Introduction

Regurgitation and aspiration of gastric contents is a rare but dangerous perioperative complication threating with lung damage and mortality [1-4]. According to the Closed Claims project of American Society of Anesthesiologists, aspiration of gastric contents accounted for 115 (5%) out of 2,496 cases. Aspiration-related mortality was 57% (n=66) [1]. Sun J. et al. [5] analyzed 166,491 cases of anesthesia. In their study, incidence of gastric content aspiration was 1: 8.325, incidence of regurgitation without aspiration — 1:8.325. Mortality associated with aspiration of gastric contents was 1:55,497. Volume, nature (liquid or dense content) and gastric acidity are considered to be important factors for the outcomes [1, 3-6]. Modern prevention strategies are based on accurate fasting periods before elective surgery [4, 6–12]. To date, patients do not eat dense food for at least 6 hours and liquid food for at least 2 hours before anesthesia in Europe. In North America, it is recommended to avoid eating fried/fatty foods or meats within at least 8 hours, while light meals such as tea and toast should not be taken within 6 hours before surgery. However, food passage is individual and depends on food nature, as well as conditions or diseases slowing down gastric emptying and predisposing to more gastric contents despite appropriate fasting intervals. Therefore, fasting does not guarantee an empty stomach and the risk of aspiration remains [12–17].

Ultrasound of the stomach has been used in medicine for a long time. However, bedside ultrasound assessment of gastric contents has become a new direction in the work of anesthesiologists. This approach quickly entered the anesthesiology residency program abroad [18–21]. Ultrasound of gastric contents (empty, liquid or dense content) is valuable for more accurate analysis of the risk of aspiration compared to adherence to fasting period [3, 14, 22].

The purpose of the study was ultrasound assessment of stomach content prior to elective and emergency surgery depending on the timing of the last meal and fluid intake.

Material and methods

An observational prospective clinical study was carried out at the St. Petersburg Elizabethan Hospital. The Ethics Committee of the Kirov Military Medical Academy approved the study (protocol No. 219 dated February 26, 2019). Inclusion criteria: informed consent, age 18-75 years, forthcoming general anesthesia with tracheal intubation and mechanical ventilation throughout elective and emergency abdominal or gynecological surgery. Exclusion criteria: postoperative scar on abdominal wall along the midline, specific comorbidities (tuberculosis, sarcoidosis), inadequate imaging of antrum of the stomach.

There were 138 patients divided into 2 groups. The first group included 64 patients operated on for emergency indications; the second group enrolled 74 patients operated on in elective manner. General characteristics of patients upon admission to operating theatre are presented in Table 1.

Table 1. General characteristics of patients

Note. Data are presented as mean and standard deviation, as well as absolute numbers and percentages; * — t-test; ** — Mann-Whitney test; BMI — body mass index.

All patients were asked about the timing of the last meal and food nature (dense, liquid). In the 2nd group, anesthesiologists advised not to take dense food since the evening before surgery, light food within 6 hours before anesthesia; clear fluids were discontinued 2 hours before anesthesia (non-structural fat-free fluids such as water, clear fruit juice, tea or coffee).

We performed ultrasound examination in operating theatre before induction of anesthesia using a Mindray M5 portable ultrasound machine (Shenzhen Mindray Bio-Medical Electronic Co. Ltd., China) and C5-3s convex probe (2.5–5 MHz). Epigastric scanning was performed in supine position with operating table tilted 30° to the right side (Fig. 1).

Fig. 1. Patient position during ultrasound examination of the stomach.

During ultrasound, we obtained a cross section of the antrum in sagittal plane between the left liver lobe (anteriorly) and the pancreas (posteriorly). Additional landmarks were abdominal aorta and mesenteric artery. Four main ultrasound signs were determined [20]. Sign of empty stomach (Fig. 2a) was flat or round antrum with collapsed adjacent anterior and posterior walls. In case of round or ovoid appearance, the image is called "bull's eye" or "target". Sign of dense food in early phase of digestion (Fig. 2b) is air profile artifact just below the mucosa along anterior wall of enlarged antrum like "annular" air artifacts covering gastric content, posterior wall of the antrum, pancreas and aorta. This is because air mixes with dense food in oral cavity during chewing and this air-saturated food bolus enters the stomach. This ultrasonic feature is called ground glass pattern. Sign of dense food in delayed phase of digestion (Fig. 2c) is enlarged antrum with mixed echogenicity following forcing out the air from food bolus after some time. Clear fluids (water, tea, apple juice, black coffee) appear anechoic or hypoechoic (Fig. 2d). As the volume increases, the antrum becomes round and enlarged. There are still gas bubbles immediately after taking the liquid, which are visualized as small dotted hyperechoic inclusions disappearing within a few minutes after ingestion (sign of a starry night).

Fig. 2. Ultrasound pattern and scheme of the main ultrasonic signs.

a — ultrasound sign of an empty stomach; b — ultrasound sign of dense food in early phase of digestion; c — ultrasonic sign of dense food in late phase of digestion; d — ultrasonic sign of transparent liquid; A — antrum; Ao — aorta; D — diaphragm; L — liver; P — pancreas; R — rectus abdominis muscle; Sma — superior mesenteric artery (adapted from Perlas A. et al. (2018) [23]).

We excluded 8 patients from the 1^st group and 4 patients from the 2nd group due to impossible imaging of the antrum. Study design is shown in Fig. 3.

Fig. 3. Study design.

Statistical analysis

Data analysis was carried out using the SPSS-26 software for Windows (SPSS Inc., USA). Distribution normality was tested using visual (histograms, plots) and analytical methods (Kolmogorov-Smirnov/Shapiro-Wilk test). Normally distributed data are presented as mean and standard deviation. In case of abnormal distribution of quantitative data, we used median, 25^th and 75^th percentiles. An analysis was carried out to describe and generalize distribution of variables. Frequencies were described as n (%); 95% confidence interval (CI) was calculated using the UCALC calculator. Between-group comparison of parametric data was performed using t-test, non-parametric data — Mann-Whitney U-test. Differences were considered significant at p-value <0.05.

Results

Survey of patients in both groups showed that fluid intake was the last before surgery in most cases. Moreover, timing of fluid intake significantly differed in emergency group (p=0.025), while no differences were observed in elective group (p=0.063) (Table 2).

Comparison of timing of liquid or dense food intake between elective and emergency groups revealed no significant differences (p=0.632 and p=0.487, respectively).

The study included 138 patients. However, ultrasound failed to visualize antrum of the stomach in some cases due to bowel pneumatization and abdominal compartment syndrome in emergency patients. It is likely that higher incidence of difficulties in visualizing the antrum of the stomach is associated with abdominal compartment syndrome in emergency patients. We excluded 8 (12.5%) people from emergency group and 4 (5.4%) patients from elective group (p=0.025). Thus, difficult visualization was more common in emergency patients.

Ultrasonic sign of empty stomach was the main sign in both groups. Moreover, this sign was more common in elective group (84.3% vs. 69%) (Table 2). However, not all elective patients had an empty stomach. Ultrasonic sign of delayed phase of digestion was found in 14.3% of cases; early phase of digestion was detected in 1 patient. Ultrasonic signs of early and delayed phases of digestion were more common in emergency patients without significant between-group differences. Clear fluid in the stomach was detected only in 3 emergency patients (Table 3).

Table 2. Timing of the last meal and liquid intake prior to surgery

Note. Data are presented as median, 25^th and 75^th percentiles; * — within-group comparison of timing of dense food and fluid intake.

Table 3. Stomach content in elective and emergency patients

Note. Data are presented as n (% (95% CI); * — Mann-Whitney test.

Discussion

The problem of a "full" stomach in patients eligible for emergency and elective surgery is still important. To date, anesthesiologists focus only on survey and timing of the last meal when determining the risk of regurgitation of gastric contents [4, 8-12]. Our study revealed that patients undergoing elective surgery have longer mean fasting time than prescribed one. Timing of the last fluid and dense food intake was 8 (6.75; 10) and 9 (8; 12) hours respectively despite permission to take clear liquid 2 hours before surgery. Other studies identified similar data. Patients limited themselves in dense food and liquid intake [23]. Cestonaro T. et al. [24] analyzed 135 patients aged 19 — 89 who underwent elective surgery. Timing of the last dense food intake was 16.5 (5.5-56.9) hours, clear liquid — 15.8 (2.5-56.9) hours. Al Maqbali M.A. et al. [25] analyzed fasting before elective surgery in 169 patients. Timing of the last dense food intake was 12.1 ± 2.3 hours, liquid — 11.9 ± 2.2 hours. The authors emphasized that all patients fasted and did not take liquid for longer than the recommended time. Francisco S.C. et al. [26] also noted longer fasting time than the recommended one (16 (9.5-41.6) hours). Patients who underwent surgery in the afternoon had even longer fasting period compared to those who underwent surgery in the morning. Witt L. et al. [27] reported significantly longer duration of actual preoperative fasting for clear liquid and dense food than compared to the recommended time. Mean preoperative fasting period for clear liquid was 11.3 (6.8-14.3) hours, for dense food — 14.5 (12.1–17.2) hours. In our study, timing of food and clear liquid intake was similar in elective and emergency patients.

Ultrasound examination of the stomach ensures more objective information compared to survey, although this approach is not possible in all cases. We excluded 4 (5.4%) patients from elective group and 8 (12.5%) patients from emergency group due to inadequate imaging of the antrum (p = 0.036). Thus, difficult visualization of the stomach was more common in emergency patients. Delamarre L. et al. [28] performed ultrasound examination of the stomach in 149 (76%) out of 196 cases, Cozza V. et al. [29] — in 66% of cases. Bouvet L. et al. [30] identified gastric content in half-sitting position in 98% of patients. Van de Putte P. et al. [31] reported successful visualization of the antrum in 95% and 90% of subjects in the right lateral and supine position, respectively. Perlas A. et al. [23] reported successful ultrasound imaging of the stomach in the right lateral position in all patients. Khalil A.M. et al. [32] obtained similar data regarding imaging of the antrum in the right lateral position.

Despite the long periods after the last intake of dense food and clear liquid, not all patients had an empty stomach. Empty stomach was more common in elective patients in our study although there were no significant between-group differences in timing of the last meal (69% (57.2-80.8) vs. 84.3% (75.6-92.4), p=0.016). Kaydu A. et al. [33] found dense gastric content in 65.0% of elective patients, clear liquid in 20.0% and empty stomach only in 15.0% of patients. Cozza V. et al. [29] performed ultrasound of the stomach in supine position and found empty stomach in 43% of patients, clear liquid in 11%, early phase of digestion in 3%, delayed phase of digestion in 2% of patients. Imaging of gastric content was unsuccessful in 41% of patients. Shorbagy M.S. et al. [34] observed an empty stomach in 10 (22.2%) patients, while other 35 (77.7%) patients had a full stomach. Of these, 29 patients had dense gastric content and 6 ones had clear liquid. Khalil A.M. et al. [32] found empty stomach in 52% of patients with normal body weight and 58% of obese patients.

Delamarre L. et al. [28] reported inaccurate clinical diagnosis established by anesthesiologists in 58% of patients considered with a full stomach and 21% of patients considered with an empty stomach. Ultrasound revealed a full stomach in 53 (27%) out of 196 cases, and 36 (68%) of these ones had dense gastric content [28].

Limitations

Ultrasound examination of the stomach is associated with certain limitations. Imaging of the antrum may be inconclusive or inaccessible even with expert equipment and experienced specialists. This may be caused by anatomical changes, misinterpretation of other hollow internal structures or presence of air in nearby structures, such as bowel, that complicates imaging of the antrum. In addition, sonographic findings may be inaccurate or unreliable in patients with previous gastric surgery and large hiatal hernia.

Conclusion

Patients undergoing elective surgery have longer mean fasting time compared to prescribed time. Timing of the last fluid and dense food intake was 8 (6.75; 10) and 9 (8; 12) hours respectively despite permission to take clear liquid 2 hours before surgery. Difficult imaging of the antrum was more common in emergency patients (8 (12.5%) vs. 4 (5.4%) ones, p=0.036). Despite the long periods after the last dense food and clear liquid intake, not all patients had an empty stomach. The last one was more common in elective patients (84.3% (75.6-92.4) vs. 69% (57.2-80.8), p=0.016).

Author contribution:

Concept and design of the study — Lakhin R.E., Gettuev A.T.

Collection and analysis of data — Gettuev A.T., Arsentiev L.V., Mikhailyuk Yu.Yu.

Statistical analysis — Arsentiev L.V., Andreenko A.A.

Writing the text — Lakhin R.E., Gettuev A.T., Andreenko A.A.

Editing — Levshankov A.I., Strukov E.Yu.

The authors declare no conflicts of interest.