Introduction

The superior mesenteric artery syndrome (SMAS), also known as “Wilkie syndrome”, is a complication of scoliosis surgery [1,2,3]. This clinical condition was first illustrated by Carl von Rokitasky in 1861 and later in greater detail by Wilkie in 1927 [4,5,6,7]. The reported incidence of SMAS after scoliosis surgery is approximately 0.013–4.7% [8,9,10,11]. The aetiology of SMAS is a change of the anatomical relationship between the third part of the duodenum, lumbar spine, and superior mesenteric artery (SMA) [12, 13]. The SMA arises from the anterior aspect of the aorta at the level of L1–L2 vertebral bodies [14, 15]. At its origin, the SMA is encased in fat and lymphatic tissue and descends downwards the mesentery [8, 11, 16]. On the other hand, the duodenum traverses the aorta at the level of the L3 vertebral body and is suspended by the ligament of Treitz between the aorta and the SMA [8, 16, 17]. Any factor which modifies this anatomical relationship could lead to an extrinsic compression of the duodenum [11, 17,18,19]. SMA syndrome presents clinically unspecific simulating other factors and causes of upper gastrointestinal obstruction [4,5,6,7]. The correct diagnosis of SMAS may be challenging, requiring a proper correlation of clinical signs and symptoms with radiographic findings [4,5,6,7]. A timely diagnosis and appropriate management of SMAS is recommended since delayed treatment has a mortality rate of 33% [1, 6, 20]. Its management is conservative initially, with the rationale of weight gain to increase the retro-peritoneal fat and the aortomesenteric angle [13, 21, 22]. If the conservative therapy fails, surgical options are represented by gastrojejunostomy, duodenojejunostomy, or Ladd procedure [13, 21, 22].

Evidence on SMAS is limited. A proper therapeutic algorithm and internationally accepted guidelines are missing. This investigation systematically reviewed current evidence on pathogenesis, risk factors, management, and outcomes of SMAS following correction spine surgery for adolescent idiopathic scoliosis (AIS).

Methods

Eligibility criteria

All the clinical studies investigating SMAS following spine surgery for AIS were accessed. According to the author´s language capabilities, English, German, Italian, and Spanish articles were eligible. According to the Oxford Centre of Evidence-Based Medicine, we included only studies with levels I to IV of evidence [23]. Studies published in grey literature or without full text were not eligible. Opinions, letters, and editorials were not considered. Only clinical studies were eligible. Missing quantitative data under the outcomes of interests warranted the exclusion of the study.

Search strategy

This study was conducted according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses: the 2020 PRISMA statement [24]. The following framework was used to guide the literature search:

  • P (Problem): SMAS following AIS;

  • I (Intervention): scoliosis correction surgery;

  • C (Comparison): height, weight, length of fusion as risk factor for SMAS;

  • T (Time): interval between the surgical treatment and presenting the symptoms for SMAS.

In December 2023, the following databases were accessed: PubMed, Web of Science, and Google Scholar. No time constraint was set for the search. The medical subject headings (MeSH) used in each database for the search are reported in the appendix. Additional filters were not used for the database search.

Selection and data collection

Two authors (G.P. and L.S.) independently executed the database search. All matched titles were screened by hand, and the abstract was examined if suitable. The full text of the abstracts matched the topic and was accessed. Studies without accessible or available full text were not included. A cross-reference of the bibliography of the full-text manuscripts was also performed for the inclusion. Authors debated controversies. A third senior author (F.M.) took the final decision in case of additional differences of opinion.

Data items

Two authors (G.P. and L.S.) separately performed data examination and analysis. The following data at baseline were extracted: author, year and journal of publication, study design, and patient characteristics, including age, gender, weight, height, and BMI. The post-operative day (POD) when symptoms of SMAS started was recorded. Data regarding the type of surgical scoliosis correction, signs and symptoms, diagnosis and management of SMAS were retrieved. Data were extracted in Microsoft Office Excel version 16.72 (Microsoft Corporation, Redmond, USA).

Assessment of the risk of bias

The risk of bias was examined following the guidelines in the Cochrane Handbook for Systematic Reviews of Interventions [25]. The Nonrandomised Studies of Interventions (ROBINS-I) tool was used for all included non-RCTs [26]. Seven domains of potential bias in non-RCTs were assessed. Possible confounding and the nature of patient selection before the comparative intervention are evaluated by two domains. A further domain is used to evaluate bias in the classification during the intervention. The final four domains assess the methodological quality after the intervention comparison has been implemented and relate to deviations from previously intended interventions, missing data, erroneous measurement of outcomes, and bias in the selection of reported outcomes. The figure of the ROBINS-I was elaborated using the Robvis Software (Risk-of-bias VISualization, Riskofbias.info, Bristol, UK) [27]. Case reports included in this investigation were evaluated using the Joanna Briggs Institute (JBI) criteria appraisal tools for case reports [28]. Possible answers were “yes”, “no”, “unclear”, or “not applicable”. Eight questions were answered to evaluate the methodological quality in accordance with the JBI checklist. The included studies were assessed based on the following bias: low risk of bias: studies with more than six “yes”; moderate risk of bias: studies with five to six “yes”; and high risk of bias: studies with less than five “yes”. Two reviewers (G.P. and L.S.) evaluated the risk of bias in the extracted studies separately. Disagreements were solved by a third senior author (F.M.).

Results

Study selection

The literature search resulted in 276 articles. Of them, 192 were excluded as they were duplicates. Additionally, 44 studies were excluded for the following reasons: study design (N = 7), low level of evidence (N = 6), full text not available or not in peer-reviewed journals (N = 16), and language limitations (N = 4). An additional 11 studies were excluded because they did not offer quantitative data on the outcomes of interest. Finally, 29 articles were included in the present investigation. Of them, nine had a retrospective study design, and 20 were case reports. The results of the literature search are visible in Fig. 1.

Fig. 1
figure 1

PRISMA flow chart of the literature search

Full size image

Methodological quality assessment

The ROBINS-I was applied to investigate the risk of bias in 31.0% (9 of 29) of the included studies, as they were non-RCTs. Given the small number of patients studied in the included investigations, the risk of bias based on confounding and the selection of participants was rated in four studies critical because they reported data from only one patient. The residuary studies were rated with a moderate to serious risk of bias in the cited domains. The intervention protocol was well reported in most studies, and no significant deviation from the interventions was identified, showing a primarily low risk of bias in the classification of interventions and deviation from intended interventions. Data were adequately reported in the most included studies, and the measurement of the outcomes was equivalent among the groups. Given the lack of randomisation of the investigators and patients, the measurement of the outcomes was evaluated with a moderate risk of bias in all of the studies. The reported results corresponded to the planned protocol in most included studies. Given the overall poor methodological quality in the included studies, the overall risk of bias was predominantly serious to critical. The assessments of the methodological quality are given in Fig. 2.

Fig. 2
figure 2

ROBINS-I of non-RCTs

Full size image

The risk of bias using the JBI criteria appraisal tool for case reports was evaluated in 69.0% (20 of 29) of the included studies, as they were case reports. While seven of the included studies received a score of six and five, respectively, indicating moderate quality, 12 received a score of seven or eight, demonstrating good-quality evidence. The case report by Fiorini et al. received a score of three out of eight, suggesting low-quality evidence with a high risk of bias. The detailed assessment steps are shown in Fig. 3.

Fig. 3
figure 3

JBI criteria appraisal tool for case reports

Full size image

Study characteristics and results of individual studies

A total of 61 patients were included in the present study. 62% (38 of 61 patients) were women. The mean age was 15.8 ± 7.2 years. The mean weight was 45.3 ± 8.0 kg, the mean height 159.6 ± 13.6 cm, and the mean BMI 16.5 ± 2.9 kg/m2. The generalities and demographics of the included studies are shown in Table 1.

Table 1 Patient generalities of the included studies
Full size table

Synthesis of results

The mean interval between spine surgery and symptoms occurrence was 69.1 ± 262.4 days (3 days to 4 years). The mean duration of the treatment for SMAS was 21.6 ± 10.3 days. Table 2 provides an overview of the main findings of the included studies.

Table 2 Overview of the main findings of the included studies
Full size table

Discussion

The systematic review found that the mean interval between spine surgery and symptoms occurrence was 69 days, with high between-studies variability (3 days to 4 years). The mean duration of the management for a diagnosed SMAS was 21.6 days.

Following spinal surgery in AIS, SMAS is a complication provoked by extrinsic compression of the third part of the duodenum as it crosses between the SMA, which lies anteriorly, and the abdominal aorta and lumbar spine posteriorly [7, 13, 50,51,52]. The SMA supplies blood to the entire small intestine except the duodenal bulb, cecum, ascending, and transverse colon. The vessel arises from the anterior aspect of the aorta at the level of the L1 vertebral body and extends caudally into the small bowel mesentery [53]. The third part of the duodenum passes between the aorta and SMA and is suspended by the ligament of Treitz, also known as the suspensory muscle of the duodenum [54,55,56,57]. The aortomesenteric angle and the aortomesenteric distance usually range from 28° to 65° and 10 to 34 mm, respectively. In SMAS, they are reduced, having values of 6–15° and 2–8 mm, respectively [13, 51, 58,59,60,61]. The third part of the duodenum is generally surrounded by periduodenal fat at the aortomesenteric angle. Modifying this anatomy could lead to vascular compression of the third part of the duodenum, which might lead to SMAS [7, 8, 13, 41, 62]. The risk factors for SMAS can be congenital or acquired, and identifying patients at risk for SMAS is pivotal to avoiding complications. A congenital short ligament of Treitz could pull the duodenum upward into the vascular angle between the SMA and the aorta, causing vascular compression [11, 30]. Acquired causes related to a decrease in intestinal padding include weight loss (weight < 25% percentile, BMI < 25th percentile), eating disorders, cachexia, severely debilitating conditions, and incapacitating illnesses such as malignancies [22, 63,64,65]. A lower BMI, height, or weight than 5%, 50%, or 25% percentile, sagittal kyphosis, and heavy and quick halofemoral traction after anterior spinal release are other important risk factors for SMAS [41]. Acquired and iatrogenic modifications of the aortomesenteric angle due to rapid height gain (> 50% percentile) or postoperative spinal lengthening could also lead to SMAS. Children mostly attend a surgical correction of AIS in their most rapid longitudinal growth phase. Moreover, children or adolescents with scoliosis and thoracic curves also have hyperkyphosis, which leads to a more extended spine and reduces the aortomesenteric angle [8, 11, 18, 19, 29, 30, 41]. Willner et al. [66] showed that a spurt of growth is particularly accelerated in the year before the diagnosis of AIS, and these patients tend to be taller and more slender than their age-matched counterparts. Moreover, there may be an acute increase in vertebral column length after posterior and anterior spinal instrumentation [67, 68]. The traction on the SMA and, consequently, narrowing of the aortomesenteric angle can be caused by the acute increase in vertebral column length following spinal instrumentation. Finally, postoperative weight loss causes retroperitoneal fat reduction that protects the duodenum from compression [8, 11, 18, 19, 29, 30, 41]. During the retroperitoneal dissection in the anterior approach to the thoracolumbar spine, the disruption of the autonomic nerve supply to the small intestine may also trigger the development of the condition [19]. The mean interval between spine surgery and the symptoms of SMAS was 69 days, with high variability between studies (3 days to 4 years). Prompt identification of risk factors and an early diagnosis are necessary to manage SMAS and reduce the risk of complications. Delay in the development of the syndrome might be related to the progressive weight loss occurring in the postoperative period, resulting in gradual loss of the retroperitoneal fat [42, 69, 70]. Symptoms include nausea, anorexia, bilious vomiting, and intestinal constipation, with partial relief of them with postural changes and epigastric pain [71,72,73,74]. During the physical examination, the peristalsis can be normal or hyperkinetic, and the abdomen is soft with occasional tenderness in the epigastrium at deep palpation [19, 41]. Symptoms of SMAS are similar to paralytic ileus [18, 41, 68, 75]. Recurrent vomiting with gastric dilatation could eventually lead to severe hypovolemia, progressive dehydration, oliguria, and electrolyte disorders, such as hypokalaemia and metabolic alkalosis, or even death [68, 76, 77]. Death in SMAS can result from chemical pneumonia by vomitus inhalation or gastric perforation [8, 44, 78].

SMAS is characterised by three criteria: dilated duodenum, reduction of the aortomesenteric angle to less than 25°, and compression of the third part of the duodenum by the SMA [13, 77, 79,80,81]. Barium imaging can assist in the diagnosis, showing the characteristic duodenal dilatation with an abrupt vertical cutoff in the third part of the duodenum and a 4–6-h delay in the gastroduodenal transit [22, 82]. The gold standard radiological methods in SMAS diagnosis is represented by computer tomography angiography (CTA), which reveals compression of the third part of the duodenum by the SMA with subsequent proximal duodenal dilatation and a reduced aortomesenteric angle and aortomesenteric distance [61, 62, 83, 84]. In most patients, conservative management is associated with improving the symptoms, usually occurring after 2–3 days [85]. Oral intake should be restricted, and under radiographic assistance, nasojejunal feeding should be passed distally from the duodenal obstruction to achieve enteral feeding and progressive weight gain with high-caloric nutritional supplements [85]. If the symptomatology persists or enteral feeding is impossible, total parenteral nutrition should be started.

Failure of the conservative treatment can lead to life-threatening conditions such as metabolic alkalosis, electrolyte imbalance, and aspiration pneumonia [13, 20, 86, 87].

If conservative management fails, gastrojejunostomy, duodenojejunostomy, or Ladd procedure is advocated [88,89,90,91]. Although conservative management is the first therapeutic approach in patients with SMAS, the current evidence lacks large clinical trials and recommendations. An internationally accepted therapeutic algorithm for the management of SMAS is necessary. Given the rarity of SMAS, the current evidence is mostly case reports. Therefore, the generalisation of these results is limited, and an estimate of the prevalence of SMAS is unreliable. Moreover, the limited evidence hinders the establishment of reliable risk factors and a proper algorithm for diagnosing and treating SMAS. Additional investigations are required to establish risk factors and diagnostic criteria.

Conclusion

SMAS is a complication of spinal surgery in AIS caused by extrinsic compression of the third part of the duodenum. The mean interval between spine surgery and symptoms occurrence was 69 days, with high between-studies variability (3 days to 4 years). Prompt identification of risk factors and an early diagnosis are necessary to manage SMAS and reduce the risk of complications. Additional investigations are required to establish risk factors and diagnostic criteria.