Gastrointestinal Dysfunction Criteria in Critically Ill Children: The PODIUM Consensus Conference

Abstract

Context:

Prior criteria to define pediatric multiple organ dysfunction syndrome (MODS) did not include gastrointestinal dysfunction.

Objectives:

Our objective was to evaluate current evidence and to develop consensus criteria for gastrointestinal dysfunction in critically ill children.

Data Sources:

Electronic searches of PubMed and EMBASE were conducted from January 1992 to January 2020, using medical subject heading terms and text words to define gastrointestinal dysfunction, pediatric critical illness, and outcomes.

Study Selection:

Studies were included if they evaluated critically ill children with gastrointestinal dysfunction, performance characteristics of assessment/scoring tools to screen for gastrointestinal dysfunction, and assessed outcomes related to mortality, functional status, organ-specific outcomes, or other patient-centered outcomes. Studies of adults or premature infants, animal studies, reviews/commentaries, case series with sample size ≤10, and non-English language studies with inability to determine eligibility criteria were excluded.

Data Extraction:

Data were abstracted from each eligible study into a standard data extraction form along with risk of bias assessment by a task force member.

Results:

The systematic review supports the following criteria for severe gastrointestinal dysfunction: 1a) bowel perforation, 1b) pneumatosis intestinalis, or 1c) bowel ischemia, present on plain abdominal radiograph, computed tomography (CT) scan, magnetic resonance imaging (MRI), or gross surgical inspection, OR 2) rectal sloughing of gut mucosa.

Limitations:

The validity of the consensus criteria for gastrointestinal dysfunction are limited by the quantity and quality of current evidence.

Conclusions:

Understanding the role of gastrointestinal dysfunction in the pathophysiology and outcomes of MODS is important in pediatric critical illness.

Table of Contents Summary:

We provide modern, evidence-based definitions of gastrointestinal dysfunction in critically ill children after systematic review of the literature.

INTRODUCTION

The gastrointestinal (GI) tract has long been hypothesized to act as a ‘motor’ of organ dysfunction’, but is not included in prior organ-specific criteria for pediatric multiple organ dysfunction syndrome (MODS).¹ The primary functions of the gastrointestinal (GI) tract are extensive and include transport, digestion, and absorption of food and fluids, barrier functions, immune functions, and maintenance of homeostasis between the host and gut microbes.² In pediatric critical illness, biomarkers to directly assess these primary functions of the GI tract are either not validated or not available for clinical use, nor do validated composite GI dysfunction or failure scoring systems exist as they do for adults.^3–7 Feeding intolerance may be an ideal measure of gastrointestinal dysfunction; however, there is variability in both the measures and thresholds used to define intolerance and its impact on nutrition delivery.^5,8 In addition, feeding tolerance is further complicated by heterogeneity in prescriber practices for initiation and advancement of enteral feeding.⁹ Given the high variability of definitions for feeding intolerance between studies, there are no validated criteria that can be used in a current pediatric GI dysfunction definition.

We sought to identify an objective, pragmatic definition of pediatric GI dysfunction that was well supported by existing literature and could be easily measured and implemented with existing medical and clinical infrastructure in most pediatric intensive care units (PICU) or acute care centers. We excluded criteria for GI dysfunction defined solely or predominantly in premature neonates. Based on initial review of existing literature, a definition for pediatric GI dysfunction necessitated use of signs and symptoms of GI dysfunction rather than biomarkers or specific scoring systems to directly interrogate loss of primary GI functions.

METHODS

The PODIUM collaborative sought to develop evidence-based criteria for organ dysfunction in critically ill children. The present manuscript reports on the systematic review on gastrointestinal dysfunction scoring tools performed as part of PODIUM, provides a critical evaluation of the available literature, proposes evidence-based criteria for gastrointestinal dysfunction in critically ill children, as well as recommendations for future research. The PODIUM Executive Summary details Population, Interventions, Comparators, and Outcomes (PICO) questions, search strategies, study inclusion and exclusion criteria, and processes for risk of bias assessment, data abstraction and synthesis, and for drafting and developing agreement for criteria indicating gastrointestinal dysfunction.^{refPODIUMExecutiveSummary}

RESULTS

Proposed Consensus Criteria with Rationale

Of 1722 unique citations published between 1992 and 2020, 39 studies were eligible for inclusion, as shown in the PRISMA flowchart (Fig. 1). Data tables (Supplemental Tables 1 and 2) and risk of bias assessment summaries (Supplemental Fig. 1) are detailed in Supplemental Digital Content 1 and 2. Criteria for severe gastrointestinal dysfunction in critically ill children informed by the evaluated evidence are listed in Table 1.

Figure 1.

Study flow diagram according to the Preferred Reporting Items for Systematic review and Meta-Analysis Protocols recommendations.

Table 1.

PODIUM: Criteria for gastrointestinal dysfunction in pediatric critical illness

Organ system	Criterion for organ dysfunction	Suggested thresholds	Conditions	Severity
GI	Bowel ischemia	Bowel perforation OR Pneumatosis intestinalis OR ischemia present on gross inspection (surgical) or by plain abdominal film, CT, or MRI Sloughing of gut	None	Severe

CT, computed tomography; MRI, magnetic resonance imaging

Based on initial literature review, no validated or widely available scores, biomarkers, or direct assays of the primary functions of the GI tract for transport, digestion, absorption, barrier functions, immune function or host-microbial interactions were identified. Therefore, the assembled expert panel identified surrogate signs and symptoms of primary GI dysfunction, and literature was reviewed. Evidence to support a proposed consensus definition was obtained from small cohorts of pediatric patients and administrative datasets. Consensus expert opinion informed but did not determine initial proposed GI dysfunction criteria, as did existing adult GI failure scores. The final approved consensus criteria reflect signs and symptoms of severe GI dysfunction due to GI ischemia, classified as the presence of any one of the following: 1a) bowel perforation, 1b) pneumatosis intestinalis, or 1c) bowel ischemia, present on plain abdominal radiograph, computed tomography (CT) scan, magnetic resonance imaging (MRI), or gross surgical inspection, OR 2) rectal sloughing of gut mucosa. No criteria to diagnose mild or moderate GI dysfunction had sufficient evidence to be included in the final consensus definition.

GI Ischemia

Bowel perforation, pneumatosis intestinalis, or bowel ischemia

Of surrogate signs and symptoms for primary GI dysfunction reviewed, only GI ischemia was considered to exclusively represent GI dysfunction and have sufficient supportive literature after consensus voting. Inclusion of bowel perforation, pneumatosis intestinalis, or other findings consistent with bowel ischemia by radiographic or surgical inspection was based on ten studies in heterogeneous critically ill pediatric populations (Supplemental Tables 1 and 2). While bowel perforation and pneumatosis intestinalis can occur in the absence of ischemia and reperfusion injury, both were considered clear indicators of GI dysfunction and were included in the definition. In infants outside of the neonatal age group, risk factors for pneumatosis intestinalis and/or perforation were graft vs host disease, congenital heart disease, hypotension, GI bleeding, and ischemia-reperfusion injury.¹⁰ In a retrospective chart review of 37 episodes of pneumatosis intestinalis in 32 pediatric patients where neonates were excluded, the most common inciting events were infectious colitis, acute enteric infection or toxin, bowel ischemia, and gastrointestinal dysmotility. Preceding ischemia or graft versus host disease was associated with need for surgery or mortality.¹⁰ In a separate review of 1120 surgical procedures for correction of congenital heart disease, 2.8% had GI complications associated with longer hospital length of stay.¹¹ In a review of 69 children with small bowel obstruction, intestinal strangulation was evident on radiologic imaging by thickened bowel wall or reduced bowel wall enhancement indicating bowel ischemia, these findings were associated with ICU admission and longer hospital length of stay.¹² In three studies evaluating outcomes of bowel perforation after liver transplantation, bowel perforation was associated with 50% mortality.^13–15 Several studies describing use of somatic Near-InfraRed Spectroscopy (NIRS) to identify decreased GI perfusion were reviewed.^16,17 While decreased NIRS may reflect under-perfusion of the GI tract, it can also indicate reduced renal perfusion, making it nonspecific to the GI tract, so these studies were not included.

Rectal Sloughing Of Gut Mucosa

Gut mucosal sloughing was included as criteria for bowel ischemia based on expert consensus. Bowel ischemia following prolonged cardiac arrest and/or colonic blood flow restriction resulting in sloughing is a known clinical sign of restriction of blood supply in the gut. While arguably a subjective process to measure gut sloughing, its presence is a clear marker of severe GI injury.

Candidate Variables Considered but not Included in Final Proposed Criteria

Existing literature for additional candidate variables considered as criteria to define pediatric GI organ dysfunction during critical illness were reviewed. These included: feeding intolerance, GI bleeding, ileus, partial and total bowel obstruction, biomarkers to assess intestinal barrier function, measurement of intra-abdominal hypertension (IAH) or abdominal compartment syndrome (ACS), variables for GI immune function, and variables for host-microbe interactions in the GI tract. Several GI dysfunction or failure scores existed in the adult ICU and surgical literature, but none included critically ill pediatric patients.^3,4 Only criteria for feeding intolerance, ileus or small bowel obstruction, GI bleeding, intra-abdominal hypertension (IAH) or abdominal compartment syndrome (ACS), and GI ischemia met eligibility on literature review as initial variables for consideration for consensus voting. As noted, GI ischemia criteria represent the final consensus criteria. The expert panel identified Scientific Priorities after completion of literature review to address key gaps in knowledge or promising novel targets to characterize GI dysfunction in pediatric critical illness, these are listed in the Data Supplement.

Feeding Intolerance

Several large, retrospective studies demonstrate that enteral nutrition (EN) adequacy in the first 48 hours of critical illness is associated with lower 60- or 90-day mortality in mechanically ventilated children.^18,19 However, tolerance of EN as currently defined is subjective, provider-dependent, and therefore not a good objective marker for GI dysfunction.⁵ Literature for variables of EN adequacy and tolerance such as; gastric residual volumes, emesis, diarrhea, constipation, impaired gastric emptying, impaired intestinal motility, and malabsorption were reviewed. Feeding intolerance was variably determined and there were no validated scales of clinical or biomarker values to reliably assess feeding intolerance indicative of GI dysfunction in the PICU. We excluded delivery of EN as GI dysfunction criteria as this was a marker of provider discretion, rather than true GI dysfunction. In current US enteral feeding guidelines, measurement of gastric residual volumes (GRV) is not recommended as it is not an accurate method to assess EN tolerance or risk of EN-associated complications.²⁰

Ileus or Small Bowel Obstruction

Ileus is defined as failure of passage of enteric contents through the small bowel and colon that are not mechanically obstructed. Ileus is recognized on plain abdominal radiograph or Computed Tomography (CT) by proportional dilation of loops of small and large bowel with lack of a transition point. Small bowel obstruction is recognized by proximal dilation of bowel with an identified transition point. Most patients who present to the PICU with abdominal complaints undergo a flat, plain abdominal radiograph because it is widely available, accurate, and inexpensive. We excluded plain radiographic imaging evidence and clinical characteristics of ileus and small/total bowel obstruction due to limited number and size of studies that defined the association between specific radiographic and clinical findings and patient outcomes in pediatric critical illness.^12,21 A definition of pediatric GI dysfunction which includes plain abdominal radiograph assessment of ileus and partial or total bowel obstruction has face validity and is pragmatic, but would be based on expert opinion. Therefore, ileus was not included in the final consensus pediatric GI dysfunction definition. The severe consequences of partial or total small bowel obstruction to indicate bowel ischemia remain in the definition as they have literature to support a clear association with patient outcome.¹²

GI Bleeding

Several large pediatric cohort studies and one RCT of clinically significant upper and lower GI (UGI; LGI respectively) bleeding were reviewed (Supplemental Tables 1 and 2). Risk factors for UGI bleeding were respiratory failure, coagulopathy, and Pediatric Risk of Mortality Score ≥ 10.²² Clinically significant UGI bleeding was associated with mortality. Coagulopathy may increase risk for UGI and LGI bleeding. While clinically significant bleeding would indicate co-existing GI dysfunction due to impaired intestinal absorption and barrier functions, it is potentially confounded by coagulopathy so was not included in the final consensus definition.

Intra-Abdominal Hypertension

Nine studies characterizing intra-abdominal hypertension or abdominal compartment syndrome in critically ill children were reviewed (Supplemental Tables 1 and 2). Despite an association with patient outcomes, abdominal compartment syndrome (ACS) was excluded as it was defined by the presence of additional organ dysfunction. Expert agreement was that a definition of GI dysfunction should solely reflect the GI system and not be dependent upon the presence of additional organ dysfunction. Intra-abdominal hypertension (IAH) may be due to primary GI causes or due to external pathology (sepsis, trauma, burns, fluid resuscitation, capillary leak) and is characterized on abdominal radiographic imaging by findings consistent with GI dysfunction, such as bowel wall edema, bowel distention, and is associated with feeding intolerance.²³ In a prospective cohort of 175 patients, 12.6% had IAH defined as bladder pressure ≥10 mmHg.²⁴ The presence of IAH was associated with higher mortality (40.9% vs 15.6%; P = .01) and prolonged PICU stay (19.5 [3–97] vs 8 [1–104] days, OR 1.02; 95% CI, 1.00–1.04; P = .02). IAH was an independent risk factor for mortality (OR 6.98; 95% CI, 1.75–27.86; P = .006). Using the Pediatric Guidelines Sub-Committee for the World Society of the Abdominal Compartment Syndrome recommendations, a threshold of bladder pressure of ≥10 mmHg was considered but not included in the final consensus criteria.²⁵ Additional studies are needed to identify whether a threshold of bladder pressure ≥10 mmHg is sensitive and specific to primary GI dysfunction and is association with patient outcomes. We also evaluated use of abdominal perfusion pressure as criteria for GI dysfunction. In a prospective cohort of 35 patients one month-18 years, serial measurements of bladder pressure and mean arterial pressure (MAP) were obtained to determine abdominal perfusion pressure.²⁶ By receiver operator curve (ROC) analysis, optimal cutoff point for abdominal perfusion pressure was 53 mmHg, and low perfusion pressure was associated with mortality. Additional studies of abdominal perfusion pressure as a criteria for primary GI dysfunction are needed.

Gut Host-Microbial Interactions, Maintenance of Barrier Function, and Immune Homeostasis

The gut microbiome is profoundly altered during pediatric critical illness and is characterized by decreased relative abundance of beneficial commensal bacteria and increased relative abundance of harmful pathobionts.^27,28 This dysbiosis and altered homeostasis between host and microbes is implicated in the pathophysiology of intestinal and extraintestinal organ dysfunction.^1,29 The intestinal epithelial barrier is the largest surface area in contact with the human external environment and is the critical interface between the gut microbiome and host immune system². Impaired intestinal barrier function either due to critical illness or in response to dysbiosis can result in translocation of bacteria or their metabolites, influence host immune responses, and increase exposure to pathogens.^1,30 Minimally invasive urine or plasma biomarkers of intestinal barrier functions exist, but are not validated outside of research use.⁶ Whether a multi-omics approach will yield key diagnostic indicators of gut dysbiosis, impaired intestinal immune functions, or pathologic host-microbial interactions measured in the urine, blood stream, or feces relevant to patient outcomes is yet to be determined.²⁸

CONCLUSIONS

The current literature for pediatric GI dysfunction during critical illness is limited and impacts the validity of the proposed consensus criteria. We propose consensus criteria that reflect signs and symptoms of severe GI dysfunction due to GI ischemia, classified as the presence of any one of the following: 1a) bowel perforation, 1b) pneumatosis intestinalis, or 1c) bowel ischemia, present on plain abdominal radiograph, CT scan, MRI, or gross surgical inspection, OR 2) rectal sloughing of gut mucosa. The limitations of supporting literature result in a weak recommendation, reflecting expert consensus. Given the emerging links between MODS pathophysiology and GI dysfunction, prospective evaluation of these criteria are urgently needed to identify and define the incidence and outcomes of GI dysfunction during pediatric critical illness. Further research is needed to develop and test whether criteria scaled according to the severity of GI dysfunction and incorporating validated tests or biomarkers of primary GI functions are associated with patient outcomes. Additionally, it is essential to better understand the emerging role of gut-host microbial interactions on GI dysfunction and patient outcomes of pediatric critical illness.

Supplementary Material

Data Supplement

Supplemental Figure 1. Risk of Bias Assessment Summary for Studies Included in the PODIUM Gastrointestinal Dysfunction Systematic Review (n=39 studies)

Supplemental Table 1. Studies Included in the PODIUM Gastrointestinal Dysfunction Systematic Review (n=39 studies)

Supplemental Table 2. Performance Characteristics for Assessment Tools and Scores for Gastrointestinal Dysfunction in Critically Ill Children (n=39 studies)

Abbreviations:

MODS

multiple organ dysfunction syndrome

PODIUM

Pediatric Organ Dysfunction Information Update Mandate