CT facilitates improved diagnosis of adult intestinal malrotation: a 7-year retrospective study based on 332 cases

Intestinal malrotation is a congenital anomaly, often considered a pediatric disorder since most patients are diagnosed within the first year of life [1]. Definitive diagnosis can be made by typical symptoms of bilious vomiting in combination with an abnormal upper gastrointestinal (UGI) series [2‐4]. In adult patients, however, the diagnosis is often incidental. Because of atypical symptoms, these patients are often initially diagnosed with other diseases [5] with further imaging studies or surgery revealing intestinal abnormalities. In fact, the number of adult malrotation patients is far underestimated. Nehra et al. [6] found adults accounting for 48% (82/170) of malrotation cases with 17% of these asymptomatic. However, being asymptomatic at a given point in time does not guarantee a patient will remain asymptomatic for life. In some cases, coexisting diseases or changes in physical condition could trigger a series of complications, including life-threatening volvulus [7, 8]. There are a lack of research on asymptomatic patients and still no clear indication which patients may develop dangerous complications in the future requiring surgery [9, 10].

The UGI series remains the current gold standard for the diagnosis of intestinal malrotation and is widely used in pediatrics [11]. In addition, ultrasound (US) is also recommended for the screening of malrotation with volvulus [12]. Some studies have been conducted in pediatric patients using UGI or US [13‐16]. However, in adult patients, computed tomography (CT) scans, especially with intravenous and oral contrast, have greater diagnostic value, and some studies recommend CT as the first choice in adult patients with suspected malrotation [5, 8, 17]. CT not only displays abnormal signs visible on UGI and US, but also avoids the influence of intestinal gas, enabling acquisition of complete anatomical information regarding the bowel. There have also been studies using CT to classify intestinal variation in adult patients [18, 19], but these included fewer than 20 patients. In addition, the literature on adult intestinal malrotation mostly consists of case reports.

In order to improve the awareness of radiologists and clinicians about adult intestinal malrotation, in the current study, records from all adult patients diagnosed with intestinal malrotation at our institution were reviewed since 2013 and analysis of relevant clinical data as well as CT images performed.

A final 332 eligible adult patients with intestinal malrotation were included in the study. Since 188 of these patients were incidentally identified and did not exhibit symptoms related to intestinal malrotation or receive relevant treatment, they were included in the asymptomatic group, and the course of their treatment was not further documented. Of the 144 patients whose symptoms were recorded, 8 patients had comorbidities that may be associated with symptoms (gastroptosis, liver cysts, irritable bowel syndrome, ileocecal tumors, rectal cancer). Since the possibility that the symptoms were caused by malrotation could not be ruled out, these symptoms were still recorded. In addition, treatment information was not available for 53 outpatients. Therefore, detailed clinical and treatment information was available for 91 patients, who were categorized into surgical and conservative management groups for further analysis (Additional file 1: Fig. S1).

In this study, intestinal variations of malrotation were categorized into ten groups based on three common anatomical landmarks (duodenum, jejunum, and cecum) on CT images (Fig. 2), a classification scheme easily adaptable to clinical practice and easily understandable by clinicians. Furthermore, CT results from these patients were further analyzed with respect to their clinical manifestations and treatments received in an effort to facilitate identification of high-risk patients by their CT presentation, thus enabling individualized treatment and follow-ups. To our knowledge, this is the largest clinical and radiological study of adult intestinal malrotation to date.

Compared with infants and children, the existence of adult intestinal malrotation has been historically ignored due to the lack of characteristic clinical features and is often described as an unexpected discovery in the literature [19, 20]. However, the current study confirms that malrotation is not only a condition to be recognized by pediatricians, but also need to be taken into account by gastroenterologists and surgeons when diagnosing adult patients. In 332 adult patients, this abnormality occurred in any age (mean age 51.0 ± 15.3 years, range 18–97) and was twice as common in men (male/female: 1.91:1) a finding significantly different from previous studies (Durkin et al. 0.43:1 [5], Nehra et al. 0.78:1 [6], and Anand et al. 1:1 [4]).

Imaging studies of intestinal malrotation are infrequent, and studies with large sample sizes are even rarer. Long et al. [13] performed a typology of intestinal malrotation by analyzing the intestinal distribution on pediatric UGI series, after which Fay et al. [21] utilized radiation-free MRI for further study of malrotation in children, confirming its diagnostic reliability. However, the value of CT has traditionally been underestimated. Although the current gold standard for the diagnosis remains UGI series, a number of authors including Zissin et al. [19] have highlighted the unique advantages of abdominal CT in the diagnosis of malrotation in adults. In Yang et al. [18], CT manifestations associated with an increased incidence of complications were described; however, that work only included 14 patients. Based on the analysis of 332 patients in the present study, the advantages of CT in detection of asymptomatic malrotation are further confirmed. Moreover, the D_NJ_RC_R subtype (1a–b) was most common in patients undergoing surgery, whereas the D_YJ_RC_R subtype (4a–b) was most common in those managed conservatively. These findings may help clinicians detect more insidious forms of malrotation, guiding subsequent treatment and follow-up.

The current findings suggest that the presence of the jejunum in the right abdomen is a consistent feature of intestinal malrotation, present in 98.2% (326/332) of patients. Cecal ectopia was found in 12% (40/332) of patients with 29 cases located on the left, 7 in the pelvis, and 4 in the middle of the abdominal cavity, a more common finding than in Yang et al. (1/14) [18]. Cecal ectopia has been described in several case reports [22‐25], and patients may develop atypical abdominal pain in the presence of concurrent appendicitis, requiring prompt identification. Sonomura et al. emphasized the high diagnostic value of coronal CT in acute appendicitis with intestinal malrotation [25] which seems reasonable considering the uncertainty regarding the location of the appendix in patients with malrotation. Although an abnormal SMA/SMV relationship was only observed in 8.4% (28/332) of patients, 20 cases were asymptomatic (Additional file 1: Table S1). Abnormal SMV angulation was observed in 98.7% of patients with an otherwise normal SMA/SMV relationship. Anatomic inversion between SMA and SMV found on CT and US has been proposed as a feature of intestinal malrotation [26], but this abnormal anatomical relationship has been shown to be inadequate in the diagnosis of isolated malrotation (observed in only 28% of patients with malrotation) [16] and can also be observed in normal people [27]. Therefore, in the diagnosis of malrotation, not only the relative position of the SMA and SMV should be noted, but also abnormal angulation of the SMV.

Furthermore, 91 patients with detailed clinical information and treatment history were analyzed. Not surprisingly the incidence of both intestinal obstruction and volvulus were significantly higher in the surgically managed group (p = 0.002 and 0.046, respectively). In addition, although the mean age of the two groups was similar, the conservatively managed group encompassed a wider age range. Older patients in particular were more likely to be treated conservatively, consistent with previous findings [28]. This may be attributable to older patients, in general, being poorer surgical candidates. Interestingly, although half of the patients in the surgical group had a short duration of illness, 33.3% (4/12) also had a duration of illness longer than one month and two of them even longer than one year. In addition patients in the conservative treatment group had more symptoms compared to the surgical group. This could be due to two factors: The conservative group had a larger number of cases, and patients in it generally had a longer duration of disease, which makes them more likely to have various gastrointestinal symptoms. Although further analysis showed that this difference was not statistically significant, such factors should still be taken into account.

Significant differences in CT classification were observed between the two groups. D_YJ_RC_R subtype (4a-b) was present in up to 64.5% of the conservatively managed group but only in 25% of the surgically managed group (p = 0.013). In Yang et al., volvulus occurred more frequently in patients with partial rotation of the duodenum and colon [18], consistent with the current results. However, although six of the seven patients [D_YJ_RC_R (Fig. 2, type 4a–b): 5, D_YJ_RC_M (Fig. 2, type 6): 1, and D_NJ_RC_R (Fig. 2, type 1a): 1] who developed volvulus exhibited duodenal partial rotation, four of them with D_YJ_RC_R subtype had less severe volvulus that improved with conservative treatment. These results suggest that patients with a partially rotated duodenum are less likely to require surgery when the cecum is in a normal position. But the small number of patients with volvulus in both studies limits the strength of these findings. Moreover, although none of the patients with volvulus had an abnormality in the relative position of SMV and SMA, we observed SMV angulation in all of their CT images, and in addition, most of them (6/7) showed the typical “whirlpool sign.” Previous studies have confirmed the sensitivity of the “whirlpool sign” in CT and US images in the diagnosis of volvulus [12, 29], and the current study also supports this.

Another unexpected finding in the current study was that the proportion of asymptomatic patients was as high as 56.6% (188/332), much higher than the 17% (14/82, p < 0.001) in a previous study [6]. Schey et al. [30] have suggested that patients with duodenal nonrotation are more often asymptomatic, which is further confirmed by the radiological study by Zissin et al. [19]. However, in the current study, only 19.1% (36/188) of asymptomatic patients exhibited duodenal nonrotation, and further, such patients tended to have a variety of other intestinal abnormalities (Additional file 1: Table S1). In addition, the “whirlpool sign” was observed in four asymptomatic patients, which mainly showed mild rotation of the vessels and mesentery, without intestinal torsion, explaining their lack of symptoms upon presentation. In fact, many asymptomatic patients diagnosed incidentally can recall bouts of abdominal discomfort potentially related to their malrotation if their past symptoms are thoroughly interrogated [2, 31]. In the present study, the majority of asymptomatic cases, except for some diagnosed incidentally during routine physical examinations, were detected as a result of an examination or treatment for coexisting disease, and even if there were symptoms that might be associated with malrotation, it was difficult to definitely attribute them as such. In another CT study [19], the proportion of asymptomatic patients also reached 94.4% (17/18). However, these patients likewise could have been only temporarily asymptomatic or only had only mild symptoms that were insufficiently severe to report. Yet, these individuals would still potentially be at risk for future volvulus. The incidence of volvulus in the current study was only 2.1% (7/332), much lower than the 12% in Nehra et al. (n = 82, p < 0.001) [6]. The increasing adoption of abdominal CT may improve reader confidence in identifying asymptomatic cases of malrotation, thus reducing the proportion found in the setting of volvulus.

Currently, the Ladd’s procedure is a widely accepted standard treatment for intestinal malrotation, and some studies have emphasized that the Ladd’s procedure should be routinely used in patients with intestinal malrotation, regardless of age [32, 33]. However, in the present study, the rate of surgery was only 13.2% (12/91), much lower than previously reported (Additional file 1: Fig. S3), and of the 12 patients who underwent surgery, only three were treated with Ladd’s procedure (Fig. 6), which seems to suggest an overall inadequate understanding of the disease and its treatment. In fact, although surgery is sometimes necessary, it also should be pursued cautiously. Compared to infants and children, adult patients are more likely to have major postoperative complications and have a higher reoperation rate [5, 34]. In addition, some patients with intestinal necrosis will undergo partial small bowel resection. In this setting, inadequate identification or evaluation of malrotation preoperatively may lead to a more complicated surgery and adverse outcomes. Hsu et al. have reported on short bowel syndrome caused by undiagnosed intestinal malrotation resulting from inappropriate surgical maneuvers and emphasize that CT should be utilized to adequately evaluate each patient preoperatively, especially for procedures with limited field of view, such as laparoscopic and Da Vinci robotic surgery [35]. Given the increasing popularity of minimally invasive surgery, number of asymptomatic patients with malrotation, and various malrotation subtypes, adequate scrutiny of the proximal bowel and mesenteric vessels on preoperative CT is increasingly important.

The current study has several limitations. First of all, although 332 patients were included in this study, detailed clinical data could only be examined in 91. However, this remains the largest study of adult malrotation to date. Secondly, a large number of patients were asymptomatic; although, since this was a retrospective study spanning 7 years and most patients had comorbidities, this was difficult to truly validate. In addition, we must emphasize that the CT classification we proposed is a summary description of common anatomical features and the anatomy of the same subtype may also vary, which requires a specific analysis by the radiologist. Finally, follow-up data could not be reliably obtained, preventing assessment of the effects of various treatments. Long-term follow-up of such patients will be crucial in future studies.

In summary, adult intestinal malrotation is much more common than typically considered. CT not only allows the detection of a greater number of asymptomatic patients but also facilitates diagnostic precision enabling classification of malrotation into subtypes.