Introduction

The midgut is supplied by the superior mesenteric artery (SMA) during normal embryologic development and passes through the herniation, rotation, retraction, and fixation stages. Rotation occurs around the SMA anti-clockwise and 270° in total. If all goes well, the duodenojejunal junction becomes fixed at the left upper quadrant, the cecum at the right lower quadrant, and the midgut widely attached to the retroperitoneum with the oblique mesentery extending from the top left to the bottom right. The midgut is attached to the retroperitoneum with a narrow mesentery and becomes prone to volvulus if rotation and fixation do not take place properly. Correct diagnosis of malrotation has critical importance due to the potential catastrophic course [1].

Although barium contrast study of the upper gastrointestinal system (UGCS) is accepted as the gold standard for the diagnosis of malrotation, it may yield false-negative and false-positive results. Additional disadvantages such as application difficulties and radiation exposure are also of concern. The utility of ultrasonography in the diagnosis of malrotation is still controversial. As it is a noninvasive, easy, and fast imaging modality, the value of color Doppler ultrasonography (CDUS) for the diagnosis of malrotation was investigated in this study.

Materials and methods

Patients who were investigated for presumed malrotation between 2007 and 2014 were evaluated retrospectively. The study was approved by the Institutional Review Board. Age, gender, symptoms, surgery findings, and the results of all imaging studies were recorded. Patients with a history of gastroschisis and omphalocele were excluded. Patients with presumed malrotation were evaluated with CDUS, UGCS after upright plain radiography, and occasionally colonic contrast studies. First, gray-scale Doppler ultrasonography was performed to determine the superior mesenteric artery (SMA) and vein (SMV) by scanning their course from the aorta and portal venous system, respectively. Then, CDUS was performed to obtain the whirlpool sign (WS), which is characteristic for volvulus diagnosis.

CDUS was performed with a linear probe at a frequency of 7.5 MHz (Toshiba Aplio 500). Imaging was performed by a pediatric radiologist with more than 10 years of experience. When the SMV was located at the 12 o’clock position in relation to the SMA during CDUS imaging, it was defined as “vertically located” by the radiologist. When the SMA and SMV relation appeared normal at the proximal segment but disturbed at the distal part, it was defined as suspicious. Volvulus was diagnosed when the SMV and intestinal mesentery wrapped around the SMA in a clockwise direction like a corkscrew.

As the main objective of this study was to investigate the diagnostic value of the inverted SMA/SMV relationship on ultrasound, patients who could not be evaluated due to gas artifact and patients identified as ‘vertical’ and ‘suspicious’ were excluded from the statistical analysis.

Results

A total of 82 patients (49 males and 33 females) were included in the study. The median age of the patients was 1.2 years (range 1 day–15.6 years). Vomiting was the main complaint in 77 patients, while eight had additional symptoms such as chronic diarrhea and bloody stool. All patients underwent CDUS, and the SMA and SMV relationship was normal in 44% (n = 36) (Fig. 1). It was not possible to evaluate the SMA/SMV relation in 19 patients (23%) due to gas artifact in the upper abdomen. A final diagnosis of malrotation was reached in 15 patients (18%) when the SMV was located on the left side of the SMA (Fig. 2), while the SMV was located at the 12 o’clock position in relation to the SMA in nine patients (11%) (Fig. 3). Three of the patients (4%) were regarded as ‘suspicious for malrotation’.

Fig. 1
figure 1

Normal the SMV is seen in the normal location to the right of the SMA

Full size image
Fig. 2
figure 2

Malrotation the SMA is observed to the right of the SMV

Full size image
Fig. 3
figure 3

The SMV and SMA are aligned at 12 o’clock

Full size image

The sensitivity and specificity of CDUS for the diagnosis of malrotation were 93.8 and 100%, respectively. The positive predictive value was 100% and the negative predictive value was 97.9%.

An upper gastrointestinal barium study was performed in 75 patients. The result was normal in 33 patients (44%), while 12 (16%) were diagnosed with malrotation, eight (11%) had surgical conditions other than malrotation, and 22 (29%) had non-surgical conditions such as gastroesophageal reflux and organoaxial gastric volvulus.

Of the 15 patients diagnosed with malrotation by CDUS, UGCS was also confirmative in nine patients, and the diagnosis was confirmed during surgery in further six patients. Of the 36 patients with normal CDUS imaging, UGCS indicated malrotation in only one patient, and the condition was confirmed during surgery (Table 1).

Table 1 Comprasion of UGCS and CDUS in diagnosis of malrotation
Full size table

A ‘double bubble’ sign and/or decreased gas shadows at distal intestinal segments were encountered on plain abdominal radiography in eight patients. Among these patients, two had duodenal web, two had malrotation, one had reverse rotation, and three had malrotation complicated by midgut volvulus. Air-fluid levels were encountered in two patients, one of whom was ultimately diagnosed with malrotation and the other with Hirschsprung disease. Gasless abdomen was encountered on plain abdominal radiography in one patient found to have volvulus. Direct abdominal radiography was considered to be normal in 71 patients.

A barium colon study was performed in 12 patients. In this group, five studies were consistent with malrotation, five were normal, and two indicated conditions other than malrotation.

Discussion

Intestinal malrotation is characterized by a narrow small intestinal mesentery together with abnormal position and fixation of the intestine. It may be asymptomatic and may be found incidentally, or patients may present with vague abdominal symptoms, non-specific abdominal pain, repeated vomiting attacks, and chronic diarrhea. The extreme and life-threatening presentation is acute ischemic midgut volvulus that emerges suddenly [2, 3]. It is difficult to predict the real incidence due to the patients who are asymptomatic or have vague symptoms. The incidence is estimated to be one per 6000 live births. The incidence may be up to 0.2% in upper gastrointestinal barium studies and 1% in autopsy series [1].

Malrotation has two major types: incomplete rotation and non-rotation. The usual intestinal rotation is arrested during the first 90° in non-rotation, and all the small intestine is located at the right side of the abdomen with the colon on the left side. The mesentery is relatively wider in these patients and the midgut volvulus risk is lower. The usual anti-clockwise rotation of the intestine is interrupted in incomplete rotation at a point between 90° and 270°. The cecum is located higher and close to the midline, a part of the small intestine is located at the left of the midline, and Ladd bands extending from the cecum to the abdominal rear wall compress the duodenum. These patients have a high risk of midgut volvulus due to the significantly narrow small intestinal mesentery [1, 4]. A rare form of malrotation is reverse rotation, in which the duodenum and colon show clockwise rotation according to the axis of the SMA and SMV, unlike the normal state in this anomaly, and the duodenum and jejunum are located in front of these vessels with the transverse colon at the back as a result. Colonic obstruction findings emerge due to vessel compression in this condition [5].

The duodenojejunal junction (DJJ) is usually located on the left side of the vertebral body or its pedicle at the level of the duodenal bulbus or higher; or between L1 and L2 vertebrae on anterior–posterior UGCS images. Proximal jejunal segments are located at the left upper quadrant. The 3rd part of the duodenum is located at the same level as the second part of the duodenum in the lateral view [4]. On upper gastrointestinal contrast series, the duodenojejunal junction fails to cross the midline and lies below the level of the duodenal bulb, and the proximal jejunum is located on the right side of the abdomen in malrotation (Fig. 4). Although UGCS has been accepted as the gold standard in the diagnosis of malrotation for a long time, it can yield false-negative and false-positive results, with the rates being reported as 4–15 and 1–15%, respectively [4, 6, 7]. Upper gastrointestinal system barium studies were performed in 91% of the patients with presumed malrotation in this study (n = 75). Of these patients, 16% (n = 12) were diagnosed with malrotation, 11% with conditions other than malrotation (e.g., annular pancreas, duodenal web), and 29% with gastroesophageal reflux. No specific condition was demonstrated in 44% of the patients. The diagnosis was confirmed during surgery in 11 of the 12 patients reported as malrotation, and intestinal obstruction due to congenital bands without malrotation was found during surgery in one patient. One of the 63 patients diagnosed with a condition other than malrotation was concluded to have malrotation during surgery. The sensitivity and specificity of UGCS for diagnosis of malrotation were 91.7 and 98.4%, respectively. The positive and negative predictive values were 91.7 and 98.4%, respectively.

Fig. 4
figure 4

The duodenojejunal junction (arrowhead) is right of the midline and lies below the level of the duodenal bulb (arrow). The proximal jejunum is located on the right side of the abdomen (small arrows)

Full size image

Although not performed in our study, the position of the duodenum in relation to the SMA is investigated by ultrasonography [4]. According to an investigation by Yousefzadeh, demonstrating the 3rd part of the duodenum (D3) at the retromesenteric area is the safest way to rule out the possibility of malrotation. They state that D3 is always in front of the SMA and intraperitoneal in patients with malrotation. D3 is always located between the aorta and the SMA independent of age in normal subjects, and this rules out malrotation. The retromesenteric position of D3 can be demonstrated with US, CT, or MR. The authors previously used US to show retromesenteric D3 and explained the technique in detail [8]. It was suggested that such ultrasonographic evaluation could be conducted more efficiently by filling the duodenum with water, with the false-negative and false-positive rates reported to be 0% [9]. The false-negative rate was reported to be 2.6% in another study where retromesenteric D3 was evaluated with CT [10].

The SMV located at the right and at the front of the SMA is accepted as the normal position. Therefore, the possibility of malrotation is believed to be ruled out when this relationship is normal or confirmed in the case of an abnormal relationship. However, it has been reported that a normal SMA-SMV relationship does not rule out the possibility of malrotation, with false-negative and false-positive rates of 2–33 and 21%, respectively [4, 11].

All 82 patients in our study underwent CDUS. A conclusion could not be made in 19 (23%) patients due to gas artifact. The SMA-SMV relationship was normal in 36 patients (44%), the SMV was located at the left of the SMA in 15 patients (18%), it was vertically located in relation to the SMA in nine patients (11%), and a suspicious SMA–SMV relationship was seen in three patients (4%). The accuracy or inaccuracy of the diagnosis was cross-checked with UGCS or intraoperatively in all of the patients who underwent CDUS (Table 1). Malrotation was detected with UGCS in only one of the 36 patients whose SMA–SMV relationship was reported to be normal on CDUS, and the diagnosis was also verified during surgery. The diagnosis was verified intraoperatively in all 15 patients whose SMV was observed at the left of the SMA and who had been diagnosed with malrotation. In a systematic review by Graziano et al., it was emphasized that the diagnosis should be confirmed by UGCS after a diagnosis of malrotation with ultrasonography, and also that ultrasound cannot precisely exclude malrotation due to false-negative results [12]. In our study, it was also shown that the method could yield false-negative results. For this reason, it was concluded that malrotation should be excluded with UGCS, especially in patients with symptoms of malrotation such as vomiting and chronic diarrhea. Meanwhile, there were no false-positive results in our study as all patients with inversion of the SMA and SMV on CDUS were confirmed to have malrotation during surgery. Accordingly, there is no need to confirm the diagnosis with UGCS after a diagnosis of malrotation with CDUS.

Reversal of the SMA and SMV relationship is the ultrasonic criterion of malrotation. However, two conditions were encountered that did not fully comply with this criterion and did not show a normal anatomic relationship in our study. No malrotation was detected in any of the 12 patients in the “vertical group” and “suspected group”. To the best of our knowledge, there are no sufficient data to clarify or classify these conditions in the English literature. Although the number of subjects was limited in our study, we believe that the SMV located at the 12 o’clock position compared to the SMA or a normal SMV at its origin that is disturbed at a distal site does not indicate a diagnosis of malrotation.

In our study, malrotation suggested by CDUS was confirmed with barium colon study in one of the seven patients in whom UGCS was not used, and this patient ultimately underwent surgery, during which malrotation was confirmed. The other six patients with a specific diagnosis of volvulus based on CDUS underwent surgical correction without further radiological evaluation. The whirlpool sign (WS) detected by CDUS is a well-defined finding that indicates a diagnosis of volvulus. The value of CDUS has been reported to be higher in cases of malrotation with volvulus than in cases of isolated malrotation [13, 14]. Appearance of a whirlpool sign, which indicates volvulus and represents the wrapping of the SMV around the SMA, was found in six patients in our series with CDUS. During surgery, we found malrotation without volvulus in two cases and malrotation with volvulus that did not demonstrate acute ischemic findings in four cases. The patients underwent detorsion and standard Ladd’s procedure.

Colonic barium studies have been employed less often recently for the diagnosis of malrotation. Colonic barium studies were employed in 12 patients in our series and were normal in seven patients. None of the patients with normal colonic barium studies had malrotation. Among the patients with colonic barium studies suggesting malrotation, malrotation was confirmed in four patients and no malrotation was found in one patient.

Erect abdominal radiographs were obtained in all patients, with no pathology found in 71 patients, but 10 of them were diagnosed with malrotation. While a surgical condition other than malrotation was found in two of eight patients with a double bubble sign on plain radiography, malrotation with volvulus was confirmed in three patients and malrotation alone in three patients. Malrotation with volvulus was encountered in a patient with the appearance of gasless abdomen on plain radiography. As many patients with malrotation had normal findings on plain abdominal radiography, one should not rule out the possibility of malrotation based on abdominal radiography. On the other hand, identification of double bubble or gasless abdomen signs on direct abdominal radiography should ‘ring the bells’ for the possibility of midgut volvulus.

We have changed our approach to the management of these patients in our clinic following this study. UGCS has been accepted as the gold standard for the diagnosis of malrotation in these patients for many years, and accordingly we used UGCS in most patients and colonic barium studies in a few patients (other than the six patients who had volvulus findings on CDUS) to confirm the diagnosis. The fact that the diagnosis was confirmed in all patients where a malrotation diagnosis had been made with CDUS created a tendency to use CDUS as the only imaging modality in these patients. However, we continue to use UGCS in cases with ‘vertical’ or ‘suspicious’ findings on CDUS to avoid false-negative results. Exclusion or confirmation of malrotation with UGCS in patients with normal CDUS findings depends on the personal preference of the attending physician.

Conclusion

Correct and timely recognition of malrotation is of critical importance as it can result in serious conditions such as midgut volvulus. Although UGCS has been accepted as the gold standard in the diagnosis of malrotation for many years, this method has a significant rate of false-negative results according to the recent literature and the results presented in this study.

The sensitivity of CDUS and UGCS in diagnosis of malrotation was found to be 91.7 and 93.8%, respectively, in the current study. However, gas artifacts may reduce the quality of imaging as 31 cases could not be evaluated due to gas artifacts on CDUS and were reported as suspicious in our study. Prospective studies including other diagnostic methods such as the determination of retromesenteric D3 are needed to determine the most proper method.