Sonographic Features Related to Volvulus in Neonatal Intestinal Malrotation

Sonographic Features Related to Volvulus in Neonatal Intestinal Malrotation Hsun-Chin Chao, MD, Man-Shan Kong, MD, Ju-Yi Chen, MD, Syh-Jae Lin, MD, Jer-Nan Lin, MD This 3 year prospective study evaluated the sensitivity and specificity of abdominal ultrasonography and color Doppler ultrasonography in 31 neonates with suspected malrotation or malrotation with volvulus. Water instillation was used to detect duodenal dilatation, edema, and malrotated bowels. Twenty patients with ultrasonographic characteristics of inversion of the superior mesenteric artery and superior mesenteric vein were later surgically proved to have malrotation. Nine of these 20 patients also had volvulus. Sonographic features suggestive of volvulus included duodenal dilation with tapering configuration (8 of 9 cases, 89%), fixed midline bowel (8 of 9 cases, 89%), whirlpool sign (8 of 9 cases, 89%), and dilation of the distal ABBREVIATIONS SMA, Superior mesenteric artery; SMV, Superior mesenteric vein; UGI, Upper gastrointestinal; CDUS, Color Doppler ultrasonography; CT, Computed tomography; MR, Magnetic resonance Received August 13, 1999, from the Divisions of Gastroenterology (H.-C.C., M.-S.K., S-J.L.), Neonatology (J.-Y.C.), and Surgery (J.-N.L.), Department of Pediatrics, Chang Gung Children s Hospital, Taoyuan, Taiwan. Address correspondence to Syh-Jae Lin, MD, Department of Pediatrics, Chang Gung Children s Hospital, 5-Fu Hsing Street, Kwei-Shan 333, Taoyuan, Taiwan. Address reprint requests to Hsun-Chin Chao, MD, Division of Gastroenterology, Department of Pediatrics, Chang Gung Children s Hospital, 5-Fu Hsing Street, Kwei-Shan 333, Taoyuan, Taiwan. superior mesenteric vein (5 of 5 cases, 100%). The sensitivity and specificity of duodenal dilation with tapering configuration for detecting volvulus were 89% and 92%, respectively; of fixed midline bowel, 89% and 92%; of whirlpool sign, 89% and 92%; and of dilation of distal superior mesenteric vein, 56% and 73%. The results of this study indicate that ultrasonographic features of inversion of the superior mesenteric artery and superior mesenteric vein could aid in the diagnosis of malrotation, and certain sonographic features can also be used to evaluate volvulus, a condition requiring emergent operation. KEY WORDS: Malrotation; Volvulus; Bowel, malrotation; Neonate; Superior mesenteric artery, inversion. Malrotation represents an incomplete rotation of the intestine during fetal development. Because intestinal malrotation can become a life-threatening emergency if the common complications of obstruction by Ladd bands or midgut volvulus are not detected and treated promptly, early diagnosis of this disorder is important. 1 Bilious vomiting in a neonate should always be dealt with cautiously and urgent investigations undertaken to establish the cause of vomiting. 2 Conventionally, the diagnosis of malrotation is made by a UGI series showing an abnormal position of the duodenojejunal junction, usually to the right of the midline. Barium enema may also show an abnormal position of the cecum. However, the results of radiologic studies are not always conclusive. 3,4 As in previous studies, the mortality of volvulus associated with malrotation could be significantly decreased if an early diagnosis was made. This study examined the sensitivity and specificity of ultra- 2000 by the American Institute of Ultrasound in Medicine J Ultrasound Med 19:371 376, 2000 0278-4297/00/$3.50

372 INTESTINAL MALROTATION AND VOLVULUS J Ultrasound Med 19:371 376, 2000 sonography in detecting intestinal malrotation or volvulus. The surgical findings were correlated with the sonographic features of the patients to determine the accuracy of the sonographic diagnosis. The value of ultrasonographic features in predicting volvulus was also evaluated. The predictive features may indicate the need of surgical intervention. MATERIALS AND METHODS A 3 year prospective study of the ultrasonographic diagnosis of intestinal malrotation was undertaken from July of 1996 to June of 1999. A total of 31 neonates with clinical symptoms suggestive of intestinal malrotation were recruited into this study. Plain abdominal radiography, sonography, and UGI barium examinations were performed immediately when the patients had symptoms suggestive of malrotation. The results of sonographic studies, radiologic examinations, and operative findings of the patients were reviewed. The double bubble appearance with sparse distal air was used as a characteristic finding on plain abdominal radiographs for the diagnosis of malrotation. Ultrasonographic characteristics of inversion of the SMA and SMV were used to diagnose malrotation in each patient, and the sonographic diagnosis of malrotation was compared with the results of a UGI barium study. Malrotation was diagnosed by UGI barium examination when an abnormal configuration of the duodenal C-loop was found, or when the duodenojejunal flexures were located to the right of the vertebral bodies. The sonographic study was performed by a gastroenterologist (H.-C.C.) using an Acuson TX 128 machine (Acuson, Mountain View, CA) with a 5 MHz curved array transducer for children and a 7 MHz curved array transducer for newborn infants. All of the patients were examined in the supine posture. Because air may decrease the sensitivity of the sonographic demonstration of the duodenum and other areas of the bowel, water instillation was used to help in evaluating the dilation, duodenal edema, and malrotated bowel. We routinely use 50 ml distilled water for instillation via a nasogastric tube in each ultrasonographic examination. The anteroposterior diameter of the proximal SMV (the segment of the SMV above the umbilical area) and the distal SMV (the segment of the SMV below the umbilical area) were measured. The distal SMV was considered dilated if its diameter was greater than that of the proximal SMV. CDUS was used to confirm the SMA with arterial flow and SMV with venous flow and to detect the presence of whirlpool sign. The CDUS criterion for whirlpool sign showed the SMV winding around the SMA. Color gain was adjusted dynamically during examination to maximize the visualization of blood flow. Precautions were taken to avoid misinterpretation of vascular images caused by color noise and motion artifacts induced by patient movements. In addition to diagnosing malrotation by ultrasonography, we also evaluated specific sonographic and CDUS features that can be used to predict the presence of volvulus. The sonographic and barium studies were correlated with the operative findings. RESULTS A total of 20 patients (11 male, 9 female) had ultrasonographic diagnoses of malrotation, which were subsequently proved by surgery. Nine cases were complicated by volvulus. All patients had bilious vomiting, and most of them (80%) had clinical presentation during the first 2 weeks of life. Table 1 lists the ultrasonographic diagnosis, findings on plain abdominal radiographs and UGI barium studies, and surgical correlation of the subjects. Plain abdominal radiographs showed a double bubble appearance in 18 patients. Among these, 3 patients had duodenal atresia without distal sparse air, and 15 patients did have distal sparse air, including 13 with malrotation (9 complicated by volvulus) and 2 with duodenal web. A positive UGI barium finding Table 1: Sonographic, Radiographic, and Surgical Correlation in 31 Newborn Infants with Suspected Malrotation Surgically Negative Diagnosis of Proved Cases of cases Malrotation Malrotation (n = 20) (n=11) Sonographic diagnosis Inversion of SMA and 20 0 SMV (n = 20) Plain abdominal radiographs Double bubble sign With sparse distal air* 13* 2* Without distal air 0 3 Nonspecific 4 9 UGI barium diagnosis Positive findings (n = 18) 18 0 Negative findings (n = 13) 2 11 *The double bubble sign with sparse distal air was considered a characteristic feature for malrotation. Among the 15 patients, 13 had malrotation (9 cases complicated by volvulus), and 2 had duodenal web.

J Ultrasound Med 19:371 376, 2000 CHAO ET AL 373 (abnormal duodenal C-loop or the duodenojejunal flexure located to right of the vertebral bodies) for malrotation was found in 18 cases (all surgically proved); the remaining 2 cases were found to have normal duodenal C-loops, in which the location of duodenojejunal flexures could not be clearly demonstrated. The ultrasonographic and CDUS features in the patients with intestinal malrotation are summarized in Table 2. Inversions of the SMA and SMV were found in 20 patients, and all of them were surgically proved to have malrotation. Associated sonographic features in these patients included duodenal dilatation with tapering configuration (Fig. 1) in 9 cases, thickness of duodenal wall greater than 2 mm in 9 cases, fixed midline bowel (Fig. 2) in 9 cases, whirlpool sign (Fig. 3) in 8 cases, dilation of the distal SMV in 5 cases (Fig. 4), and ascites in 5 cases. The duodenal dilation and thickness of the duodenal wall were measured after water instillation. The instances of fixed midline bowel were found close to the duodenum, and water was visualized passing through duodenum to the midline fixed bowel (Fig 2). The fixed midline bowel was considered jejunum, since water was visualized passing through the duodenum into these midline bowel segments. Table 3 shows the results of analysis of sonographic features related to volvulus. The statistical analysis was done using chi-square test. A P value lower than 0.01 was considered statistically significant. Sonographic features of the nine patients with volvulus showed duodenal dilation with tapering configuration, 8 cases (P = 0.002); thickness of duodenal wall greater than 2 mm, 6 cases (P = 0.143); fixed midline bowel, 8 cases (P = 0.002); whirlpool sign, 8 cases (P = 0.002); dilation of the distal SMV, 5 cases (P = 0.02); and ascites, 3 cases (P = 0.795). The sensitivity, specificity, positive predictive value, and negative predictive value of these sonographic features for detecting volvulus in malrotation are summarized in Table 4. Duodenal dilation with distal tapering, fixed midline bowel, and whirlpool sign were sensitive and specific for detecting volvulus (sensitivity 89%, specificity 92%), and had a high positive predictive value for volvulus (89%). Dilation of distal SMV showed 100% positive predictive value, 56% (5 of 9) sensitivity, and 73% (11 of 15) specificity for detecting the presence of volvulus. The sensitivity, specificity, and positive predictive value of duodenal wall thickness greater than 2 mm were 66%, 75%, and 66%, respectively. Presence of ascites showed only 33% sensitivity and 60% positive predictive value, but an 82% specificity. No bowel gangrene was found at operation or during follow-up evaluation in these 20 patients. The remaining 11 neonates who did not have ultrasonographic features of malrotation were eventually proved to have other clinical conditions, including duodenal atresia (3 cases), duodenal web (2 cases), and functional ileus (6 cases). DISCUSSION Early diagnosis of intestinal malrotation is a major diagnostic challenge to clinicians and is imperative to prevent the catastrophic complication of midgut volvulus and bowel necrosis. Plain abdominal radio- Table 2: Ultrasonographic and CDUS Features of Malrotation in 20 Newborn Infants Figure 1 Water instillation test helps to evaluate duodenal (D) configuration. Duodenal dilation with tapering (arrows) is a feature relating to volvulus in neonatal intestinal malrotation. Sonographic and CDUS Features Number of Patients Position of SMA and SMV Inversion of the SMA and SMV 20 Associated features Duodenal dilation with distal tapering* 9 Duodenal wall over 2 mm* 6 Fixed midline bowel 9 Ascites 5 Whirlpool sign 9 Dilation of distal SMV 5 *Duodenal configuration and edema are demonstrated by the aid of water instillation, and the thickness of duodenal wall was measured when the duodenum was filled with water. CDUS helped demonstrate whirlpool sign and dilation of the SMV.

374 INTESTINAL MALROTATION AND VOLVULUS J Ultrasound Med 19:371 376, 2000 Figure 2 The fixed midline bowel loops are close to duodenum (D) and are highly suggestive of jejunum (J), since little water passed through the duodenum into the midline bowel loops. Figure 3 Whirlpool sign, with the SMV (arrowheads) winding around the SMA, is a sonographic feature associated with volvulus in cases of intestinal malrotation. graphs provide essential information but usually are nonspecific. Some neonates with malrotation were reported to have a normal gas pattern, thus delaying the diagnosis. 5,6 Volvulus may be recognized by the presence of double bubble sign (air collection in the stomach and duodenum) while the rest of the abdomen is fairly gasless. However, our results showed that the double bubble appearance with sparse distal air was found only in 65% (13 of 20 patients) of patients with malrotation, and the predictive rate of this characteristic radiographic finding for volvulus is only 62% (8 of 13 patients). In addition, two patients with duodenal web had similar radiologic pictures. UGI contrast examination is the preferred diagnostic method in cases of symptomatic malrotation. 7 11 In addition, cholescintigraphy has been used to diagnose malrotation by the demonstration of malrotated bowel, 12,13 but this method is more time consuming than barium examination. CT, 14 16 MR imaging, 17 and sonographic examination 17 21 of abnormal positions of the SMA and SMV have all been used to diagnose intestinal malrotation with good sensitivity. Nichols and Li postulate that patients with malrotation have had a failure of counterclockwise rotation of the SMV around the SMA, which should have occurred during normal embryologic bowel development. 14 An abnormal position of the SMA and SMV has been demonstrated to be reliable in the diagnosis of intestinal malrotation. 3,4,15 22 Our results confirm this observation and show that the inversion of the SMA and SMV was 100% sensitive in detecting neonatal intestinal malrotation. Barium UGI studies had been unable to accurately predict which patients with malrotaion are at risk for midgut volvulus. 19 Ultrasonographic features for early detection of volvulus in patients with malrotation have been reported previously. 20 The abnormal position of the duodenum and SMA, with a whorllike pattern around the SMA, was suggestive of volvulus. Our study confirms previous reports that the whirlpool sign (SMV winding around the SMA) demonstrated sonographically is sensitive in detecting volvulus. We also found that duodenal dilation with distal tapering configuration, duodenal edema, fixed midline bowel, and dilation of the SMV were other significant predictors of volvulus with high sensitivity and specificity. Figure 4 Dilation of the distal SMV was found in only 5 of 9 patients with volvulus, but this feature is very specific in the detection of volvulus in cases of malrotation.

J Ultrasound Med 19:371 376, 2000 CHAO ET AL 375 Table 3: Analysis of Sonographic Features Relating to Volvulus in Children with Malrotation Surgically Proved Volvulus Ultrasound Features Positive Negative P Value in Detecting Volvulus Duodenal dilation with distal tapering 0.002* Positive 8 1 Negative 1 10 Wall of duodenum greater than 2 mm 0.143 Positive 6 3 Negative 3 9 Fixed midline bowel 0.002* Positive 8 1 Negative 1 10 Whirlpool sign 0.002* Positive 8 1 Negative 1 10 Distal SMV dilation 0.02 Positive 5 0 Negative 4 11 Ascites 0.795 Positive 3 2 Negative 6 9 *P < 0.01: marked significance of the chi-square analysis. Water instillation appeared to be helpful in visualizing the superior mesenteric vessels, the level of bowel obstruction, duodenal configuration, and fixed midline bowel (malrotated bowel). With the aid of water instillation into the stomach and duodenum, these additional sonographic features of malrotation could be demonstrated more clearly. In conclusion, ultrasonography provides good diagnostic results in neonatal intestinal malrotation. Specific sonographic features relating to volvulus should be evaluated as potential indicators of the need for an emergent operation. REFERENCES 1. Weinberger E, Winters WD, Liddell RM, et al: Sonographic diagnosis of intestinal malrotation in infants: Importance of the relative positions of the superior mesenteric vein and artery. AJR 159:825, 1992 2. Yanez R, Spitz L: Intestinal malrotation presenting outside the neonatal period. Arch Dis Child 61:682, 1986 3. Zerin JM, DiPietro MA: Superior mesenteric vascular anatomy at US in patients with surgically proved malrotation of the midgut. Radiology 183:693, 1992 4. Torres AM, Ziegler MM: Malrotation of the intestine. World J Surg 17: 326, 1993 5. Long FR, Kramer SS, Markowitz RI, et al: Radiographic patterns of intestinal malrotation in children. Radio- Graphics 16:547, 1996 Table 4 Statistical Data of Volvulus Detection: Sonographic Features and Surgical Correlation Sensitivity (%) Specificity (%) PPV (%) NPV (%) Duodenal dilation with distal tapering 89 92 89 11 Duodenal wall thickness greater than 2 mm 66 75 66 75 Fixed midline bowel 89 92 89 11 Whirlpool sign 86 92 89 11 Distal SMV engorgement 56 73 100 0 Ascites 33 82 60 60 PPV, Positive predictive value; NPV, Negative predictive value. Data are included for patients with volvulus.

376 INTESTINAL MALROTATION AND VOLVULUS J Ultrasound Med 19:371 376, 2000 6. Lin JN, Lou CC, Wang KL: Intestinal malrotation and midgut volvulus: A 15-year review. J Formos Med Assoc 94:178, 1995 7. Ford EG, Senac MO Jr: Malrotation of the intestine in children. Ann Surg 215:172, 1992 8. Torres AM, Ziegler MM: Malrotation of the intestine. World J Surg 17:326, 1993 9. Millar AJW, Rode H, Brown RA, et al: The deadly vomit: Malrotation and midgut volvulus; a review of 137 cases. Pediatr Surg Int 2:172, 1987 10. Simpson AJ, Leonidas JC, Krasna IH, et at: Roentgen diagnosis of midgut malrotation: Value of upper gastrointestinal radiographic study. J Pediatr Surg 7:243, 1972 11. Powell DM, Othersen HB, Smith CD: Malrotation of the intestines in children: The effect of age on presentation and therapy. J Pediatr Surg 24:777, 1989 12. Glowniak JV: Intestinal malrotation diagnosed by cholescintigraphy. Clin Nucl Med 13:835, 1988 13. Elliott MW, Williamson MR, Davis M, et al: Simultaneous diagnosis of acute cholecystitis, intestinal malrotation, and duodenal diverticulum by cholescintigraphy. Clin Nucl Med 18:355, 1993 14. Nichols DM, Li DK: Superior mesenteric vein rotation: A CT sign of midgut malrotation. AJR 141:707, 1983 15. Fisher JK: Computed tomographic diagnosis of volvulus in intestinal malrotation. AJR 141: 707, 1981 16. Mori H, Hayashi K, Futagawa S, et al: Vascular compromise in chronic volvulus with midgut malrotation. Pediatr Radiol 17:277, 1987 17. Gaines P, Saunders A, Drake D: Midgut malrotation diagnosed by ultrasound. Clin Radiol 38:51, 1987 18. Shatzkes D, Gordon DH, Haller JO, et al: Malrotation of the bowel: Malalignment of the superior mesenteric artery-vein complex shown by CT and MR. J Comput Assist Tomogr 14:93, 1990 19. Dufour D, Delaet MH, Dassonville M, et al: Midgut malrotation, the reliability of sonographic diagnosis. Pediatr Radiol 22:21, 1992 20. Pacros JP, Sann L, Genin G, et al: Ultrasound diagnosis of midgut volvulus: The whirlpool sign. Pediatr Radiol 22:18, 1992 21. Zerin JM, DiPietro MA: Mesenteric vascular anatomy at CT: Normal and abnormal appearances. Radiology 179:739, 1991 22. Loyer E, Dunne Eggli K: Sonographic evaluation of superior mesenteric vascular relationship in malrotation. Pediatr Radiol 19:173, 1989