In this retrospective case-control study, we investigated the difference in gastric ultrasound findings between elderly patients and young patients. Among patients with Perlas grade 0, which refers to the empty stomach, the CSA supine and CSA RLDP were greater in the elderly group than in the young group. The CSA supine could not predict the risk of pulmonary aspiration in the young group. However, the CSA RLDP in the young group and the CSA supine and CSA RLDP in the old group could predict the risk of pulmonary aspiration. The CSA RLDP cutoff value for predicting high risk of pulmonary aspiration in the young group was 6.92 cm2. In the elderly group, the CSA supine and CSA RLDP cutoff values for predicting high risk of pulmonary aspiration were 7.03 cm2 and 10.96 cm2, respectively.
Aspiration pneumonia is one of the major complications related to general anesthesia.[14–16] Therefore, it is essential for patients to fast prior to surgery to prevent pulmonary aspiration.[15, 17] Nowadays, the emptiness of the stomach can be easily assessed with gastric ultrasound at the bedside. The assessment of pulmonary aspiration risk with gastric ultrasound involves several steps. First, qualitative assessment of gastric contents (i.e., whether the stomach is empty or contains fluid or solid) is performed. The risk of pulmonary aspiration is considered high if solids are present in the stomach and is considered low if the stomach is empty. However, if liquid is present in the stomach, additional quantitative evaluation is required. The risk of pulmonary aspiration is evaluated using a semi-quantitative three-point grading scale (grades 0, 1, and 2) that considers the difference in qualitative assessments performed in two different positions (supine position and RLDP).[4, 10] In the quantitative assessment, the CSA cutoff value is used to discriminate patients with a high risk of pulmonary aspiration. Alternatively, the CSA is used to calculate the estimated GV; an estimated GV per kg of >1.5 mL/kg is considered an indicator of high pulmonary aspiration risk.[4, 5]
Although the mathematical models for calculating GVs are known to be reliable,[2, 3] they sometimes lack reproducibility after adjusting for different patient populations. In addition, the assessment of the risk of pulmonary aspiration with mathematical models may be somewhat cumbersome compared with that with the grading system or the CSA. This is because mathematical models involve calculation by combining specific constants. Therefore, we thought that simple and intuitive evaluation methods such as grading and use of CSA cutoff values can be more easily adopted in actual clinical practice. However, because elderly patients have higher gastric compliance than younger patients, we expected that the prediction of pulmonary aspiration risk with a semi-quantitative CSA cutoff value would be different in elderly and young patients.
In our study, we investigated whether Perlas grade 2 predicted a high risk of pulmonary aspiration in each group. The positive predictive value of Perlas grade 2 was low in both the groups, with 50% in the elderly group and 19% in the young group. Based on our study results, we agree with a previous study suggesting that Perlas grade 2 should be used as a screening test and a guide for further pulmonary aspiration risk assessments. However, because our sample size was small, we believe that our results should be validated in well-designed prospective studies with larger sample sizes.
Some studies have reported the use of CSA cutoff values for the diagnosis of a high risk of pulmonary aspiration.[7, 8] For example, Bouvet et al. reported that a CSA of >320 mm2 is associated with an increased risk of pulmonary aspiration. Another study reported that a CSA of 340 mm2 indicated an increased risk of pulmonary aspiration with a sensitivity of 91% and a specificity of 71%. The AUC for the diagnosis of a high risk of pulmonary aspiration was 90%. However, the CSA cutoff values presented in these studies have a limitation in that age was not considered. In our study, the CSA cutoff value was greater in elderly patients than in young patients. We believe that this was because of the higher gastric compliance in elderly patients, as revealed in a previous study. Therefore, we believe that the CSA cutoff value for evaluating the risk of pulmonary aspiration should be greater for elderly patients than for young patients. Moreover, the application of the CSA cutoff value for the young group (CSA RLDP = 6.92 cm2) to the elderly group decreased the specificity, positive predictive value, and accuracy. Thus, we suggest the following CSA cutoff values for the elderly patients: CSA supine, 6.92 cm2 and CSA RLDP, 10.65 cm2.
It is well-known that the accuracy of the assessment of gastric residual liquid with ultrasound in the RLDP is greater than in the supine position because of gravity. However, the assessment cannot be performed in the RLDP in some patients with clinical conditions such as extreme pain hip fracture, mental disorder, and poor cooperation. In this study, the CSA supine had a good predictive value (AUC: 0.818) for high pulmonary aspiration risk. Therefore, we carefully suggest that gastric ultrasound in the supine position with a head up tilt at 45 degrees would be helpful for predicting the risk of pulmonary aspiration in elderly patients, when gastric ultrasound in the RLDP cannot be performed.
This study had some limitations. First, this was a retrospective study. Because the number of patients who underwent preoperative gastric ultrasound was small, the sample size of our study was small. Therefore, extrapolation of our results to broader patient groups with various underlying diseases should be performed with caution. Second, the residual GV was not determined based on the aspiration of actual gastric contents. However, the reproducibility of the GV calculation formula has been proven in previous studies. Moreover, the measurement of GV through gastric suction is not considered to be a reliable tool for assessing the GV. Therefore, we think that GV evaluation determined by gastric suction would not affect our study outcomes.
Third, we compared the two different populations; therefore, some differences were found in the demographic characteristics of the patients of the two groups. The proportion of female patients in the young group was 100%, whereas that in the old group was 20.1%. However, because sex does not affect the CSA, we believe that the difference in the proportion of female patients would not have affected our outcomes. The differences in weight, body mass index, and ASA class might be because of the differences in sex ratio or age between the groups. However, because these factors do not significantly affect GV assessment, these factors might not have significantly affected our results. Furthermore, comorbid diseases increase with age, and the ASA class was inevitably higher in the elderly group than in young group. The solid fasting time was longer in the elderly group (14 hours) than in the young group (12 hours). However, gastric emptying of liquid is not affected by gastric emptying of solid food because solid food is emptied by peristalsis or pressure pump mechanism. In addition, the fasting time for solids in both the groups exceeded the gastric emptying time of the solid food (5.8 ± 0.8 hours). Therefore, it is thought that a 3-hour difference in fasting time for solids between the two groups would have not affected our outcomes.