Pulmonary aspiration of gastric contents is one of the most feared complications of obstetric anaesthesia.1 Parturients are at particular risk of this complication because of increased intra-gastric pressure related to the gravid uterus favouring gastro-oesophageal regurgitations,2 increased risk of difficult intubation in case of general anaesthesia3 and an increased volume of gastric contents mainly related to delayed gastric emptying during labour.4–6 Hence, the quantification of the gastric volume and contents may be of particular interest to obstetric anaesthetists when faced with a need for emergency anaesthesia in a labouring woman.
Ultrasonographic assessment of gastric contents is a non-invasive tool that has been successfully used in healthy volunteers and in surgical patients, by using both qualitative examination of the antrum and measurement of the cross-sectional antral area.7–11 Perlas et al.10 defined three qualitative grades of the ultrasound scan: grade 0 (an empty stomach), defined as an empty antrum in both the supine and the right lateral decubitus position; grade 1, visualisation of gastric fluid in the right lateral decubitus position only and grade 2, visualisation of gastric fluid in patients lying in the supine position. The qualitative ultrasound assessment of the antrum allows an easy and fast diagnosis of a full stomach whenever solid contents are visualised, or when a significant volume of liquid contents are present in the stomach in the supine position (grade 2). If no gastric contents are observed in the supine position, the patient must be turned into the right lateral decubitus position and be rescanned to either confirm that no gastric contents are present (grade 0) or that gastric fluid, not previously seen in the supine position is now visible (grade 1).10 Previous studies reported that grade 1 and grade 0 corresponded to significantly different volumes of gastric fluid in adult patients and in term pregnant women.12,13
A complementary approach, based on the measurement of the antral cross-sectional area, has also been described.7,8 Ultrasound examination of the antrum allows a standardised and easy-to-perform measurement of the observed antral area, with good intra and inter-rater reliability.14 In the non-pregnant adult, the antral area measured in the supine position correlates positively with the gastric volume. A cut-off value of 340 mm2 for detecting a ‘risk stomach’, defined by solid contents and/or a liquid volume more than 0.8 ml kg−1, has a high performance.7 Thus, alternatively, a patient at low risk of aspiration could be defined by a stomach with an antral area less than 340 mm2 and with no visualisation of gastric contents. Whether such a cut-off value of the antral area measured in the supine position exists for women in labour is unknown.
Therefore, we conducted this prospective study of women in labour to calculate the cut-off value of the antral cross-sectional area measured in the supine position for the diagnosis of an ‘empty’ stomach (grade 0).
Materials and methods
This prospective observational cohort study received approval (n 2015–028B) from the local ethics committee (Comité pour la Protection des Personnes Sud Est III, Groupement Hospitalier Est, Lyon, Chairperson Professor F. Chapuis, on 9 June 2015) and was registered in the French database for clinical trials (EudraCT number: 2015-A00682-47). The methodology followed the recommendations of the STrengthening the Reporting of OBservational studies in Epidemiology (STROBE) statement.15 The study was conducted between June and July 2015 in the Femme Mère Enfant Hospital, Lyon, France. Patients were recruited according to the availability of someone to perform the ultrasound scans: two physicians (L.J. and L.B.), with experience of at least 50 gastric ultrasound examinations in pregnant women. Written informed consent from the women was obtained on arrival in the delivery room.
Pregnant women at more than 36 weeks of gestation were recruited to take part in the study if they met the following clinical inclusion criteria: in established labour, at least 18 years of age, American Society of Anesthesiologists physical status I to III, weight 45 to 110 kg, and height more than 150 cm. Exclusion criteria were age less than 18 years, refusal to participate in the study, contraindication for the right lateral decubitus position during the labour and a history of upper gastrointestinal surgery. The women were not allowed any oral intake during labour.
A portable ultrasound device (M-Turbo, SonoSite, Inc., Bothell, Washington, USA) fitted with an abdominal probe (C60 2 to 5 MHz) was used for the ultrasound examinations.
Ultrasound scans were performed within the first 2 h of the women arriving in the delivery room. The women lay in a supine, 45° head-up, ramped position for the first scan and then in the full right lateral decubitus position for the second. To obtain a consistent, standardised scanning level in the sagittal plane, the abdominal aorta and the left lobe of liver were used as internal landmarks, as described in the literature.7,16,17
An ultrasound assessment of the antrum was performed by the first operator and the stomach contents scored using a modification of the qualitative grading scale previously described by Perlas et al.10 Grade 0 was defined as the lack of visualisation of any gastric contents in the antrum, in both the supine and the right lateral decubitus positions. Grade 1 was defined as the visualisation of gastric fluid in the right lateral decubitus position but not in the supine position, and grade 2 was defined as the visualisation of fluid content in both the supine and right lateral decubitus positions. For the present study, we defined grade 3 as the visualisation of solid or of solid + liquid contents in the antrum, whatever the position (Fig. 1).
Immediately after this first examination, a second operator, blinded to the results of the first assessment, performed an ultrasound measurement of the antral area, in both the supine and the right lateral decubitus position. The maximal anteroposterior diameter (D1) and longitudinal diameter (D2) of a single section of the gastric antrum were determined and the antral area was calculated using the formula, antral area = (π × D1 × D2)/4.18 The mean value of three consecutive measurements of each diameter was used to estimate the antral area for this study.
Any failure for the qualitative and/or the quantitative assessment of the antral area was noted and the patients were excluded from the analysis.
Patient characteristics (age, height, current weight, parity and gestational age), and medical history (gestational diabetes, gastro-oesophageal reflux and neurological disease) were recorded. The presence of epidural analgesia during the ultrasound examination, and the duration of liquid and solid fasting, was also noted.
Statistical analysis was performed using MedCalc for Windows, version 18.104.22.168 (MedCalc Software, Ostend, Belgium).
After a Shapiro–Wilk's W test for normality of distribution of the data, data were expressed as mean ± SD. One-way analysis of variance followed by post hoc test (Bonferroni adjusted comparisons) whenever a statistically significant difference was observed, or t-tests, were used as appropriate to perform the comparisons of normally distributed quantitative data. Incidence data were analysed using χ2 or Fisher exact tests.
The performance of the measurement of the antral area in the supine position to detect a gastric antrum grade 0 was assessed using receiver operating characteristics (ROCs) curve, constructed by plotting the sensitivity as a function of the false positive rate (100-specificity). The area under the ROC curve was calculated. The antral area cut-off value was determined for sensitivity more than 80% with the highest corresponding specificity for detecting an antral grade 0.
For each test, P < 0.05 was considered as statistically significant.
Assuming that the area under the ROC curve for the detection of a gastric antrum grade 0 was 0.80, required 32 women to be included in each group (antral grade = 0 or antral grade >0) to create the ROC curve with a significance level of 0.05 and a power of 0.85. Because of foreseeable protocol violations and difficult ultrasound examinations, we planned to recruit 90 patients.
Ninety patients met the inclusion criteria. Figure 2 shows the recruitment and loss of patients in the trial. Four (4.4%) were excluded because of protocol violation and 13 (14.4%) were excluded because the antrum was not clearly identified during the first ultrasound assessment, thus preventing the determination of antral grading (Fig. 2): the antrum was not visualised in the right lateral position in four patients, in the supine position in four patients and in neither the right lateral decubitus nor the supine position in five patients.
Seventy-three parturients were included and analysed (Table 1). All but two had epidural analgesia. Two parturients had gestational diabetes but none suffered from neurological disease nor from gastro-oesophageal reflux. The ultrasound examination findings are reported in Table 2. No parturient had haemodynamic issues while lying in the right lateral decubitus. The frequencies of the antral area measured in the supine position in the whole population studied, and in parturients with antral grade = 0, 1 and more than 1, are set out in Fig. 3.
The ROC curve for the diagnosis of antral grade = 0 is set out in Fig. 4. The cut-off value of the antral area measured in the supine position to detect a grade 0, was 381 mm2, with a sensitivity [95% confidence interval (CI)] of 81 (64 to 92) %, a specificity (95% CI) of 76 (59 to 88) %, a negative predictive value (95% CI) of 80 (63 to 92) % and a positive predictive value (95% CI) of 76 (60 to 89) %.
Thirty-four parturients (46.6%) had an antral area less than 381 mm2. The distribution of the parturients with an antral area less than or more than 381 mm2 according to the antral grade is summarised in Table 3. The fasting duration was not recorded in four patients. The mean fasting duration for solids and non-clear liquids was significantly longer in patients with antral grade 0 vs. other grades, and in patients with an antral area less than 381 mm2 compared with patients with an antral area more than 381 mm2 (Table 4). Twenty-three patients out of 34 (67%) with an antral area less than 381 mm2 had a fasting duration for solids more than 8 h, whereas 25 patients out of 35 (71%) with an antral area more than 381 mm2 had a fasting duration for solids less than 8 h (P = 0.002). However, no statistical significant relationship was found between clear liquid fasting and antral status (Table 4).
This study calculated a threshold antral area measured in the supine position for the detection of an antral grade 0 in labouring women.
The qualitative antral grading score previously described by Perlas et al.10 in non-pregnant adults may be useful to predict the amount of gastric fluid in preoperative period. In fact, it was reported in adult surgical patients that a grade 0 corresponded to a gastric fluid volume of 0 ml, whereas a grade 1 was associated with fluid volume less than 100 ml in 77% of the study participants, and a grade 2 was associated with fluid volume more than 100 ml in 75% of the patients.12 More recently, Arzola et al.13 reported that the predicted fluid volume ranged from 0.4 ml kg−1 (grade 0) to 1 ml kg−1 (grade 1) and 2.7 ml kg−1 (grade 2) in term non-labouring pregnant women. These values were calculated using a mathematical model validated in non-pregnant adults only, and should therefore be considered with caution. Nevertheless, they suggest that a grade 0 may correspond to a low gastric fluid volume, less than 0.8 ml kg−1, and may consequently correspond to a low risk of pulmonary aspiration of gastric content, whereas a grade 2 may be associated with a significant gastric fluid volume and increased risk of aspiration. However, the interpretation of a grade 1 remains difficult. A gastric volume of 50 ml (or 0.8 ml kg−1) may be considered as a critical volume for severe aspiration in humans.5,19 The combination of a gastric volume more than 0.8 ml kg−1 with other risk factors, such as difficulties in airway management or inappropriate anaesthetic technique forcing air into the stomach, may be sufficient to cause significant regurgitation and aspiration with subsequent pulmonary damage.20,21 In the absence of more accurate data concerning the relationship between the gastric volume and the risk of aspiration of gastric content in anaesthesia, one cannot exclude that a qualitative grade 1 score may be associated with a significant risk for aspiration of gastric contents: 23% of grade 1 patients may have a gastric fluid volume more than 100 ml, that is, more than 1.5 ml kg−1.12 Therefore, in our study, an empty stomach was defined as antral grade = 0.
The discrimination between antral grade 0 and grade 1 requires two ultrasound examinations, turning the patient into the right lateral decubitus position between the two scans. This latter manoeuvre can be performed easily for most elective surgical patients. However, in clinical practice, the assessment of gastric contents in the labouring woman may be needed in association with an obstetric emergency, for example, in the context of acute foetal compromise or of maternal haemodynamic alteration and these situations may contraindicate turning the parturient into the right lateral position. Therefore, for the assessment of gastric contents, it is particularly appropriate to describe an ultrasound tool for use in the supine position. This will allow either the fast diagnosis of a full stomach (grade 2 or 3, clear visualisation of gastric contents in the supine position), or the fast diagnosis of an ‘empty’ stomach when the antral area is less than 381 mm2 and no gastric contents are observed in the supine position. Several studies have reported that, despite the potential risk of aspiration, tracheal intubation was not always performed routinely during obstetric general anaesthesia (during or after vaginal delivery).22,23 The ultrasound measurement of the antral area may constitute a new point-of-care examination to enable a patient's airway management to be individualised, based on evidence of gastric contents.
The area under the ROC curve was 85%, corresponding to a good performance of the ultrasonographic measurement of the antral area for the diagnosis of an empty stomach. The cut-off value of the antral area was chosen for sensitivity more than 80% with the highest corresponding specificity. High sensitivity is of particular interest to minimise the risk of concluding falsely that the stomach is empty. In clinical practice, for the fast diagnosis of an empty stomach, the measurement of the antral area would only be performed as a supplement to the qualitative analysis when no gastric contents were observed in the supine position (grade 0 or 1). The risk of concluding that the stomach was empty while there were significant gastric contents (grade 2 or 3) is therefore less important than suggested by the 81% sensitivity. Therefore, in clinical practice, when the right lateral decubitus position is contraindicated, or time is limited, we suggest that whenever gastric contents are not visualised in the supine position (grade 0 or grade 1) an antral area cut-off value of 381 mm2 may be used to define an empty stomach.
The calculated cut-off value of the antral area measured in the supine position (381 mm2) is higher than the value previously reported in the non-obstetric population (340 mm2).7 This may be because of changes in the anatomical position of the stomach: it is displaced in a cephalad direction by the gravid uterus in obstetric patients. Bataille et al.24 previously reported that the cut-off value of the antral area was 320 mm2 in parturients. However, these authors did not use an appropriate methodology for the calculation of the cut-off value. In the present study, 54% of patients with a prolonged fasting time (last solid meal >8 h) had an antral area more than 320 mm2, whereas the antral area was more than 320 mm2 for 80% of patients with a shorter fasting time (last solid meal <8 h). Similar to the results published by Arzola et al.13 in term pregnant women, the antral area measured in the right lateral decubitus was less than 960 mm2 for 92% of patients with a prolonged fasting time, but the antral area was more than 960 mm2 for only 39% of patients with a shorter fasting time. The cut-off value of the antral area measured in the right lateral decubitus position was not calculated in our study because this calculation was not relevant in our study design. Nevertheless, the antral area measured in the right lateral decubitus position probably correlates better with the gastric fluid volume than when the measurement is performed in the supine position.9,25 Further studies in parturients should be conducted to assess the relationship between the antral area measured in the right lateral decubitus position and the gastric fluid volume, even if, in clinical practice, the right lateral decubitus position may be not desirable for some parturients.
In our study, about two-thirds of women with a shorter fasting period (last solid meal <8 h) had an antral area more than 381 mm2, whereas two-thirds of women after a prolonged fasting period (>8 h) had an antral area less than 381 mm2. Unfortunately, the last food ingested, and especially its caloric value, were not recorded so we cannot comment on gastric volume and food type. Nevertheless, our result supports the hypothesis that gastric emptying is delayed,6 but is not interrupted,24 during labour. Consequently, if general anaesthesia is required during established labour, although a prolonged fasting period for solid foods (>8 h) may reduce the likelihood of pulmonary aspiration of gastric contents, it does not guarantee that the stomach is empty: hence, the interest in the ultrasound assessment of gastric contents before induction of anaesthesia.26 In the present observational study, we noted that the mean fasting duration for solids was 10 ± 5 h, almost 48% of the patients fasted for more than 8 h, and 45.2% of the patients had an empty stomach during the ultrasound examination. These data may explain why no aspiration was reported in those parturients receiving general anaesthesia without tracheal intubation.22,27
In the present study, the overall success rate for the ultrasound examination of the antrum was 85.6%. Nevertheless, among the 13 inconclusive ultrasound examinations, in only five women was it not possible to identify the antrum in either the supine or the right lateral position. In the remaining eight women, the antrum could still be assessed in one position – either the supine or the right lateral decubitus position. Arzola et al.28 previously reported that ultrasound qualitative assessment of gastric contents was challenging in non-labouring pregnant women, particularly when BMI and gestational age were increased and he reported an overall incidence of 12.5% of incorrect diagnoses. In labouring women, Bataille et al.24 reported a success rate of 96% when measuring the antral cross-sectional area in the supine position, whereas in our study, the success rate for this measurement was 90% (81/90). Thus, for the large majority of women in active labour, ultrasound assessment of gastric contents is feasible and may therefore be useful in clinical practice.
Our study has several limitations. Our results are based on the comparison of data provided by one single technique (ultrasound examination of the antral area), and the actual gastric volume of each parturient was not confirmed by other non-invasive methods such as gastric tomodensitometry or MRI. For ethical reasons, it is difficult to justify an invasive method for the assessment of gastric contents in the parturient. To minimise confounding factors, the ultrasound examinations were performed by two operators blinded to each other's results. Furthermore, the hypothesis that antral grade 0 may correspond to low gastric volume, ranging from 0 to 0.4 ± 0.2 ml kg−1, depends on a mathematical model validated in non-pregnant adults only.12 The cut-off value of the antral area may be useful only for defining an empty stomach, corresponding to an antral area less than 381 mm2. The visualisation of gastric contents when the ultrasound is performed in the supine position (antral grade 2 or 3) is probably sufficient for defining the stomach contents as being ‘at risk’ of regurgitation and subsequent aspiration. However, a grey zone (defined by the combination of an antral area >381 mm2 with no visualisation of gastric contents) remains difficult to interpret and required further investigation.
In conclusion, the 381 mm2 cut-off value of the antral area measured in the supine position in the parturient allows discrimination of antral grade 0 (empty stomach) from antral grade at least 1 with an acceptable performance. A single examination of the antrum, performed in the supine position in the parturient, allows both qualitative analysis of gastric content for a fast diagnosis of a full stomach (grades 2 and 3), and measurement of the antral area for the fast diagnosis of an empty stomach (area <381mm2 and no visualisation of gastric content). Although probably less accurate than the measurement performed in the right lateral decubitus position, this point-of-care tool may be used by anaesthetists to help them to choose the most appropriate anaesthetic technique, according to the potential risk of regurgitation. Further studies are required in parturients to confirm our results and to define more precisely the usefulness of the ultrasonographic measurement of the antral area in the prevention strategy of pulmonary aspiration of gastric contents.
Acknowledgements relating to this article
Assistance with the study: none.
Conflicts of interest: none.
Financial sponsorship and support: none.
1. McClure JH, Cooper GM, Clutton-Brock TH. Saving mothers’ lives: reviewing maternal deaths to make motherhood safer: 2006-8: a review. Br J Anaesth
2. Vanner RG, Goodman NW. Gastro-oesophageal reflux in pregnancy at term and after delivery. Anaesthesia
3. Kinsella SM, Winton AL, Mushambi MC, et al. Failed tracheal intubation during obstetric general anaesthesia: a literature review. Int J Obstet Anesth
4. Wright PM, Allen RW, Moore J, Donnelly JP. Gastric emptying during lumbar extradural analgesia in labour: effect of fentanyl supplementation. Br J Anaesth
5. Raidoo DM, Rocke DA, Brock-Utne JG, et al. Critical volume for pulmonary acid aspiration: reappraisal in a primate model. Br J Anaesth
6. Carp H, Jayaram A, Stoll M. Ultrasound examination of the stomach contents of parturients. Anesth Analg
7. Bouvet L, Mazoit JX, Chassard D, et al. Clinical assessment of the ultrasonographic measurement of antral area for estimating preoperative gastric content and volume. Anesthesiology
8. Bouvet L, Miquel A, Chassard D, et al. Could a single standardized ultrasonographic measurement of antral area be of interest for assessing gastric contents? A preliminary report. Eur J Anaesthesiol
9. Perlas A, Chan VW, Lupu CM, et al. Ultrasound assessment of gastric content and volume. Anesthesiology
10. Perlas A, Davis L, Khan M, et al. Gastric sonography in the fasted surgical patient: a prospective descriptive study. Anesth Analg
11. Bouvet L, Chassard D. Contribution of ultrasonography for the preoperative assessment of gastric contents. Ann Fr Anesth Reanim
12. Perlas A, Mitsakakis N, Liu L, et al. Validation of a mathematical model for ultrasound assessment of gastric volume by gastroscopic examination. Anesth Analg
13. Arzola C, Perlas A, Siddiqui NT, Carvalho JC. Bedside gastric ultrasonography in term pregnant women before elective cesarean delivery: a prospective cohort study. Anesth Analg
14. Kruisselbrink R, Arzola C, Endersby R, et al. Intra- and interrater reliability of ultrasound assessment of gastric volume. Anesthesiology
15. Vandenbroucke JP, von Elm E, Altman DG, et al. Strengthening the Reporting of Observational Studies in Epidemiology (STROBE): explanation and elaboration. PLoS Med
16. Bolondi L, Bortolotti M, Santi V, et al. Measurement of gastric emptying time by real-time ultrasonography. Gastroenterology
17. Cubillos J, Tse C, Chan VW, Perlas A. Bedside ultrasound assessment of gastric content: an observational study. Can J Anaesth
18. Darwiche G, Almer LO, Bjorgell O, et al. Measurement of gastric emptying by standardized real-time ultrasonography in healthy subjects and diabetic patients. J Ultrasound Med
19. Engelhardt T, Webster NR. Pulmonary aspiration of gastric contents in anaesthesia. Br J Anaesth
20. Kluger MT, Short TG. Aspiration during anaesthesia: a review of 133 cases from the Australian Anaesthetic Incident Monitoring Study (AIMS). Anaesthesia
21. Sakai T, Planinsic RM, Quinlan JJ, et al. The incidence and outcome of perioperative pulmonary aspiration in a university hospital: a 4-year retrospective analysis. Anesth Analg
22. Zieleskiewicz L, Bellefleur JP, Antonini F, et al. Airway management for anaesthesia performed at the end of labour: survey of practices. Ann Fr Anesth Reanim
23. Zieleskiewicz L, Bellefleur JP, Leone M. Upper airway management in obsterics: results of a French survey. Can J Anaesth
24. Bataille A, Rousset J, Marret E, Bonnet F. Ultrasonographic evaluation of gastric content during labour under epidural analgesia: a prospective cohort study. Br J Anaesth
25. Van de Putte P, Perlas A. Ultrasound assessment of gastric content and volume. Br J Anaesth
26. American Society of Anesthesiologists Task Force on Obstetric Anesthesia. Practice guidelines for obstetric anesthesia: an updated report by the American Society of Anesthesiologists Task Force on Obstetric Anesthesia. Anesthesiology
27. Ezri T, Szmuk P, Stein A, et al. Peripartum general anasthesia without tracheal intubation: incidence of aspiration pneumonia. Anaesthesia
28. Arzola C, Cubillos J, Perlas A, et al. Interrater reliability of qualitative ultrasound assessment of gastric content in the third trimester of pregnancy. Br J Anaesth