Children hospitalized for critical asthma (CA) in the pediatric ICU (PICU) are commonly prescribed stress ulcer prophylaxis (SUP) to mitigate risk of gastrointestinal (GI) bleeding. We sought to describe trends for SUP prescribing and explore for differences in rates of GI bleeding, gastritis, and SUP-related complications for those with and without SUP exposure.
We performed a retrospective, multicenter cohort study using the Pediatric Hospital Information System registry among 42 children’s hospitals from 2010 to 2019 including children 3 to 17 years of age admitted to the PICU for CA. Primary outcomes were chronologic and regional variation in SUP prescribing assessed by Joinpoint regression and Pearson’s correlation. Rates of GI bleeding, gastritis, enteric ulceration, and SUP-related complications (C. difficile colitis, necrotizing enterocolitis, and thrombocytopenia) were compared for those with and without SUP exposure.
Of 30 177 children studied, 10 387 (34.4%) received SUP. No episodes of GI bleeding were recorded. One subject developed gastric ulceration and 32 (0.1%) gastritis. Linear trends for SUP were observed with rates increasing from 25.5% in 2010 to 42.1% in 2019 (+1.9% annually). Prescribing varied by institution (range: 5.5% to 97.2%) without correlation to admission volumes. Extremely rare rates of SUP-related complications were noted.
Although children hospitalized for CA routinely receive SUP, no episodes of GI bleeding were noted over a 10-year period. SUP solely for corticosteroid exposure may be unwarranted. We advocate for a targeted approach to SUP considering alternative risk factors for GI bleeding.
Stress ulcer prophylaxis is administered to hospitalized children receiving corticosteroids for critical asthma. However, the incidence of clinically important bleeding and potential risk reduction for bleeding as a result of prophylaxis administration has yet to be described for this population.
In this largest cohort study to date assessing children hospitalized for critical asthma over a 10-year period, we observed a stress ulcer prophylaxis rate of 10 387 (34.4%) that increased linearly by +1.9% annually without any episodes of major gastrointestinal bleeding.
Major gastrointestinal (GI) bleeding, defined by the International Society for Thrombosis and Haemostasis, includes (1) fatal bleeding, (2) a fall in hemoglobin of 20 g/L, or (3) bleeding requiring transfusion of ≥2 U of packed red blood cells.1 Acquired bleeding is rare in the pediatric ICU (PICU) setting and is reported in between 0.1% to 2.6%2–6 but contributes to mortality, morbidity, and length of stay (LOS).7 Critically illness alone may be a risk for stress-related GI ulceration and bleeding related to impaired splanchnic perfusion, systemic hypoxemia, low gastric pH, and dysregulated mucosal cytoprotection.8–10 Previously described factors associated with stress ulcer-related GI bleeding include mechanical ventilation (MV) >48 hours, hepatic dysfunction, acute kidney injury, shock, severity of illness, multiple comorbidities, and coagulopathy (platelet count <50 × 109/L or international normalized ratio >1.5).5,9,11,12
Children admitted to the PICU for asthma exacerbation, termed critical asthma (CA), are believed to be at risk for stress ulcer-related GI bleeding from concomitant systemic corticosteroids administered as standard asthma management.13,14 As a result, children hospitalized for CA are frequently prescribed stress ulcer prophylaxis (SUP) with histamine-2 blocker (H2B) or proton pump inhibitor (PPI) exposure. To date, limited data suggest a risk reduction for major GI bleeding from SUP in this population.6,12,13,15–17 Furthermore, SUP administration is associated with ventilator associated pneumonia, C. difficile colitis, necrotizing enterocolitis (NEC), and thrombocytopenia.18–24 In a single-center, retrospective study comparing rates of GI bleeding among children admitted for CA with and without exposure to SUP, a recent publication revealed no episodes of GI bleeding or SUP-related adverse events.17
To build on existing levels of evidence, we queried the Pediatric Health Information System registry to identify national trends in SUP prophylaxis prescribing over a decade and estimate rates of major GI bleeding and gastritis among children hospitalized for CA. We hypothesized GI bleeding and gastritis were rare and no different for those prescribed SUP.
Methods
Data Source
We conducted a retrospective, cohort study assessing inpatient encounters from January 2010 through December 2019 using the Pediatric Health Information System (PHIS) registry. This administrative database contains inpatient, emergency department, ambulatory surgical, and observational encounter-level data from over 50 not-for-profit, tertiary care pediatric hospitals affiliated with the Children’s Hospital Association (Lenexa, KS). A complete listing of hospital demographics can be found at www.childrenshospitals.org/PHIS/. For external benchmarking, participating hospitals provide discharge and encounter data including demographics, diagnoses, and procedures. Encounter data encompasses demographics and up to 41 International Classification of Diseases, 9th/10th revisions, Clinical Modification diagnostic codes (ICD-9/10-CM). The registry includes clinical transaction classification codes used to identify pharmaceutical, supply, laboratory, and radiographic therapies and procedures. Data are deidentified and subjected to reliability and validity checks before inclusion. For this study, data from 42 centers were included. This protocol was reviewed and approved by our local institutional review board (00150245).
Study Participants and Cohorts
Inclusion criteria were encounters from children 3 through 17 years of age with the principal diagnosis of CA identified by ICD-9/10-CM, admitted to the PICU, and prescribed systemic corticosteroids including dexamethasone, methylprednisolone, prednisone, prednisolone, and hydrocortisone. Exclusion criteria included presence of GI bleeding on admission, gastroesophageal reflux disease (GERD), eosinophilic esophagitis, acute chest syndrome, congenital heart disease, interstitial lung disease, bronchopulmonary dysplasia, cystic fibrosis, pulmonary hypertension, tracheostomy dependence, neuromuscular disease, and if the LOS was <1 day. Children <3 years of age were excluded to limit misclassification for respiratory illness related to bronchiolitis that mimics asthma exacerbation in young children. Encounters with a preadmission diagnosis of GERD and eosinophilic esophagitis were excluded to limit bias from alternative indications for acid suppressive therapies. Study cohorts were identified by exposure to SUP (either H2Bs or PPIs) charged on the first encounter day to indicate providers’ intention to provide prophylaxis.
Study Outcomes and Definitions
Primary outcomes were (1) SUP prescribing rates assessed both chronologically and by unique hospital center and (2) major GI bleeding events identified by ICD-9/10-CM documentation (included codes: K250, K252, K2901, K2961, 5780, 5781, 5789, K921, K920, and K922) and using surrogate markers of severe bleeding including prescriptions for octreotide, PPI infusion, and packed red blood cell transfusion. Secondary outcomes, defined by ICD-9/10-CM documentation, included the frequency of gastritis, gastric ulcers, intestinal perforation, and SUP-related adverse events (thrombocytopenia, NEC, and C. difficile colitis). Descriptive data included patient age, National Heart Lung and Blood Institute (NHLBI) Asthma severity classification (recorded within PHIS after 2016), PHIS pediatric medical complexity algorithm scores, prescribed adjunctive therapies (terbutaline, aminophylline, ketamine, heliox, and bilevel positive airway pressure), LOS, MV rates, extracorporeal life support rates, and index mortality.
Statistical Analyses
Descriptive statistics are reported throughout the manuscript as percentages, means ± SDs, or medians [interquartile range] depending on data distribution and type. Comparative analyses included χ2 or Fisher’s exact test for categorical variables and student’s t or Wilcoxon rank sum tests for continuous variables. Chronologic trends in SUP prescribing were examined using Joinpoint regression software v4.7 (National Cancer Institute) reporting annual percent change (APC) and t-statics. Pearson’s correlation was applied to assess for variation in center SUP-prescribing rate explained by center CA admission volumes. Statistical tests were 2-sided, and Type I error was set at 0.05. Encounters were used as the unit of analysis and assumed independent. Descriptive and comparative analyses were completed using Stata© v15.1 software (Stata Corporation, College Station, TX).
Results
Study Sampling and Cohorts
We identified 34 544 inpatient encounters with CA among 42 children’s hospitals that met inclusion criteria. Of those, 4367 were excluded (common exclusions were preexisting GERD [n = 1544], congenital heart disease [n = 1264], oncologic malignancy [n = 494], tracheostomy dependence [n = 417], and PICU LOS < 1 day [n = 241]), resulting in an overall sample of 30 177 subjects. Of those, 10 387 (34.4%) were prescribed SUP and 19 790 (65.6%) were not. A CONSORT diagram is provided (Fig 1). The most common SUP drug class prescribed were H2Bs (n = 8477, 81.6%) as compared to PPIs (n = 1910, 18.4%).
General Characteristics and Clinical Outcomes
Study subjects had a mean age of 8.3 ± 3.8 years, were more commonly male (ratio of 1.45:1) and 66.3% had a NLHBI classification of moderate to severe persistent asthma. The mean LOS was 3.2 ± 2.1 days, 3417 (11.3%) were provided MV during the PICU hospitalization, and 5 (0.02%) encounters resulted in mortality. A complete list of patient demographics, severity of illness indices, and pharmaceutical/respiratory interventions, LOS, and mortality data are listed in Table 1. Of note, children prescribed SUP had lower frequency of severe persistent NHLBI classification (27.6% vs 22.9%, P < .001) and lower prescribing rates for pharmaceutical CA interventions including terbutaline (14.6% vs 12.4%, P < .001), aminophylline (9.6% vs 3.6%, P < .001), heliox (12.2% vs 10.3%, P < .001), and ketamine (5.4% vs 4%, P < .001). In contrast, noninvasive positive pressure ventilation (both bilevel positive airway pressure and continuous positive airway pressure) were more commonly prescribed to children receiving SUP (27.3% vs 22.1%, P < .001).
Variables . | Study Sample(n = 30 177) . | Yes SUP(n = 10 387) . | No SUP(n = 19 790) . | P . |
---|---|---|---|---|
Age, y, mean ± SD | 8.3 ± 3.8 | 8.1 ± 3.8 | 8.3 ± 3.8 | <.001 |
Sex, male:female ratio | 1.45:1 | 1.46:1 | 1.42:1 | .213 |
Pediatric medical complexity algorithm category, median (interquartile range) | 2 (2–2) | 2 (2–2) | 2 (2–2) | <.001 |
NHLBI chronic asthma classification, %a | ||||
Intermittent | 16.1 | 16.9 | 14.9 | .009 |
Mild persistent | 17.6 | 18.5 | 16.6 | .011 |
Moderate persistent | 41.4 | 41.7 | 40.9 | .407 |
Severe persistent | 24.9 | 22.9 | 27.6 | <.001 |
Prescribed medications, % | ||||
Aminophylline | 5.7 | 3.6 | 9.6 | <.001 |
Inhaled nitric oxide | 0.2 | 0.1 | 0.37 | <.001 |
Ketamine | 4.7 | 4.4 | 5.4 | <.001 |
Terbutaline | 13.2 | 12.4 | 14.6 | <.001 |
Volatile anesthetic agents | 0.03 | 0.02 | 0.05 | .291 |
Prescribed respiratory interventions, % | ||||
Bilevel positive airway pressure | 7.7 | 8.4 | 6.2 | <.001 |
Continuous positive airway pressures | 17.9 | 19 | 15.9 | <.001 |
Heated, humidified high-flow nasal cannula | 3.8 | 3.7 | 3.8 | .591 |
Heliox | 11 | 10.3 | 12.2 | <.001 |
Invasive mechanical ventilation | 11.3 | 11.4 | 11.2 | .702 |
Extracorporeal life support | 0.28 | 0.26 | 0.31 | .421 |
Length of stay, d, mean ± SD | 3.2 ± 2.1 | 3 ± 1.8 | 3.4 ± 2.6 | <.001 |
Index mortality, % | 0.02 | 0.02 | 0.02 | >.999 |
Variables . | Study Sample(n = 30 177) . | Yes SUP(n = 10 387) . | No SUP(n = 19 790) . | P . |
---|---|---|---|---|
Age, y, mean ± SD | 8.3 ± 3.8 | 8.1 ± 3.8 | 8.3 ± 3.8 | <.001 |
Sex, male:female ratio | 1.45:1 | 1.46:1 | 1.42:1 | .213 |
Pediatric medical complexity algorithm category, median (interquartile range) | 2 (2–2) | 2 (2–2) | 2 (2–2) | <.001 |
NHLBI chronic asthma classification, %a | ||||
Intermittent | 16.1 | 16.9 | 14.9 | .009 |
Mild persistent | 17.6 | 18.5 | 16.6 | .011 |
Moderate persistent | 41.4 | 41.7 | 40.9 | .407 |
Severe persistent | 24.9 | 22.9 | 27.6 | <.001 |
Prescribed medications, % | ||||
Aminophylline | 5.7 | 3.6 | 9.6 | <.001 |
Inhaled nitric oxide | 0.2 | 0.1 | 0.37 | <.001 |
Ketamine | 4.7 | 4.4 | 5.4 | <.001 |
Terbutaline | 13.2 | 12.4 | 14.6 | <.001 |
Volatile anesthetic agents | 0.03 | 0.02 | 0.05 | .291 |
Prescribed respiratory interventions, % | ||||
Bilevel positive airway pressure | 7.7 | 8.4 | 6.2 | <.001 |
Continuous positive airway pressures | 17.9 | 19 | 15.9 | <.001 |
Heated, humidified high-flow nasal cannula | 3.8 | 3.7 | 3.8 | .591 |
Heliox | 11 | 10.3 | 12.2 | <.001 |
Invasive mechanical ventilation | 11.3 | 11.4 | 11.2 | .702 |
Extracorporeal life support | 0.28 | 0.26 | 0.31 | .421 |
Length of stay, d, mean ± SD | 3.2 ± 2.1 | 3 ± 1.8 | 3.4 ± 2.6 | <.001 |
Index mortality, % | 0.02 | 0.02 | 0.02 | >.999 |
NHLBI classification data available only from 2016 onward including 10 334 encounters.
SUP Prescribing Trends
Participating site CA admission volume and SUP prescribing rates are depicted in Fig 2. Mean site SUP prescribing rates varied widely by institution with a mean of 62.4 ± 29% (ranging from 5.9% to 97.2%). A Pearson's product-moment correlation was run to assess for a potential relationship between site SUP prescribing rates and total site CA admission volume that did not reveal any relevant associations with an r2 = 0.019 (P = .279). Stress-ulcer prophylaxis prescribing rates increased from 25.5% in 2010% to 42.1% in 2019. On completing a Joinpoint regression analysis, a linear relationship (Fig 3) without breakpoints represented the most ideal chronologic trend for SUP prescribing during the study period with a mean annual increase of 1.9 ± 6%.
GI Bleeding, Gastritis, and SUP-Related Adverse Events
During the study period, only 1 subject receiving SUP was subsequently diagnosed with GI ulceration. No GI bleeding events were observed for the study sample including surrogate markers for GI bleeding such as transfusions of packed red blood cells or prescriptions for octreotide. Gastritis was noted in 32 (0.1%) subjects, and rates did not differ for children with and without prescribed SUP (0.11% vs 0.1%, P = .706). No episodes of GI ulceration were documented. Sucralfate was prescribed in 34 (0.11%) subjects and was more common among those already receiving SUP (0.22% vs 0.11%, P < .001). With regard to SUP-related adverse events, no diagnoses of C. difficile colitis or NEC were noted. Thrombocytopenia was observed in 24 (0.08%) subjects and was less frequent among those prescribed SUP (0.05% vs 0.14%, P = .008).
Discussion
In this retrospective cohort study using data from 42 children’s hospitals over a 10-year period, we observed that 34.4% of children aged 3 to 17 years hospitalized for CA in the PICU were prescribed SUP on admission. In 2019, SUP prescribing rates were 42.1%, and this practice appears to be increasing linearly by a mean of 1.9% annually since 2010. Rates of SUP were not correlated to center CA admission volume and ranged widely by participating sites from 5.9% to 97.2%. Importantly, not 1 episode of GI bleeding was recorded including surrogate markers for major bleeding such as octreotide infusions and packed red blood cell transfusions. Gastritis and GI ulceration diagnoses were equally rare and not detectably different for children with and without SUP exposure. This is the largest report to date demonstrating the potential over prescription of SUP in this extremely common PICU population. Given these data and recent reports on potential harms related to H2B and PPI exposure, the practice of applying SUP specifically to reduce the risk of bleeding related to corticosteroid exposure in CA management must be strongly reconsidered. Prospective trials, albeit unlikely feasible given the low event rates, are required to determine risk-reductions associated with SUP in this population. We suggest clinicians apply a targeted assessment including other contributing factors that may heighten the risk of stress ulcer-related major GI bleeding before prescribing SUP.
Standard of care for CA includes systemic corticosteroids for management of inflammatory dysregulation associated with asthma exacerbation.14 High-dose glucocorticoid therapy may inhibit prostaglandin synthesis, which thereby impairs gastric alkalinization and predisposes children to gastritis and ulceration.13 Stress-related mucosal disease may further be exacerbated by mucosal ischemia and can be heightened in the setting of CA if there exists concomitant hypoxemia, acidosis, and low gastric pH.14 These factors, along with apprehension related to the morbidity of acquired GI bleeding, likely contribute to the practice of SUP administration after corticosteroid exposure among intensivists caring for children with CA.
In a recent multicenter observational cohort study by Duffet et al,25 378 children admitted to 7 Canadian academic PICUs requiring mechanical ventilation were studied for trends in SUP prescription and rates of clinically important bleeding. They observed major GI bleeding among only 3 children (0.8%), none of whom received systemic corticosteroids. In the same study, complications of SUP were noted in 1% and included C. difficile–associated diarrhea. In a metanalysis of 57 randomized clinical trials by Alhazzani et al,18 rates of GI bleeding in critically ill subjects were no different in those receiving SUP versus placebo. Chaibou et al2 assessed over 800 children admitted to the PICU with acute respiratory failure and found those with a primary diagnosis of asthma and pneumonia were at increased odds of receiving SUP. In the same analysis, independent risk factors for GI bleeding were respiratory failure and preexisting coagulopathy. These findings echo our results and suggest the presence of established risk factors rather than corticosteroid exposure alone should be included in the decision to prescribe SUP for children with CA.
The absence of data to support the use of SUP for CA in pediatrics is also evident from adult publications. In a systematic review by Narum et al13 including over 33 000 adults assessing complication rates after corticosteroid exposure, the incidence of GI bleeding was observed to be 2.9% for those with corticosteroid exposure and to 2% for placebo. In a multicenter, blinded trial of adults admitted to intensive care units (the SUP-ICU trial), similar rates of mortality and clinically significant bleeding were noted regardless of treatment with SUP as compared to placebo.26 In this analysis of 3298 subjects, major GI bleeding rates were 2.5% among high-risk adults who received SUP as compared to 4.2% who received placebo (relative risk 0.58, 95% confidence interval: 0.4–0.86). The reduction in GI bleeding as compared to our data may reflect differences in sampling technique including subject age and severity of illness.
The typical dosing for systemic corticosteroids for children hospitalized with CA includes 1 to 2 mg/kg per day of enteral prednisolone/prednisone or 2 to 4 mg/kg per day of intravenous methylprednisolone.14,27 Curiously, children admitted for management of acute rheumatologic disease, autoimmune disease, or infantile spasms routinely receive pulsed, high-dose systemic corticosteroids as part of standard care.28–30 Dosing for pulsed corticosteroids are in excess of 4 to 20 times that of dosing applied for CA. Yet, SUP are not recommended by pediatric and adult gastrointestinal societies for the management of these conditions including for acute exacerbations of Crohn’s disease or ulcerative colitis (diseases characterized by the presence of enteric ulceration). Our data are consistent with adult and pediatric literature and suggest alternative risk factors and mitigating strategies should be considered.
While the presence of coagulopathy is a known risk for major GI bleeding, data published by Sochet et al17 suggest providers caring for children with CA do not routinely assess coagulation assays for children admitted for CA. In their retrospective, descriptive study assessing SUP practices for PICU patients with CA, only 1.8% had coagulation factors drawn. A metaanalysis by Huang et al31 found patients receiving enteral nutrition may be have reduced GI bleeding events as compared to those receiving pharmacologic SUP and experience lower risk of hospital-acquired pneumonia. These data are echoed in observational data from Nourian et al who showed similar findings when comparing H2Bs to enteral nutrition alone as stress ulcer prophylaxis.32 Sochet et al17 noted 51% of children in the PICU concurrently received enteral nutrition, and it remains unclear if enteral nutrition impacted primary and secondary study outcomes at all. Given the brief observed LOS, it is unclear if any children receive sufficient exposure to SUP to achieve suppression of gastric pH that could mitigate the risk of GI bleeding. Although not assessed in our study sample, previous reports suggest SUP is cost-effective in high-risk patients but likely not in low-risk patients in which drug-acquisition costs are increased but not offset by a reduction in outcome-related costs.17,33
Limitations
The retrospective methodology of this study limits the ability to determine causality, the indication for SUP as intended by prescriber, and definitive identification of primary study outcomes. The primary limitations of PHIS registry data are the absence of patient-level data granularity, asthma-specific severity of illness metrics, and detailed specifications for asthma management at each of the 42 centers included for study. The low frequency of major GI bleeding and gastritis precluded the use of associative statistics as part of our analytic plan. Data are dependent on ICD-9/10 and clinical transaction classification codes to classify encounters, and errors related to misclassification may explain our study observations. The presence of these billing codes does not explicitly verify events occurred or that patients received medications and charged therapies. We are not able to comment on other factors known to impact rates of ulcer-related GI bleeding including the timing, route, and volume of enteral nutrition. PHIS data correspond to date of service, and the exact time of day are not recorded resulting in a potential 47.9-hour window in which a unique code may represent an event, outcome, procedure, or administered medication. Specifically, the duration of exposure for SUP cannot be reliably obtained within PHIS data and, therefore, we cannot account for drug dose or intensity as it related to GI-bleeding events. We assumed encounters were independent, but subjects may represent the same patient over multiple PICU admissions. NHLBI classifications primarily designate chronic disease severity rather than acute indices of severity of illness, and NHLBI data are not recorded from encounters before 2016. Although we excluded children with alternative indications for H2B and PPI exposure (preexisting GERD, eosinophilic esophagitis, etc.), these criteria impact the generalizability of our findings to those specific populations.
Conclusion
In the last decade, SUP prescribing for children admitted to the PICU for CA is common (34.4%) and appears to be linearly increasing by 1.9% annually since 2010. Similar to existing adult and pediatric literature, we observed no major GI bleeding events for the entire study sample regardless of SUP exposure. Whereas these data represent the largest assessment of SUP in pediatric CA to date, prospective controlled trials are required to established risk factors and inform subsequent trial development for SUP in this population. At this time, we recommend a focused assessment for alternative GI bleeding risk factors at the time of hospitalization.
Dr Roberts conceptualized the study, contributed to the interpretation of results, drafted the initial manuscript, and reviewed and revised the manuscript before submission; Dr Roddy conceptualized and designed the study, contributed to the interpretation of results, assisted with initial manuscript drafting, and reviewed and revised of the manuscript before submission; Drs Wilsey, McKinley, and Sanchez-Teppa assisted with study design, contributed to the interpretations or results, and reviewed and revised the final manuscript before submission; Dr Sochet conceptualized and designed the study, extracted registry data, performed statistical analyses, assisted with the initial manuscript drafting, and reviewed and revised the final manuscript before submission; and all authors approved the final manuscript as submitted and agree to be accountable for all aspects of the work.
FUNDING: No external funding.
- CA
critical asthma
- CI
confidence interval
- GERD
gastroesophageal reflux disease
- GI
gastrointestinal
- H2B
histamine-2 blocker
- ICD-9/10-CM
International Classification of Diseases Ninth/Tenth Revision, Clinical Modification
- LOS
length of stay
- MV
mechanical ventilation
- NEC
necrotizing enterocolitis
- NHLBI
National Heart Lung and Blood Institute (NHLBI)
- PHIS
Pediatric Health Information System
- PICU
pediatric intensive care unit
- PPI
proton pump inhibitor
- SUP
stress ulcer prophylaxis
References
Competing Interests
CONFLICT OF INTEREST DISCLOSURES: The authors have indicated they have no conflicts of interest to disclose.
Comments
Rethinking stress ulcer prophylaxis for children with critical illnesses
The study tackles an important topic with implications for managing patients with critical illnesses including and beyond asthma. It appropriately challenges the current over-prescription of SUP which I observe also in my country, Republic of Korea. Further explanation about the negative effects of SUP on the stomach physiologic environment would be valuable. The importance of gastric acid in maintaining GI homeostasis has not been properly appreciated for long time. The beneficial effects of gastric acid have already been well established in literature.2 Suppression of gastric acid possibly delays gastric emptying and motility. Gastric acid facilitates protein and lipid digestion. The dissociation of food calcium complexes is dependent on gastric acidity, and the gastric acid is a prerequisite for mucosal Ca2+ absorption. Gastric acid also protects against acid-resistant enteric pathogens. These beneficial physiologic effects of gastric acid need to be acknowledged in weighing the risks and benefits of SUP.
Ulcer prophylaxis is also too often prescribed in patients receiving corticosteroids. Even though corticosteroids are known to cause gastric ulcer bleeding and delayed tissue repair, the mechanism of corticosteroid-induced GI bleeding or perforation has not been fully explained.3 Contrarily, corticosteroids have also been used to improve GI mucosal lesions caused by inflammatory process, and thus are often the first-line medication in the treatment of inflammatory bowel disease.4 Corticosteroids are also considered as treatment modality for patients with severe ulcerative GI complication of Henoch-Schönlein purpura.5
The findings of the study provide evidence for clinicians to selectively prescribe ulcer only to patients who need it. To this end, I encourage research and clinical efforts to help correctly identify patient population who are truly at elevated risk of stress ulcers. As a follow up study, it would be helpful to investigate whether there was any difference in the adverse and/or beneficial effects of long-term SUP.
Competing Interests
CONFLICT OF INTEREST: There are no potential conflicts of interest to disclose.
References
1. Roberts AR, Roddy M, Wilsey MJ, McKinley SD, Sanchez-Teppa B, Sochet AA. Stress Ulcer Prophylaxis for Critical Asthma. Pediatrics. 2022;149(4):e2021054527
2. Pohl D, Fox M, Fried M, Göke B, Prinz C, Mönnikes H, Rogler G, Dauer M, Keller J, Lippl F, Schiefke I, Seidler U, Allescher HD; Kandahar Study Group. Do we need gastric acid? Digestion. 2008;77(3-4):184-197
3. Narum S, Westergren T, Klemp M. Corticosteroids and risk of gastrointestinal bleeding: a systematic review and meta-analysis. BMJ Open. 2014;4(5):e004587
4. Kucharzik T, Ellul P, Greuter T, Rahier JF, Verstockt B, Abreu C, et al: ECCO Guidelines on the Prevention, Diagnosis, and Management of Infections in Inflammatory Bowel Disease. J Crohns Colitis. 2021;15(6):879-913.
5. Leung AKC, Barankin B, Leong KF. Henoch-Schönlein Purpura in Children: An Updated Review. Curr Pediatr Rev. 2020;16(4):265-276