Gastric Volume and Its Relationship to Underlying Pathology or Acid-suppressing Medication.

BACKGROUND: Pulmonary aspiration during sedation is a major concern for sedation providers, making identifying high-risk patients a priority. Gastric fluid volume (GFV), an accepted risk factor for aspiration, has not been well characterized in fasting children. We hypothesized that GFV would increase with gastrointestinal (GI) pathology and decrease with regular acid-suppressor use. AIMS: The primary objective was to determine baseline GFV in fasting children. The secondary objectives were to evaluate the effect of GI pathology and regular use of acid-suppressing medications on GFV. SETTINGS AND STUDY
DESIGN: This was prospective, observational study.
MATERIALS AND METHODS: We endoscopically aspirated and measured GFV of 212 children fasting for >6 h who were sedated for esophagogastroduodenoscopy (EGD). Inclusion criteria were children up to 21 years of age, with the American Society of Anesthesiologists physical Status I and II presenting for elective EGD. After determining baseline GFV, the effect of GI pathology and effect of regular acid-suppressing medication use on GFV was analyzed. STATISTICAL ANALYSIS: Analysis of variance was used to compare the GFV among ages and pathology and medication groups. Student's t-test was used to compare GFV between genders and also to compare GFV in confounder analyses.
RESULTS: For the studied 212 children, average GFV was 0.469 ± 0.448 mL/kg (0-2.663 mL/kg). We found no association between GI pathology and GFV (P = 0.147), or acid-suppressor use and GFV (P = 0.360).
CONCLUSIONS: Average GFV in this study falls within the range of prior EGD-measured GFV in fasting children. Contrary to our hypothesis, we found no association between pathologies or regular acid-suppressor use on GFV. On the basis of GFV, children with GI disorders or those using acid-suppressors do not appear to pose an increased risk of aspiration. Future studies should discern differences in effects on GFV of immediate preprocedural versus the regular use of acid-suppressing medications.

INTRODUCTION

Children undergoing sedation and general anesthesia are at risk of pulmonary aspiration of gastric contents. As there is a growing trend of providing deep sedation to children outside of the operating room for various procedures, identifying children who are at higher risk for this serious complication is pertinent to anesthesiologists, pediatric critical care (PCC) and emergency physicians, as well as other sedation-trained providers.

Gastric fluid volume (GFV) has been accepted as an independent risk factor for aspiration during sedation and anesthesia.[1] In this study, we used GFV as a surrogate marker for aspiration risk.[2] Although GFV has been extensively studied in fasting children,[345678910111213141516] most studies measure GFV by suctioning gastric contents using nasogastric tubes (blind aspiration) rather than direct visualization through an endoscope. Since blind aspiration may underestimate GFV, suction during esophagogastroduodenoscopy (EGD) may yield more accurate results.

As children with gastrointestinal (GI) pathology undergo sedation for various reasons, including EGD,[7] it is important to know whether their risk of aspiration differs from children without GI pathology. Indeed, studying GFV in children with GI disorders was identified as a need by “Modern fasting guidelines in children” to determine safe fasting guidelines for affected children.[2] Studies of pediatric subjects over 5 months of age with GI disorders have not characterized GFV based on GI pathology using biopsy results, but rather on presenting symptoms and ultrasound diagnosis.[5789] In this study, we evaluated whether GI pathology affects GFV. In addition, the use of proton-pump inhibitors (PPIs) and histamine H2-receptor antagonists (H2RA) aim to decrease gastric acidity, which is an important risk factor for pulmonary aspiration.[17] Therefore, in this study, we also compared the GFV of subjects currently using acid-suppressing medication to those who are not. Although the effect on GFV of these two classes of acid-suppressors has been studied in children, the focus has been the effect on GFV of preprocedural doses rather than the ongoing use of those medications.[151819202122] However, patients with GI pathology often use these medications regularly. We, therefore, chose to assess whether ongoing use of PPIs and/or H2RAs affects GFV.

In this prospective, observational study of pediatric subjects undergoing sedation for EGD, the primary objective was to determine the baseline value of GFV in children fasting for at least 6 h. The secondary objective was to evaluate whether GFV is affected by GI pathology or regular use of acid-suppressing medication. We hypothesized that GFV would increase with GI pathology and decrease with chronic acid-suppressor use.

MATERIALS AND METHODS

This is a prospective, observational study of pediatric patients undergoing deep sedation provided by PCC physicians for EGD procedures. When subjects arrived for endoscopy, a PCC physician performed a medical history, physical examination and clinical assessment, and obtained study consent. If signs of dehydration were present, intravenous (IV) fluids would be administered until normalized vital signs were obtained before sedation. No subjects were found to have clinical signs of dehydration, and thus, no inventions were performed. An IV line was established. Immediately before the procedure, a “time out” was called to confirm the child's name, age, allergies, and signed consents for EGD and study participation. Informed consents for the procedure and study participation were obtained from the parent or legal guardian of each subject. Assent was also received from children older than 7 years of age. The institutional review board approved this study.

The study design included children aged 6 months to 21 years, with the American Society of Anesthesiologists (ASA) physical status I and II, who presented for elective EGD with or without colonoscopy. Exclusion criteria included ASA status >III, age <6 months, active respiratory infection, gastric outlet obstruction or gastric bleeding, and all other subjects thought to be unfit for procedural sedation. Subjects were enrolled between June 2013 and February 2015. Subjects were instructed not to intake any liquids or solids by mouth for 6 h before the scheduled time of the procedure, regardless of what time of day the procedure was scheduled. Although some subjects did have a longer fasting time, the instruction given to all subjects was the same.

After deep sedation initiation with one or a combination of propofol ± (midazolam, fentanyl, and glycopyrrolate), a pediatric gastroenterologist suctioned gastric fluid under direct endoscopic vision. For the residual gastric fluid (GFV) to be measured separately from other fluid suctioned during the procedure, the container used to collect residual fluid was removed and replaced when the collection was completed.

Statistical analysis

A power calculation was performed to determine the number of subjects needed for the primary and secondary objectives of the study; 198 subjects were needed to achieve a power of at least 80%. To cover for any potential issues with data analysis, 212 subjects (5% above the target of 198) were recruited by the study team.

We documented relevant demographics (age, weight, and gender), history of acid-suppressor (PPI and/or H2RA) use, biopsy results, sedation medications used, and sedation complications. Subjects were then sorted according to pathology results into six nonmutually exclusive groups: normal, inflammation (esophagitis/gastritis/duodenitis), gastroesophageal reflux, eosinophilic esophagitism, celiac disease, and others. “Others” included inflammatory bowel disease, reactive/chemical gastropathy, Crohn's disease, foreign body removal, ulcer, and parasitic infection. The groups were created based on the number of subjects with each pathology. Accordingly, the pathologies classified as “others” were the rarest in our sample population. Analysis of variance was used to compare the GFV among ages and pathology and medication groups. Student's t-test was used to compare GFV between genders and to compare GFV in confounder analyses. The analysis was run with IBM SPSS 22.0 (IBM Corporation, Armonk, New York, United States).

RESULTS

Of 214 patients approached, 212 were successfully recruited and provided consent. GFV and the effect on GFV of acid-suppressing medications were analyzed for those subjects. Because pathology could not be determined from one subject's biopsy due to sample mishandling, the effect of GI pathology on GFV was analyzed for 211 subjects.

The mean age was 11.9 years (range 2–21 years). Neither gender nor age had a significant effect on GFV [Figure 1]. In this study, we found the overall mean gastric volume to be 0.469 ± 0.448 mL/kg (range 0–2.663 mL/kg). Neither GI pathology nor pH-altering medication use was found to have an effect on the GFV [P = 0.147, Figure 2; P = 0.360, Figure 3, respectively].

Figure 1

Demographics on gastric fluid volume. Neither age (a, P = 0.188) nor gender (b, P = 0.905) significantly affected gastric fluid volume. Error bars indicate 95% confidence interval

Figure 2

Pathology on gastric fluid volume. No association found between gastrointestinal pathology groups and value of gastric fluid volume (P = 0.147, df = 5). Total n = 211. Error bars indicate 95% confidence interval

Figure 3

Medication on gastric fluid volume. No association found between acid-suppressing medication and gastric fluid volume (P = 0.360, df = 2). Total n = 212, and average gastric fluid volume = 0.469 ml/kg. Error bars indicate 95% confidence interval. PPI: Proton pump inhibitor, H2RA: Histamine H2-receptor antagonist

DISCUSSION

Sedation is commonly used for EGD and other procedures. Fortunately, the prevalence of adverse events is low in procedural sedation for EGD in children, and that of aspiration is even lower, recently reported to be <0.1%.[23] However, it remains a feared complication that carries high morbidity.[224] Sedation providers need to know what factors increase the risk of adverse events, including aspiration. Biber et al. recently identified four predictors of adverse events in pediatric sedation for elective EGD ± colonoscopy, using the extensive pediatric sedation research consortium database.[23] Here, we examined the impact of two potential risk factors not previously addressed, in the same demographic of children presenting for elective EGD ± colonoscopy. We were specifically interested in aspiration risk. To measure the effect of a given variable on the already low incidence of aspiration, a prohibitively large sample size would be needed. Therefore, surrogate markers of aspiration, such as GFV are used.[26] GFV is an independent predictor of aspiration.[1] The effect of GI pathology or regular use of acid-suppressing medications on GFV remains unclear.

In this study, we attempted to accurately measure GFV in a group of fasting pediatric subjects undergoing EGD. The average GFV found in our study population (0.469 ± 0.448 mL/kg) was consistent with previous studies that also measured GFV using direct visualization, despite different anesthetic regimens. Cook-Sather et al. found an average GFV of 0.40 ± 0.45 mL/kg,[9] and other reports range from 0.35 to 0.68 mL/kg.[578] GFV in children has been shown not to correlate with body mass index (BMI) when corrected for ideal body weight;[25] thus, our analysis did not include BMI.

The secondary objective was to evaluate the relationship between GFV and (1) GI pathology, or (2) regular use of acid-suppressing medication. Understanding the effect of those factors on GFV will help sedation providers assess aspiration risk. Contrary to our expectations, we found that neither of those two factors were associated with a difference in GFV.

To date, no study has examined GFV in children based on pathology results; only two studies have carried out so in adults, with conflicting results.[2627] In children over 5 months of age, whether presenting symptom correlates with GFV is contested.[78] The study, measuring GFV by EGD, suggests that pathology may not affect GFV in children.

The effect of regular use of oral PPIs and/or H2RAs on gastric content has not been as widely studied as that of immediate preprocedural administration of these medications. The immediate preprocedural administration is uniformly associated with decreased gastric acidity as expected. However, its effect on GFV is contested in both adults and children.[15181920212228293031] Because we assessed regular acid-suppressor use, which is more clinically relevant to patients with GI pathologies, we cannot directly compare our results to those above. The question of whether there is a difference in acid-suppressor effect on GFV between the two timelines has not been directly studied. However, PPIs have an increased effect on proton pump inhibition over the course of 3 days and H2RAs induce drug tolerance.[32] This suggests that the effects of acid-suppressors on gastric content change with duration of use. If these pH-altering medications do in fact exert an effect on GFV, it is in the context of the day of dosage. Although regular PPI use has been shown to decrease gastric secretion in adults with gastroesophageal reflux disease,[33] our results suggest that chronic acid-suppressor use was not associated with a difference in GFV in children with or without a range of GI pathologies. Future studies should be done to discern differences in effects of immediate preprocedural versus the regular ongoing use of acid-suppressors on GFV and acidity.

A possible confounder is acid-suppressing medications artificially decreasing GFV. This is unlikely because subjects with no GI pathology (“normal”) and those with pathology had the same GFV regardless of medication use (P = 0.926 and 0.986, respectively). Second, colonoscopy preparations altering GFV due to dehydration and thereby confounding our results was unlikely. Subjects with only EGD performed, and those with EGD plus colonoscopy had the same GFV (P = 0.360). Furthermore, no subjects were found to have clinical signs of dehydration before sedation. In addition, GFV has not been associated with the subjective perception of thirst, as shorter fasting times reportedly decreased thirst but did not change GFV.[3]

The present study exhibited several shortcomings. It was performed at one institution. Neither subjects’ compliance with acid-suppressors nor the last dose administered were recorded. We did not measure gastric fluid pH. However, performing no laboratory tests on the gastric fluid helped us achieve a participation rate of 99.1%. We had a lack of healthy controls. Although we included a “normal” pathology group, these subjects presented with GI symptoms warranting an EGD. There may be a difference between GFV of subjects with normal pathology and those who are symptom-free. This difference is difficult to assess clinically and would require subjecting healthy children to an EGD procedure. We must also declare that our stratification of the study subjects into six pathology groups was arbitrary, as there is no guidance in the literature on how to meaningfully stratify this wide range of pathologies. The “other” group, which included multiple pathologies, was created because of our inability to analyze the data due to the small sample size of subjects with those pathologies. We did not remove that group from our analysis, to provide full disclosure of our subjects’ data. Finally, we did not evaluate the effect of anesthesia regimen on GFV, as it was beyond the scope of this study. All subjects received IV propofol, in keeping with institution standard. Therefore, our average GFV may be affected by the use of propofol,[34] but our comparisons among pathologies and medications should not be affected.

CONCLUSIONS

In this study, we show that amount of GFV in fasting children is 0.469 mL/kg, and is not significantly different with GI pathology or chronic acid-suppressor use. On the basis of GFV alone, children with GI pathology or those who take acid-suppressors do not appear to inherently pose an increased risk of aspiration when sedated after 6 h of fasting.

Financial support and sponsorship

Internal Departmental funding. Department of Child Health, University of Missouri, Columbia, MO, USA.

Conflicts of interest

There are no conflicts of interest.