Preoperative fasting guidelines in the paediatric population have always been a topic of interest.¹ These guidelines were created to prevent regurgitation and aspiration during airway manipulation. Traditional fasting guidelines advocated the 6-4-2 h rule for solids, breastmilk and clear fluids, respectively.² Recently, societies have adapted to a 1-h fasting for clear fluids.^1,3 A clear fluid is defined as water, pulp-free fruit juice, squash or cordial, ready-diluted drinks, and non-fizzy sports drinks, which are non-thickened and non-carbonated.¹

The prescribed fasting duration for carbohydrate-rich fluid is less clear. The European Society of Anaesthesiology and Intensive Care (ESAIC) considers carbohydrate-rich fluid as a non-clear fluid and recommends a 2-h fasting period before induction of anaesthesia in paediatric patients.¹ Similarly, the American Society of Anesthesiologists (ASA) Practice Guideline recommends a 2-h fasting period after ingestion of carbohydrate-rich drinks containing complex carbohydrates in adult and paediatric patients.² In contrast, the Australian and New Zealand College of Anaesthetists considers both simple and complex carbohydrate-containing drinks as clear fluids and recommends a 1-h fasting period after ingestion of carbohydrate-rich drinks in paediatric patients.⁴

The consumption of oral carbohydrates preoperatively has shown several benefits, including stabilised blood glucose levels, improved metabolic responses, and a reduction in anxiety and postoperative nausea and vomiting (PONV).^2,3,5 Perioperative carbohydrate loading has been established as part of the enhanced recovery after surgery guidelines in paediatric surgery.⁶

Gastric ultrasound has emerged as a novel way of assessing gastric residual volume (GRV) preoperatively and can play a vital role in determining the risk of aspiration.^5,7 It provides an objective, real-time assessment of gastric content and volume and has been proposed as the gold standard for assessing gastric contents and has been endorsed by the ASA and ESAIC.^1,2,8 A gastric volume of less than 1.5 mL/kg is unlikely to cause aspiration risk.⁸

The effect of carbohydrate-rich fluids on gastric volume has been found to promote gastric emptying 2 h post-ingestion.^8,9 Objective measurements on gastric ultrasound using GRV or cross-sectional area (CSA) of the antrum were also reduced.^5,8,9

The effect of carbohydrate-containing fluid drinks on gastric ultrasound measurements at 1-h post-ingestion is limited.^5,10 Our study aimed to compare preoperative GRVs at baseline, 1- and 2-h intervals in elective paediatric orthopaedic patients after consumption of OCD. Secondary objectives included comparing the groups based on nutritional status, determining thirst preoperatively, nausea and vomiting postoperatively, and the effect on blood glucose levels at the onset of surgery.

This prospective single-centre observational study was conducted at Charlotte Maxeke Johannesburg Academic Hospital between 01 November 2023 and 30 April 2024.

Paediatric patients between the ages of 2 years and 18 years old who adhered to standard fasting guidelines and consumed the OCD as per the study protocol, ASA physical status I or II and undergoing elective orthopaedic procedure were included. Patients who did not adhere to fasting guidelines or study protocol, those with morbidities that delay gastric motility, emergency procedures or where informed consent was not obtained, were excluded. The first booked patient on the theatre list was also excluded to avoid theatre delays.

The primary researcher obtained written informed consent from the parent or caregiver on the day preceding surgery. Children who were 7 years and above provided additional assent. The process of the gastric examination was explained, and patients were asked to maintain fasting for 6 h for solids and 2 h for clear liquids as per the study protocol. On the morning of the surgery, the primary researcher confirmed adherence to the fasting protocol. Demographic details were collected, including weight and height. Nutritional status was determined using the Centre for Disease Control body mass index (BMI) calculator for children and teens.¹¹ An experienced consultant radiologist (H.M.) specialising in ultrasound imaging in paediatric patients performed the assessments.

Three assessments were performed at the patient’s bedside, placing the patient in the supine position and, thereafter, in the right lateral decubitus (RLD) position. The first scan was a baseline assessment of the gastric antrum (T0). The patient was then offered the standardised OCD Preop™ (Nutricia), a non-carbonated, iso-osmolar drink, containing maltodextrin and polysaccharides. It is fat and protein-free and contains 0.5 kcal/mL. The patients were given a volume of 3 mL/kg of body weight to consume.³ We used adjusted body weight to determine the required volume for patients with a nutritional status ≥ 95th percentile.¹² The second and third ultrasound assessments were performed after 1- and 2-h (T1 and T2).

A Mindray Te7 ultrasound machine, with a linear transducer probe of 6 MHz – 13 MHz frequency and abdominal presets, was used. With paediatric patients displaying larger BMI and additional subcutaneous fat, a curvilinear transducer probe was used at a lower frequency (2 MHz – 5 MHz) to obtain accurate measurements.

Qualitative and quantitative assessments of the gastric antrum were performed. For the qualitative grading, the patients were initially scanned in a supine position, then in the RLD position, using the 3-point grading system described by Perlas et al.⁷ The antrum was described as either empty or flat if the walls of the antrum were found opposing each other. Clear fluids displayed distended walls and the appearance of an anechoic or hypoechoic on ultrasound. A ‘frosted glass’ pattern was seen if solid particles were visualised, which later appeared more heterogeneous in nature and hyperechoic.⁷ A grading score was then applied: Grade 0 – no fluid visualised in either the supine or RLD positions; Grade 1 – antral fluid visualised in the RLD position only; and Grade 2 – antral fluid visualised in both the supine and RLD positions.⁷

Quantitative measurements were carried out by obtaining three images of the gastric antrum while the patient was in the RLD position. This was best displayed and timed between peristaltic contractions. This approach strongly correlated with gastric volume.⁵ Once the qualitative assessment was satisfactory, the CSA was calculated and expressed in squared centimetres (cm²). All three images were traced manually by the radiologist, and the mean values of the three values were used to calculate the gastric volume. The equation from Spencer et al. was used to calculate the gastric volume: = −7.8 + (3.5 × RLD CSA in cm²) + (0.127) × age in months.¹³ This value was then divided by each participant’s actual weight to obtain a value expressed in mL/kg. Values and assessments regarded as acceptable and low risk for aspiration risk were a qualitative grading of either 0 or 1 with a gastric residual volume of less than 1.5 mL/kg.^5,7,8,14

After the last ultrasound assessment, patients waited for surgery in the ward at the scheduled time. All measurements were completed between 6:00 and 9:00 on the morning of surgery at the convenience of the research team. Thirst was assessed outside the theatre before induction of anaesthesia. Older children were subjectively evaluated by self-reporting, and younger children were assessed through input from the parent or caregiver.

After induction, blood glucose levels were recorded for each patient, and the actual fasting time (time of ingestion of fluid to time of procedure) was noted. Where the blood glucose measured was low, the attending anaesthetist administered intravenous dextrose to correct it as per institutional practice. In the recovery room and after the patient was fully awake, nausea and vomiting were subjectively assessed by the attending anaesthetist by evaluating for signs of retching or vomiting or questioning the child with the assistance of the recovery nurse, parent or caregiver if needed.

A total of 38 participants were enrolled in this study, of which five were excluded (Figure 1). Demographics are summarised in Table 1.

All patients consumed the calculated OCD and subjectively reported satisfaction, with no reports of discomfort, nausea, or vomiting after ingestion.

Gastric residual volume means are tabulated in Table 2. There were no statistically significant differences between the mean GRVs at the three time points. Oral carbohydrate drink consumption did not affect gastric antrum volumes substantially over time using quantitative volumes. Figure 2 illustrates the changes in gastric volumes at the three time intervals.

Table 3 shows the qualitative grading with the median GRV at the three-time intervals and compares calculated gastric antrum volumes (mL/kg) at T0, T1, and T2 for pre- and post-OCD ingestion. There were no statistically significant differences among the grades at any time. While grade 2 appears to have higher median volumes at baseline and 1 h, the small sample size limits the reliability of this observation (Table 2).

When comparing gastric volume changes over the three intervals in each patient, none of the patients demonstrated a volume > 1.5 mL/kg (range 0.010 mL/kg – 0.930 mL/kg) at any time point. (Appendix 1 Figure 1-A1).

Figure 3 shows the change in qualitative grade at T0, T1 and T2. At T1, two patients converted from Grade 0 to Grade 2 and subsequently to Grade 0 and 1 at T2, respectively. Four participants presented with a Grade 2 antrum at T0; at T2, all four were safely converted to a Grade 0 antrum. Only one patient had a Grade 2 qualitative assessment at T2, but no patients exceeded a volumetric evaluation of 1.5 mL/kg.

Appendix 1 Figure 2-A1 compares GRV at T0, T1 and T2 in children underweight, average weight, or overweight for age. No significant differences were found between the groups.

For those who were underweight, gastric volume was not significantly different from T0 to T1 (t = –0.29, p = 0.776, d = 0.26), T1 to T2 (t = 1.37, p = 0.183, d = 0.26), or T0 to T2 (t = 0.83, p = 0.417, d = 0.33). For those who were average weight, gastric volume was not significantly different from T0 to T1 (t = 0.82, p = 0.498, d = 0.23), T1 to T2 (t = –0.47, p = 0.683, d = 0.20), or from T0 to T2 (t = 2.87, p = 0.103, d = 0.03). For those who were overweight, gastric volume was not significantly different from T1 to T2 (t = –1.19, p = 0.297, d = 0.12) or from T0 to T2 (t = –1.94, p = 0.125, d = 0.11). However, gastric volume was significantly increased from T0 to T1 (t = –3.69, p = 0.021, d = 0.09).

There was a significant negative correlation between time to surgery and intraoperative blood glucose (r = –0.663, p < 0.001). A longer time to surgery, post-OCD consumption, was associated with a lower blood glucose level. However, no cases of hypoglycaemia were reported during surgery (mean serum glucose 4.28; range 3.80 mmol/L – 4.80 mmol/L). The average waiting time for surgery was 5 h 11 min. Over one-third of patients experienced thirst preoperatively, and no patient experienced postoperative nausea and vomiting (Table 1).

In this prospective study, the use of bedside gastric ultrasound confirms that preoperative consumption of a conservative volume of an OCD of 3 mL/kg up to 1 h before surgery did not significantly alter GRV. Qualitative ultrasound assessment found that preoperative OCD did not significantly increase gastric content, confirming the safety of OCD ingestion. We found that GRV dropped to below baseline values at 2 h. This supports existing research suggesting that carbohydrates promote gastric emptying rather than delaying it.^5,8,10

This is consistent with previous studies by Odendaal et al., who found that consuming a 3.5 mL volume carbohydrate lollipop did not significantly change GRV after 1 h.⁵ Similarly, Zhang et al. concluded that GRV remains low 1 h after ingestion of 5 mL/kg of OCD.¹⁰

The ESAIC guideline does not clearly define a volume limit for carbohydrate drinks.¹ Volumes of OCD ingested preoperatively range between 2 mL/kg and 15 mL/kg, with the commonest volume being 5 mL/kg.¹⁵ Serial magnetic resonance imaging (MRI) scans show that GRV returns to baseline within 1 h after ingesting 3 mL/kg of sugar-containing fluid.³ For this reason, we used a conservative volume of 3 mL/kg of OCD in this study. The highest volumes of OCD ingestion were found in a study conducted by Song et al.,⁸ where 15 mL/kg was administered 2 h preoperatively for patients younger than 3 years old and 10 mL/kg for patients older than 3 years old.⁸ These volumes were well tolerated, and a reduction in CSA of the antrum was found.⁸ Garg et al.⁹ demonstrated that consuming the 8 mL/kg volume of pulp-free juice in children 2 h preoperatively was safe. However, this study was limited as serial ultrasound measurements were not performed.⁹ A recent scoping review¹⁵ concluded that with regard to safety, a volume of 10 mL/kg carbohydrate-containing fluid does not increase GRV in a range of patients, including children and infants and additionally documented that 1-h fasting duration results in a lower gastric content volume compared to a 2-h fasting duration.¹⁵

Bisnotto et al. compared glucose solutions to carbohydrate-rich solutions and demonstrated that the gastric emptying of carbohydrate drinks is slower because of the higher osmolarity and calorie content; however, the GRV is lower.¹⁶ Carbohydrate drinks were also preferred over the standard glucose solution in this study.

In children who were classified as obese, we demonstrated safe values of GRV at 2 h post-ingestion despite an increase at 1 h. GRV remained below the safe limit of 1.5 mL/kg at all time intervals. In another study that objectively measured gastric volume by aspiration of contents, no differences were found between overweight, obese and normal-weight children.¹⁷

Our study found a correlation between qualitative and quantitative GRV measurements. Patients with a higher qualitative grade (Grade 2) had significantly higher measured GRV values. However, at the 2-h interval, all previous Grade 2 patients converted to either Grade 0 or 1, with no significant differences between the grades in measured volume. This suggests that the qualitative grading system initially helped identify high-risk patients, but quantitative measurements showed no differences between the two groups. Qualitative grading via ultrasound was more straightforward, making it more practical for routine bedside use. Quantitative measurements required capturing images and calculating volumes, which was much more time-consuming. Several studies support using qualitative ultrasound assessment of GRV as a practical and more efficient method in a clinical setting.^{7,13,14,16,18}

The ESAIC guidance suggests using CSA as a surrogate for gastric volumes (Grade 2B) and prefers qualitative grading to quantitative grading (Grade 2B) as this allows for simple clinical decisions to be made.¹

Just over one-third of patients reported thirst before the procedure, while no patients reported any discomfort or PONV in our study. The beneficial effects of carbohydrate intake on preoperative thirst, hunger, stabilisation of blood glucose levels and reduction in PONV have been described in multiple studies^6,9,15,17 with no reports of hyperglycaemia.¹⁵ The positive effect of reducing thirst may not be as apparent in our study as fasting times were prolonged despite OCD consumption. The OCD was consumed early in the day, and patients remained fasted after the initial assessments. This was to allow all assessments to be performed within a set time at the convenience of the researchers.

Patients in our study who had a longer time to surgery displayed a significantly lower glucose level. Viljoen et al. examined critical events in anaesthesia in paediatric patients at a South African hospital.¹⁹ They found that hypoglycaemia, which was the most common critical event, occurred in 21% of cases.¹⁹ In our study, no child had a recorded hypoglycaemia at the time of surgery. Despite the longer durations of fasting, the OCD may have assisted in preventing this. Postoperative nausea and vomiting have been reported where volumes of 10 mL/kg – 15 mL/kg of carbohydrate fluid were used.⁸ We found no PONV as our OCD volumes were much lower.

Our study has some limitations. This was a single-centre study in a select group of elective patients with no other risk factors for aspiration. Results cannot be generalised to other paediatric patient populations. No adverse events were reported in this study, but the sample size was small. Postoperative nausea and vomiting were subjectively assessed, and other pharmacological means to prevent PONV were not recorded. This was, however, not the primary objective of the study. A radiologist conducted all ultrasound examinations to ensure data accuracy. However, all measurements were made during a fixed period, and no repeat assessment was performed prior to induction. Anaesthetists, who are novice gastric ultrasonographers, may not find precise results. Given the agreement between quantitative and qualitative measures, a novice sonographer could use quantitative assessment only.

The strengths of this study are the prospective methodology used in this study and the measurement of both quantitative and qualitative assessments by a specialist radiologist. Using participants as their own control is a crucial advantage of repeated measures designs, which helps mitigate the influence of confounding factors. Using serial ultrasound imaging at the three time intervals allowed us to predict the outcome of the OCD safely and to correlate our GRV values.

In conclusion, our study offers a prospective assessment by sonography of gastric volume and contents in elective paediatric patients following the consumption of a preoperative OCD, suggesting that consuming this drink up to 1 h before surgery does not significantly increase gastric volume in this population. Future research should quantify a safe volume of OCD in children. In addition, further research is needed to understand gastric kinetics in overweight and obese children.