Ultrasound Examination of the Stomach Contents of Parturients H. Carp,
PhD, MD,
A. Jayaram, MD, and M. Stoll,
PhD, MD
Section of Obstetric Anesthesia, Department of Anesthesia, and Department of Diagnostic Radiology, Oregon Health Sciences University, Portland, Oregon
A noninvasive method to determine the actual stomach contents of parturients may help to identify factors responsible for delayed gastric emptying as well as define the risk of aspiration of gastric contents in individual patients. Therefore, we tested the ability of ultrasound imaging to identify noninvasively the stomach contents of laboring and nonlaboring pregnant volunteers. A preliminary study demonstrated that the stomach contents could be identified by ultrasound in 20 healthy volunteers and in 34 parturients, not yet in active labor, scanned after consuming liquids or solid food. Next, 39 parturients in active labor were scanned at varying postprandial times.
T
he risk of aspiration of stomach contents during obstetric anesthesia was recognized by Mendelson (1) in 1946. Although he described the syndrome of acid aspiration, the only deaths in his clinical study were the result of aspiration of solid food remaining in the stomach many hours after the cessation of oral intake. A noninvasive method to determine the actual stomach contents of parturients may help to identify factors responsible for delayed gastric emptying as well as define the potential risk of aspiration in individual patients. Real-time ultrasound is a technique capable of highresolution imaging. This technique produces good quality images of the stomach in nonpregnant volunteers ( 2 4 ) . Therefore, we tested the ability of ultrasound imaging to identify noninvasively the stomach contents of laboring patients and healthy volunteers at varying times after the ingestion of solid food.
Methods The study had the approval of the Committee on Human Research at the Oregon Health Sciences Accepted for publication February 7, 1992. Address correspondence to Dr. Carp, Department of Anesthesia, Oregon Health Sciences University, 3181 S.W. Sam Jackson Park Road, Portland, OR 97201. 01992 by the International Anesthesia Research Society 0003-2999/92/$5.00
Sixteen parturients in active labor who had not eaten for &24 h still had food detected in the stomach. In fact, nearly two-thirds of the patients in active labor who were scanned had solid food present in the stomach independent of the interval between last oral intake and the ultrasound scan. The present study demonstrates that high-resolution ultrasonography is capable of noninvasively identifying the stomach contents of parturients. These results confirm the clinical impression that emptying of the stomach is delayed for many hours after the onset of labor. (Anesth Analg 1992;74:683-7)
University. Written, informed consent was obtained from all patients. A pilot study was first performed with 20 healthy, nonpregnant volunteers. The pregnant study group comprised 73 healthy women with uncomplicated, near-term single pregnancies. Thirtynine of these patients were studied in the labor ward while in active labor after a request for epidural anesthesia. Patients in labor were studied within 1 h of epidural placement; 0.25% bupivicaine (10 mL) was administered through the epidural catheter to obtain patient comfort. Thirty-four of the 73 full-term pregnant patients, not yet in labor, were studied at the time of routine antenatal testing ( n = 30) or before elective cesarean delivery ( n = 4). All patients were English speaking, over the age of 18 yr, free from gastrointestinal tract disease, and had not received narcotic analgesics before the study. Patients with a history of previous abdominal surgery, alcohol or drug abuse, infection, preterm labor, hemorrhage, hypertension, diabetes, morbid obesity, or hyperemesis or patients receiving any medication (other than prenatal vitamins) were excluded from the study. Patients participating in the study were asked to note retrospectively the time that they last ingested “solid food.” The type and quantity of food were also recorded for each patient. After admission to the labor ward, patients were allowed unrestricted access Anesth Analg 1992;?4:68>7
683
684
OBSTETRIC ANESTHESIA GASTRIC ULTRASOUND
CARP ET AL
to clear liquid beverages only. Four patients ingested solid food after the onset of labor, and they were excluded from the study analysis. All patients received 30 mL of sodium citrate at the time of epidural catheter placement. Patients in active labor were also asked to estimate the time of onset of painful uterine contractions. Volunteers undergoing serial ultrasound examinations were fed a standardized meal identical to the breakfast prepared for obstetric inpatients at our hospital. The meal consisted of fruit juice, muffin, roll, butter, jam, and cereal with milk and weighed approximately 800 g. Ultrasound examination was performed with a real-time ultrasound transducer with a linear phased array probe (5 MHz) (model RT3000, General Electric, Milwaukee, Wis.). The ultrasonographer was unaware of the time of the last oral intake of the patient. Gastric ultrasound examination was performed according to a modification of the technique previously described by Holt et al. (2), in which elevation of the patient's head aids in visualization of the stomach by displacing gastric air upward out of the scanning field, thereby preventing scattering of the ultrasound beam by the air. Transverse views of the stomach were obtained by initially positioning the scanner under the left costal margin and scanning up to the xiphisternum. Identification of the stomach was always confirmed by directly observing the entry of 1-2 cups of water into the stomach. To displace the gravid uterus away from the ultrasound beam, pregnant patients were scanned in the right lateral position, with the head of the bed elevated to 45", rather than in the sitting position described by Holt et al. (2). For convenience, nonpregnant volunteers were scanned in the sitting position after it was determined that identical results were obtained on ultrasound examination of five healthy volunteers in both the sitting and right lateral position. With this technique, the stomach could be clearly identified in approximately 60% of the pregnant patients and in all of the nonpregnant volunteers tested. The ingestion of as little as 2 oz of beef, fish, chicken, or bread could be detected in the stomachs of healthy volunteers ( n = lo), and these foods were all similar in appearance on ultrasound. The ingestion of carbonated liquids interfered with ultrasound visualization of gastric contents for 10-15 min. Therefore no patient was scanned for at least 15 min after ingestion of a carbonated beverage. Data analysis was performed with the ?-test and Student's unpaired t-test. A P value < 0.05 was considered statistically significant.
ANESTH ANALG 1992;74:68>7
Figure 1. Ultrasound images of the stomach of a healthy volunteer under varying conditions. (A) Fluid-filled stomach (sto). (B) Stomach (STO) containing a prenatal vitamin pill (white arrow).
Results We first performed a pilot study with healthy nonpregnant volunteers to confirm the appearance of the stomach after the consumption of liquids or solid food and after an overnight fast. As shown in Figure lA, the stomach appears as a large cystic structure in a fasting volunteer after the ingestion of 2 cups of water before the scan. In real time, water could be seen streaming into the stomach; however, before ingestion of water, the empty stomach was not visible on ultrasound (data not shown). Next, a gastric scan was obtained after the ingestion of five prenatal vitamin pills (by the same volunteer as in Figure 1A). One of the pills is visible within the stomach (Figure 1B). One final illustration of this technique is shown in Figure 2. This scan demonstrates the appearance of the stomach immediately after the ingestion of a full lunch. Food particles are clearly visible within the stomach (Figure 2), and in real time, particles could be seen floating in the stomach. Next, 20 healthy
685
OBSTETRIC ANESTHESIA CARP ET AL. GASTRIC ULTRASOUND
ANESTH ANALG 1992;74683-7
Table 1. Relation of Time of Last Food Intake, First Ultrasound Examination, and Detection of Food i n the Stomachs of 39 Patients Hours between food intake and first ultrasound examination No. of patients with food in stomach
0-4
4-8
842
12-24
6
6"
6"
10"
(n = 6)
( n = 8)
( n = 10)
( n = 15)
n, total number of patients in group "No statistically significant difference compared with the M - h group.
Figure 2. Ultrasound image of the stomach (STO) of a healthy volunteer after ingestion of a full meal. Food particles can be seen within the stomach (dark urrows).
volunteers (10 women, mean age 31 +- 6 yr and 10 men, mean age 29 2 7 yr) underwent gastric ultrasound every 2 h after an overnight fast (8-10 h) and after ingesting a standardized meal (see Methods). In all cases, gastric ultrasound confirmed the absence of solid food in the stomach of the fasting volunteers. Furthermore, 4 h after ingestion of the meal, solid food was no longer detected in the stomach of any of the volunteers. These results demonstrate that ultrasound produces high-quality images of the stomach and permits detection of solid food within the stomach. We next studied 20 pregnant volunteers (mean maternal age 32 ?I 8 yr; mean gestational age 35 t 8 wk) not yet in labor and performed ultrasound examinations at 2-h intervals after an overnight fast (8-10 h) and after ingestion of a standardized meal. In all cases a gastric ultrasound scan confirmed the absence of solid food in the stomach of the fasting parturients. Furthermore, in each of the 20 parturients, solid food was no longer detected in the stomach by 4 h after ingestion of the meal. Fourteen additional pregnant volunteers (mean maternal age 28 2 6 yr; mean gestational age 35 5 8 wk) not yet in labor had a single ultrasound examination performed at varying times after eating solid food. In all cases, these patients did not have solid food detected in the stomach, provided that they had not eaten for at least 4 h before the ultrasound scan. Next, 39 healthy, full-term parturients in active labor (mean maternal age 28 k 10 yr; mean gestational age 36 k 9 wk) who requested epidural anal-
gesia had a single gastric ultrasound scan performed at varying times after eating solid food. All patients were studied within 1 h of epidural placement, and none of the patients had received narcotic analgesics. As shown in Table 1, gastric ultrasound examination demonstrated solid food within the stomachs of nearly two-thirds of these patients, independent of the time interval between the last oral intake and the ultrasound examination. Sixteen parturients who reported not eating for 8-24 h had solid food detected by gastric ultrasound examination (Table 1). The ultrasound results were consistent with the contents of vomitus obtained in four patients after ultrasound examination (three had had solid food and one liquid). In addition, the contents present in a large-bore gastric tube aspirate from three patients who subsequently required cesarean delivery under general anesthesia were consistent with the previous ultrasound results (two had had solid food and one liquid).
Discussion Mendelson (1)first reported in 1946 the development of pneumonitis and death after the inhalation of vomitus in pregnant patients undergoing general anesthesia. Since that time, numerous reports on maternal morbidity and mortality have drawn attention to this syndrome (5-7). The volume and acidity of aspirated vomitus have been identified in laboratory studies as important factors in determining the severity of the resulting pneumonitis (1,s). In addition, the presence of chunks of solid food within the aspirated vomitus is particularly dangerous. Aspiration of large chunks of food may result in mechanical obstruction of the airway, hypoxia, and death. One of the principles of anesthetic management is to consider all parturients at risk for aspiration of gastric contents because of delayed gastric emptying. The results of the present study using ultrasound to directly image solid food present in the stomachs of parturients confirm the
686
OBSTETRIC ANESTHESIA GASTRIC ULTRASOUND
CARP ET AL
clinical impression that stomach emptying is delayed for many hours after the onset of labor. High-resolution ultrasonography permits the noninvasive measurement of gastric structure and contents, and this technique has been used by radiologists to quantitate gastric motility in healthy volunteers ( 2 4 ) . Under our study conditions, the empty stomach is not easily visualized on ultrasound. Using the technique described previously by Holt et al. (2), we were able to detect 2 oz of chewed food in the stomach. Furthermore, ultrasound was able to correctly identify full or empty stomachs in nonlaboring parturients fed known amounts of solid food and subjected to repeated examinations. In addition, the contents of gastric tube aspirate or vomitus, when present, was also found to be consistent with the results of a previous ultrasound scan in laboring parturients. However, further study using other previously established techniques to measure gastric contents (e.g., endoscopy) will be required to validate this study and to determine the incidence of falsepositive or false-negative ultrasound results. In our study, ultrasound did not permit quantitative evaluation of gastric volume, as ingestion of 1-2 cups of liquid was required to identify the stomach. In fact, with ultrasound we were unable to identify the stomach in approximately 40% of the parturients scanned, and these patients were excluded from the study. Perhaps these technically inadequate scans actually represented patients whose stomachs were empty, and our results overestimated the percent of patients in active labor with food present in their stomachs. Nevertheless, even if all of the unvisualized stomachs represented empty stomachs and were included in the study, nearly 30% of patients in active labor would still have solid food in their stomachs. In contrast to the results with parturients, under the conditions of our study, ultrasound was able to visualize the stomach contents of all of the nonpregnant volunteers examined. This effect is probably the result of the stomach being displaced upward by the gravid uterus such that the rib cage interferes with penetration of the ultrasound beam. The reduced ability of ultrasound to identify the stomach in parturients will severely limit the clinical utility of this technique in pregnant patients. However, gastric ultrasound may be a useful technique to identify food in the stomachs of nonpregnant patients undergoing surgery after trauma, in diabetic patients with gastroparesis, or in patients undergoing postpartum tuba1 ligation. In addition, this technique may be useful to measure the effects of pharmacologic interventions designed to improve gastric function in nonpregnant pa tien ts. In our study, nearly two-thirds of parturients in
ANESTH ANALG 1992;74683-7
active labor had solid food detected in their stomachs, independent of the time of last oral intake (see Table 1). In fact, 16 parturients who had not eaten for 8-24 h before gastric ultrasound had food detected in their stomachs (see Table 1).Each of these 16 patients reported the onset of painful contractions within 4 h of last ingesting solid foods, suggesting that gastric emptying of solid food slows after the onset of painful contractions (data not shown). However, retrospective data provided by parturients describing the time of onset of contractions or oral intake may introduce considerable error. Further prospective study of the relation between the onset of uterine contractions and gastric emptying is required to confirm these preliminary observations. All parturients underwent gastric ultrasound within 1 h of the institution of epidural analgesia (using local anesthetic only). These patients were comfortable and cooperated with the examination. It is unlikely that epidural analgesia could be responsible for solid food appearing again in the stomachs of parturients who had not eaten for many hours before the ultrasound examination. Epidural analgesia with local anesthetic has actually been shown to hasten the return of gastrointestinal tract function after abdominal surgery (9). In view of this effect, serial gastric ultrasound examinations were performed every 2 h in five additional (mean maternal age 26 +- 7 yr; mean gestational age 37 t 8 wk) patients throughout the course of their labor (8-12 h) and u p to 8 h postpartum. These scans demonstrated that despite effective epidural analgesia, food was present in the stomach throughout labor; however, the food emptied rapidly (i.e., within 4-6 h) after delivery. Further studies will be required to determine whether epidural analgesia speeds gastric emptying in patients in active labor. Finally, it is not clear to what extent the presence of solid food in the stomach, demonstrated on gastric ultrasound, increases the risk of aspiration of gastric contents during anesthesia, as acidic gastric juice alone may cause severe pneumonitis (1,6,8). Our data do not permit measurement of gastric volume or acidity, and both of these factors may be increased in pregnant patients (6). Further studies will be required to determine the clinical significance of these radiographic findings in patients in active labor. In summary, the present study demonstrated that high-resolution ultrasonography is capable of noninvasively identifying the stomach contents of parturients. These radiographic findings confirm the clinical impression that solid food remains in the stomach for many hours after the onset of labor.
ANESTH ANALG 1992;7468>7
References 1. Mendelson CL. The aspiration of stomach contents into the lungs during obstetric anesthesia. Am J Obstet Gynecol 1946; 52~191-205. 2. Holt S, McDicken WN, Anderson T, Steward IG, Heading RC. Dynamic imaging of the stomach by real-time ultrasound-a method for the study of gastric motility. Gut 1980;21:597-601. 3. Bateman DN, Whittingham TA. Measurement of gastric emptying by real-time ultrasound. Gut 1982;23:52&7. 4. Bateman DN, Keenan S, Metreweli C, Wilson K. A noninvasive technique for gastric motility measurement. Br J Radio1 1977;50:526-7.
OBSTETRIC ANESTHESIA CARP ET AL. GASTRIC ULTRASOUND
687
5. Tomkins J, Turnbull A, Robson G. Report on confidential enquiries into maternal deaths in England and Wales, 19731975. London: Her Majesty’s Stationary Office, 1979. 6. Roberts RB, Shirley MA. Reducing the risk of acid aspiration during cesarean section. Anesth Analg 1974;53:85948. 7. Moir DD. Maternal mortality and anaesthesia. Br J Anaesth 1980;52:1-3. 8. Raidoo DM, Rocke DA, Brock-Utne JG, Marszalek A, Engelbrecht HE. Critical volume for pulmonary aspiration: reappraisal in a primate model. Br J Anaesth 1990;65:248-50. 9. Wattwil M, Tho& T, Hennerdal S, Garvill, JE. Epidural analgesia with bupivicaine reduces postoperative paralytic ileas after hysterectomy. Anesth Analg 1989;68:353-8.
PhD, MD,
A. Jayaram, MD, and M. Stoll,
PhD, MD
Section of Obstetric Anesthesia, Department of Anesthesia, and Department of Diagnostic Radiology, Oregon Health Sciences University, Portland, Oregon
A noninvasive method to determine the actual stomach contents of parturients may help to identify factors responsible for delayed gastric emptying as well as define the risk of aspiration of gastric contents in individual patients. Therefore, we tested the ability of ultrasound imaging to identify noninvasively the stomach contents of laboring and nonlaboring pregnant volunteers. A preliminary study demonstrated that the stomach contents could be identified by ultrasound in 20 healthy volunteers and in 34 parturients, not yet in active labor, scanned after consuming liquids or solid food. Next, 39 parturients in active labor were scanned at varying postprandial times.
T
he risk of aspiration of stomach contents during obstetric anesthesia was recognized by Mendelson (1) in 1946. Although he described the syndrome of acid aspiration, the only deaths in his clinical study were the result of aspiration of solid food remaining in the stomach many hours after the cessation of oral intake. A noninvasive method to determine the actual stomach contents of parturients may help to identify factors responsible for delayed gastric emptying as well as define the potential risk of aspiration in individual patients. Real-time ultrasound is a technique capable of highresolution imaging. This technique produces good quality images of the stomach in nonpregnant volunteers ( 2 4 ) . Therefore, we tested the ability of ultrasound imaging to identify noninvasively the stomach contents of laboring patients and healthy volunteers at varying times after the ingestion of solid food.
Methods The study had the approval of the Committee on Human Research at the Oregon Health Sciences Accepted for publication February 7, 1992. Address correspondence to Dr. Carp, Department of Anesthesia, Oregon Health Sciences University, 3181 S.W. Sam Jackson Park Road, Portland, OR 97201. 01992 by the International Anesthesia Research Society 0003-2999/92/$5.00
Sixteen parturients in active labor who had not eaten for &24 h still had food detected in the stomach. In fact, nearly two-thirds of the patients in active labor who were scanned had solid food present in the stomach independent of the interval between last oral intake and the ultrasound scan. The present study demonstrates that high-resolution ultrasonography is capable of noninvasively identifying the stomach contents of parturients. These results confirm the clinical impression that emptying of the stomach is delayed for many hours after the onset of labor. (Anesth Analg 1992;74:683-7)
University. Written, informed consent was obtained from all patients. A pilot study was first performed with 20 healthy, nonpregnant volunteers. The pregnant study group comprised 73 healthy women with uncomplicated, near-term single pregnancies. Thirtynine of these patients were studied in the labor ward while in active labor after a request for epidural anesthesia. Patients in labor were studied within 1 h of epidural placement; 0.25% bupivicaine (10 mL) was administered through the epidural catheter to obtain patient comfort. Thirty-four of the 73 full-term pregnant patients, not yet in labor, were studied at the time of routine antenatal testing ( n = 30) or before elective cesarean delivery ( n = 4). All patients were English speaking, over the age of 18 yr, free from gastrointestinal tract disease, and had not received narcotic analgesics before the study. Patients with a history of previous abdominal surgery, alcohol or drug abuse, infection, preterm labor, hemorrhage, hypertension, diabetes, morbid obesity, or hyperemesis or patients receiving any medication (other than prenatal vitamins) were excluded from the study. Patients participating in the study were asked to note retrospectively the time that they last ingested “solid food.” The type and quantity of food were also recorded for each patient. After admission to the labor ward, patients were allowed unrestricted access Anesth Analg 1992;?4:68>7
683
684
OBSTETRIC ANESTHESIA GASTRIC ULTRASOUND
CARP ET AL
to clear liquid beverages only. Four patients ingested solid food after the onset of labor, and they were excluded from the study analysis. All patients received 30 mL of sodium citrate at the time of epidural catheter placement. Patients in active labor were also asked to estimate the time of onset of painful uterine contractions. Volunteers undergoing serial ultrasound examinations were fed a standardized meal identical to the breakfast prepared for obstetric inpatients at our hospital. The meal consisted of fruit juice, muffin, roll, butter, jam, and cereal with milk and weighed approximately 800 g. Ultrasound examination was performed with a real-time ultrasound transducer with a linear phased array probe (5 MHz) (model RT3000, General Electric, Milwaukee, Wis.). The ultrasonographer was unaware of the time of the last oral intake of the patient. Gastric ultrasound examination was performed according to a modification of the technique previously described by Holt et al. (2), in which elevation of the patient's head aids in visualization of the stomach by displacing gastric air upward out of the scanning field, thereby preventing scattering of the ultrasound beam by the air. Transverse views of the stomach were obtained by initially positioning the scanner under the left costal margin and scanning up to the xiphisternum. Identification of the stomach was always confirmed by directly observing the entry of 1-2 cups of water into the stomach. To displace the gravid uterus away from the ultrasound beam, pregnant patients were scanned in the right lateral position, with the head of the bed elevated to 45", rather than in the sitting position described by Holt et al. (2). For convenience, nonpregnant volunteers were scanned in the sitting position after it was determined that identical results were obtained on ultrasound examination of five healthy volunteers in both the sitting and right lateral position. With this technique, the stomach could be clearly identified in approximately 60% of the pregnant patients and in all of the nonpregnant volunteers tested. The ingestion of as little as 2 oz of beef, fish, chicken, or bread could be detected in the stomachs of healthy volunteers ( n = lo), and these foods were all similar in appearance on ultrasound. The ingestion of carbonated liquids interfered with ultrasound visualization of gastric contents for 10-15 min. Therefore no patient was scanned for at least 15 min after ingestion of a carbonated beverage. Data analysis was performed with the ?-test and Student's unpaired t-test. A P value < 0.05 was considered statistically significant.
ANESTH ANALG 1992;74:68>7
Figure 1. Ultrasound images of the stomach of a healthy volunteer under varying conditions. (A) Fluid-filled stomach (sto). (B) Stomach (STO) containing a prenatal vitamin pill (white arrow).
Results We first performed a pilot study with healthy nonpregnant volunteers to confirm the appearance of the stomach after the consumption of liquids or solid food and after an overnight fast. As shown in Figure lA, the stomach appears as a large cystic structure in a fasting volunteer after the ingestion of 2 cups of water before the scan. In real time, water could be seen streaming into the stomach; however, before ingestion of water, the empty stomach was not visible on ultrasound (data not shown). Next, a gastric scan was obtained after the ingestion of five prenatal vitamin pills (by the same volunteer as in Figure 1A). One of the pills is visible within the stomach (Figure 1B). One final illustration of this technique is shown in Figure 2. This scan demonstrates the appearance of the stomach immediately after the ingestion of a full lunch. Food particles are clearly visible within the stomach (Figure 2), and in real time, particles could be seen floating in the stomach. Next, 20 healthy
685
OBSTETRIC ANESTHESIA CARP ET AL. GASTRIC ULTRASOUND
ANESTH ANALG 1992;74683-7
Table 1. Relation of Time of Last Food Intake, First Ultrasound Examination, and Detection of Food i n the Stomachs of 39 Patients Hours between food intake and first ultrasound examination No. of patients with food in stomach
0-4
4-8
842
12-24
6
6"
6"
10"
(n = 6)
( n = 8)
( n = 10)
( n = 15)
n, total number of patients in group "No statistically significant difference compared with the M - h group.
Figure 2. Ultrasound image of the stomach (STO) of a healthy volunteer after ingestion of a full meal. Food particles can be seen within the stomach (dark urrows).
volunteers (10 women, mean age 31 +- 6 yr and 10 men, mean age 29 2 7 yr) underwent gastric ultrasound every 2 h after an overnight fast (8-10 h) and after ingesting a standardized meal (see Methods). In all cases, gastric ultrasound confirmed the absence of solid food in the stomach of the fasting volunteers. Furthermore, 4 h after ingestion of the meal, solid food was no longer detected in the stomach of any of the volunteers. These results demonstrate that ultrasound produces high-quality images of the stomach and permits detection of solid food within the stomach. We next studied 20 pregnant volunteers (mean maternal age 32 ?I 8 yr; mean gestational age 35 t 8 wk) not yet in labor and performed ultrasound examinations at 2-h intervals after an overnight fast (8-10 h) and after ingestion of a standardized meal. In all cases a gastric ultrasound scan confirmed the absence of solid food in the stomach of the fasting parturients. Furthermore, in each of the 20 parturients, solid food was no longer detected in the stomach by 4 h after ingestion of the meal. Fourteen additional pregnant volunteers (mean maternal age 28 2 6 yr; mean gestational age 35 5 8 wk) not yet in labor had a single ultrasound examination performed at varying times after eating solid food. In all cases, these patients did not have solid food detected in the stomach, provided that they had not eaten for at least 4 h before the ultrasound scan. Next, 39 healthy, full-term parturients in active labor (mean maternal age 28 k 10 yr; mean gestational age 36 k 9 wk) who requested epidural anal-
gesia had a single gastric ultrasound scan performed at varying times after eating solid food. All patients were studied within 1 h of epidural placement, and none of the patients had received narcotic analgesics. As shown in Table 1, gastric ultrasound examination demonstrated solid food within the stomachs of nearly two-thirds of these patients, independent of the time interval between the last oral intake and the ultrasound examination. Sixteen parturients who reported not eating for 8-24 h had solid food detected by gastric ultrasound examination (Table 1). The ultrasound results were consistent with the contents of vomitus obtained in four patients after ultrasound examination (three had had solid food and one liquid). In addition, the contents present in a large-bore gastric tube aspirate from three patients who subsequently required cesarean delivery under general anesthesia were consistent with the previous ultrasound results (two had had solid food and one liquid).
Discussion Mendelson (1)first reported in 1946 the development of pneumonitis and death after the inhalation of vomitus in pregnant patients undergoing general anesthesia. Since that time, numerous reports on maternal morbidity and mortality have drawn attention to this syndrome (5-7). The volume and acidity of aspirated vomitus have been identified in laboratory studies as important factors in determining the severity of the resulting pneumonitis (1,s). In addition, the presence of chunks of solid food within the aspirated vomitus is particularly dangerous. Aspiration of large chunks of food may result in mechanical obstruction of the airway, hypoxia, and death. One of the principles of anesthetic management is to consider all parturients at risk for aspiration of gastric contents because of delayed gastric emptying. The results of the present study using ultrasound to directly image solid food present in the stomachs of parturients confirm the
686
OBSTETRIC ANESTHESIA GASTRIC ULTRASOUND
CARP ET AL
clinical impression that stomach emptying is delayed for many hours after the onset of labor. High-resolution ultrasonography permits the noninvasive measurement of gastric structure and contents, and this technique has been used by radiologists to quantitate gastric motility in healthy volunteers ( 2 4 ) . Under our study conditions, the empty stomach is not easily visualized on ultrasound. Using the technique described previously by Holt et al. (2), we were able to detect 2 oz of chewed food in the stomach. Furthermore, ultrasound was able to correctly identify full or empty stomachs in nonlaboring parturients fed known amounts of solid food and subjected to repeated examinations. In addition, the contents of gastric tube aspirate or vomitus, when present, was also found to be consistent with the results of a previous ultrasound scan in laboring parturients. However, further study using other previously established techniques to measure gastric contents (e.g., endoscopy) will be required to validate this study and to determine the incidence of falsepositive or false-negative ultrasound results. In our study, ultrasound did not permit quantitative evaluation of gastric volume, as ingestion of 1-2 cups of liquid was required to identify the stomach. In fact, with ultrasound we were unable to identify the stomach in approximately 40% of the parturients scanned, and these patients were excluded from the study. Perhaps these technically inadequate scans actually represented patients whose stomachs were empty, and our results overestimated the percent of patients in active labor with food present in their stomachs. Nevertheless, even if all of the unvisualized stomachs represented empty stomachs and were included in the study, nearly 30% of patients in active labor would still have solid food in their stomachs. In contrast to the results with parturients, under the conditions of our study, ultrasound was able to visualize the stomach contents of all of the nonpregnant volunteers examined. This effect is probably the result of the stomach being displaced upward by the gravid uterus such that the rib cage interferes with penetration of the ultrasound beam. The reduced ability of ultrasound to identify the stomach in parturients will severely limit the clinical utility of this technique in pregnant patients. However, gastric ultrasound may be a useful technique to identify food in the stomachs of nonpregnant patients undergoing surgery after trauma, in diabetic patients with gastroparesis, or in patients undergoing postpartum tuba1 ligation. In addition, this technique may be useful to measure the effects of pharmacologic interventions designed to improve gastric function in nonpregnant pa tien ts. In our study, nearly two-thirds of parturients in
ANESTH ANALG 1992;74683-7
active labor had solid food detected in their stomachs, independent of the time of last oral intake (see Table 1). In fact, 16 parturients who had not eaten for 8-24 h before gastric ultrasound had food detected in their stomachs (see Table 1).Each of these 16 patients reported the onset of painful contractions within 4 h of last ingesting solid foods, suggesting that gastric emptying of solid food slows after the onset of painful contractions (data not shown). However, retrospective data provided by parturients describing the time of onset of contractions or oral intake may introduce considerable error. Further prospective study of the relation between the onset of uterine contractions and gastric emptying is required to confirm these preliminary observations. All parturients underwent gastric ultrasound within 1 h of the institution of epidural analgesia (using local anesthetic only). These patients were comfortable and cooperated with the examination. It is unlikely that epidural analgesia could be responsible for solid food appearing again in the stomachs of parturients who had not eaten for many hours before the ultrasound examination. Epidural analgesia with local anesthetic has actually been shown to hasten the return of gastrointestinal tract function after abdominal surgery (9). In view of this effect, serial gastric ultrasound examinations were performed every 2 h in five additional (mean maternal age 26 +- 7 yr; mean gestational age 37 t 8 wk) patients throughout the course of their labor (8-12 h) and u p to 8 h postpartum. These scans demonstrated that despite effective epidural analgesia, food was present in the stomach throughout labor; however, the food emptied rapidly (i.e., within 4-6 h) after delivery. Further studies will be required to determine whether epidural analgesia speeds gastric emptying in patients in active labor. Finally, it is not clear to what extent the presence of solid food in the stomach, demonstrated on gastric ultrasound, increases the risk of aspiration of gastric contents during anesthesia, as acidic gastric juice alone may cause severe pneumonitis (1,6,8). Our data do not permit measurement of gastric volume or acidity, and both of these factors may be increased in pregnant patients (6). Further studies will be required to determine the clinical significance of these radiographic findings in patients in active labor. In summary, the present study demonstrated that high-resolution ultrasonography is capable of noninvasively identifying the stomach contents of parturients. These radiographic findings confirm the clinical impression that solid food remains in the stomach for many hours after the onset of labor.
ANESTH ANALG 1992;7468>7
References 1. Mendelson CL. The aspiration of stomach contents into the lungs during obstetric anesthesia. Am J Obstet Gynecol 1946; 52~191-205. 2. Holt S, McDicken WN, Anderson T, Steward IG, Heading RC. Dynamic imaging of the stomach by real-time ultrasound-a method for the study of gastric motility. Gut 1980;21:597-601. 3. Bateman DN, Whittingham TA. Measurement of gastric emptying by real-time ultrasound. Gut 1982;23:52&7. 4. Bateman DN, Keenan S, Metreweli C, Wilson K. A noninvasive technique for gastric motility measurement. Br J Radio1 1977;50:526-7.
OBSTETRIC ANESTHESIA CARP ET AL. GASTRIC ULTRASOUND
687
5. Tomkins J, Turnbull A, Robson G. Report on confidential enquiries into maternal deaths in England and Wales, 19731975. London: Her Majesty’s Stationary Office, 1979. 6. Roberts RB, Shirley MA. Reducing the risk of acid aspiration during cesarean section. Anesth Analg 1974;53:85948. 7. Moir DD. Maternal mortality and anaesthesia. Br J Anaesth 1980;52:1-3. 8. Raidoo DM, Rocke DA, Brock-Utne JG, Marszalek A, Engelbrecht HE. Critical volume for pulmonary aspiration: reappraisal in a primate model. Br J Anaesth 1990;65:248-50. 9. Wattwil M, Tho& T, Hennerdal S, Garvill, JE. Epidural analgesia with bupivicaine reduces postoperative paralytic ileas after hysterectomy. Anesth Analg 1989;68:353-8.
