure (69%), shock/hypotension (49%), and sepsis (39%). In a mixed medical/surgical PICU population, Gauvin et al4 identified that clinically significant UGIBs were more associated with coagulopathy, use of mechanical ventilation, and PRISM score = 10. Factors considered for SRMD prophylaxis by Canadian PICU attendings in our study differ slightly from those reported risk factors, most likely because of the absence of clear guidelines about SRMD prevention in children, and the extrapolation of adult data. Because of the low incidence of this condition, studies addressing that aspect would be difficult to conduct.

Daley et al12 surveyed a group of American intensive care attendings working with adult patients. Several differences between their results and those of the present study are of interest. Respondents of the study tended to initiate SRMD prevention therapy upon admission to a larger group of patients (94% of respondents initiated SRMD prevention therapy to at least 25% of their patients), compared with our respondents (48%). Guideline availability, increased data in adult patients, and differences in pathologies cared for could possibly explain this difference.

In addition, American ICU attendings prescribed ranitidine to adults less often than the respondents of our study who care for children (22.7% vs. 93%). H2-receptor antagonists were also their most popular choice of drug (63.9%). Cost and ease of use of ranitidine could explain, at least in part, its popularity over PPI and surface agents in Canadian PICUs, as stated by our respondents.

Economic aspects of SRMD prophylaxis have to be considered. Gauvin et al4 showed that hospitalization costs for pediatric patients with SRMD are significantly higher (about 4 times) than for those without this condition. In addition, in an adult care setting, a reported 3- to 4-fold increase in total hospitalization costs was described for 6 patients who had clinically significant bleeding (hemodynamic compromise or transfusion),16 though those 6 patients were already receiving ranitidine prophylaxis. The mean medication cost for SRMD prevention in this study was US $24 to US $36 per patient per stay. The authors concluded that SRMD prophylaxis needs to be rationalized to limit expenses. The relatively low occurrence of SRMD in children, the ever-increasing medication and hospitalization costs for critical care patients, and the lack of centralized statistics about PICU hospitalizations in Canada make financial comparisons even more complex. We believe that cost-effectiveness analysis would be useful in determining the need for pediatric SRMD prevention guidelines.

Although this study has multiple strengths (e.g., the high response rate and the possibility to compare results with previously reported data), it also has its limitations. For instance, the Canadian perspective may not be applied to all settings, considering the intercountry variability in health care systems. The study design could also have introduced unintentional biases. Because of the survey format, respondents could have been influenced by answer lists or the multiple-choice format, which limits the number of possible answers for some questions. Because we mostly derived our answer choices from adult literature, it may not fully represent pediatric practice and may influence results. Recall bias is another potential confounder and our study was not designed to compare stated to actual prescribing habits. Also, despite the minimal presence of formal SRMD prevention protocols within PICUs (7% of respondents), it is possible that local practice patterns within physician groups influence prescription habits. We were, however, statistically underpowered to examine that possible influence. Finally, we did not study SRMD prevention by PICU residents and fellows. Knowing that these workers are often involved in admission and follow-up prescriptions, it would have been interesting to investigate their knowledge and practice. Interestingly, previous studies17 have shown that residents' prescriptions are often inaccurate for SRMD prophylaxis.

We conclude that it is a common practice and that ranitidine is the first-line drug used by most attendings despite the lack of clinical guidelines. Our study shows that the most important risk factors considered by our respondents before prescribing SRMD prevention are prior history of gastric ulceration/bleeding, coagulopathy, and major neurologic insult. Given the low occurrence of clinically significant UGIB in Canadian PICUs and the consensus toward ranitidine use among attendings, we suspect that further studies comparing H2-receptor antagonists with PPI in this population would be unlikely to yield clinically relevant results outside of the context of a randomized clinical trial.

We would like to thank Dr Macha Bourdages, Dr Richard Bélanger, Dr Bryan Lemenchick, Dr Nathalie Laflamme, David Patten (PhD candidate), Dr Tommy Seaborn, and Dr Matthew Weiss for their help in the writing and reviewing processes of this article. An abstract was presented to the 41st Critical Care Congress of the Society of Critical Care Medicine (Houston, Texas; February 2012). An abstract was presented to the 10th Annual Congress of the Quebec Society of Intensivists (Québec, Québec; February 2012). An abstract was presented to the Annual Research Day of the Faculté de Médecine de l'Université Laval (Québec, Québec; May 2012).

ICU

intensive care unit

PICU

pediatric intensive care unit

PPI

proton pump inhibitor

PRISM

Pediatric Risk of Mortality

SRMD

stress-related mucosal disease

UGIB

upper gastrointestinal bleeding

1.
Lacroix
J
,
Nadeau
D
,
Laberge
S
,
et al
.
Frequency of upper gastrointestinal bleeding in a pediatric intensive care unit
.
Crit Care Med
.
1992
;
20
(
1
):
35
42
.
2.
Chaïbou
M
,
Tucci
M
,
Dugas
MA
,
et al
.
Clinically significant upper gastrointestinal bleeding acquired in a pediatric intensive care unit: a prospective study
.
Pediatrics
.
1998
;
102
(
4
):
933
938
.
3.
Cook
DJ
,
Fuller
HD
,
Guyatt
GH
,
et al
.
Risk factors for gastrointestinal bleeding in critically ill patients: Canadian Critical Care Trials Group
.
N Engl J Med
.
1994
;
330
(
6
):
377
381
.
4.
Gauvin
F
,
Dugas
MA
,
Chaïbou
M
,
et al
.
The impact of clinically significant upper gastrointestinal bleeding acquired in a pediatric intensive care unit
.
Pediatr Crit Care Med
.
2001
;
2
(
4
):
294
298
.
5.
Cook
D
,
Heyland
D
,
Griffith
L
,
et al
.
Risk factors for clinically important upper gastrointestinal bleeding in patients requiring mechanical ventilation: Canadian Critical Care Trials Group
.
Crit Care Med
.
1999
;
27
(
12
):
2812
2817
.
6.
Reveiz
L
,
Guerrero-Lozano
R
,
Camacho
A
,
et al
.
Stress ulcer, gastritis, and gastrointestinal bleeding prophylaxis in critically ill pediatric patients: a systematic review
.
Pediatr Crit Care Med
.
2010
;
11
(
1
):
124
132
.
7.
Cook
DJ
,
Witt
LG
,
Cook
RJ
,
Guyatt
GH.
Stress ulcer prophylaxis in the critically ill: a meta-analysis
.
Am J Med
.
1991
;
91
(
5
):
519
527
.
8.
Cook
DJ
,
Reeve
BK
,
Guyatt
GH
,
et al
.
Stress ulcer prophylaxis in critically ill patients: resolving discordant meta-analyses
.
JAMA
.
1996
;
275
(
4
):
308
314
.
9.
Tryba
M
,
Cook
D.
Current guidelines on stress ulcer prophylaxis
.
Drugs
.
1997
;
54
(
4
):
581
596
.
10.
ASHP therapeutic guidelines on stress ulcer prophylaxis: ASHP Commission on Therapeutics and approved by the ASHP Board of Directors on November 14, 1998
.
Am J Health Syst Pharm
.
1999
;
56
(
4
):
347
379
.
11.
Terdiman
JP
,
Ostroff
JW.
Gastrointestinal bleeding in the hospitalized patient: a case-control study to assess risk factors, causes, and outcome
.
Am J Med
.
1998
;
104
(
4
):
349
354
.
12.
Daley
RJ
,
Rebuck
JA
,
Welage
LS
,
Rogers
FB.
Prevention of stress ulceration: current trends in critical care
.
Crit Care Med
.
2004
;
32
(
10
):
2008
2013
.
13.
Cook
DJ
,
Reeve
BK
,
Scholes
LC.
Histamine-2-receptor antagonists and antacids in the critically ill population: stress ulceration versus nosocomial pneumonia
.
Infect Control Hosp Epidemiol
.
1994
;
15
(
7
):
437
442
.
14.
Fairman
K.
Going to the source: a guide to using surveys in health care research
.
J Manag Care Pharm
.
1999
;
5
(
2
):
150
159
.
15.
Pollack
MM
,
Ruttimann
UE
,
Getson
PR.
Pediatric risk of mortality (PRISM) score
.
Crit Care Med
.
1988
;
16
(
11
):
1110
1116
.
16.
Erstad
BL
,
Camamo
JM
,
Miller
MJ
,
et al
.
Impacting cost and appropriateness of stress ulcer prophylaxis at a university medical center
.
Crit Care Med
.
1997
;
25
(
10
):
1678
1684
.
17.
Liberman
JD
,
Whelan
CT.
Brief report: reducing inappropriate usage of stress ulcer prophylaxis among internal medicine residents
.
J Gen Intern Med
.
2006
;
21
(
5
):
498
500
.
Appendix.

Questionnaire.

Appendix.

Continued.

Appendix.

Continued.

Appendix.

Continued.

Disclosure The authors declare no conflicts or financial interest in any product or service mentioned in the manuscript, including grants, equipment, medications, employment, gifts, and honoraria. Dr Jérôme Ouellet had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. Financial support: Fondation des Étoiles.