Iterative evaluation of novel access techniques for small bowel obstruction: combining image guided, percutaneous, and endoscopic methods

Introduction

Small bowel obstruction (SBO) is a common intestinal affliction leading to approximately 350 000 admissions and $1.3 billion in expenditures annually in the USA. Surgery for SBO accounts for 5%–10% of all abdominal surgeries.1

Traditionally, operative intervention has been advocated for adhesive SBO that does not resolve spontaneously. However, operative management can lead to substantial morbidity in as many as one-third of patients.2 Since prior surgery is the most common cause of adhesive SBO, there is considerable risk of recurrent obstruction when operating for SBO. In a Swedish study of patients undergoing surgery for SBO, 20% had recurrent SBO in their lifetime. Even higher rates of lifetime recurrence are reported elsewhere.3 Surgical management of SBO, in its current form, is based on directly accessing the abdominal cavity, manually untwisting or correcting the blockage or creating a diverting stoma. These time honored concepts, while effective, have not evolved substantially for many decades and are clearly associated with substantial morbidity.4 In other disciplines, major advances in imaging and endoluminal therapies have been recently leveraged to dramatically alter the approach to diseases such as valvular heart disease and aortic aneurysm.5 6 It is timely to consider how similar concepts might be applied to the gastrointestinal tract.

For certain benign colorectal anastomotic strictures, the safety and efficacy of endoscopic multidiameter balloon dilation are well established.7 Moreover, for benign anastomotic strictures after radical gastrectomy for gastric cancer, balloon dilation is considered a primary intervention prior to proceeding with surgical revision.8 In the small bowel, obstructions due to stricturing Crohn’s disease have been successfully treated by double-balloon assisted endoscopic dilation.9 While promising, this approach is technically challenging and often not possible due to the difficulty of accessing the small bowel by retrograde advancement of the endoscope. There is also a substantial risk of recurrence after balloon dilation.

We have previously shown in preclinical trials and a case report that percutaneous access to the intestine with endoscopic clip closure of the access site is possible.10 The current series expands on this work, demonstrating that this approach is not only feasible, but that a direct cut-down of the abdominal wall overlying an upstream loop of bowel leading to the obstructed area is feasible as well.

The use of advanced imaging and contrast injection is a critical part of all these methods, helping to locate and treat the obstruction. It is now also possible to achieve three-dimensional (3-D) reconstructive imaging of the intestines in the modern operating room using fusion techniques of cone beam CT fused to preoperative imaging. This allows for accurate pinpointing of the area of abnormality such as the exact location of an obstruction.

Thus, a new approach to minimally invasive intervention for the relief of SBO is now conceivable with a combination of advanced imaging, imaging fusion software, and a set of novel access techniques. Three variations on a method for access to the small bowel are described: purely endoscopic, endoscopic/percutaneous, and cut-down, along with advanced imaging concepts in the operating theatre. These methods mix endoscopic and open surgical techniques to intervene on the small bowel without the need for extensive surgical adhesiolysis. Regardless of the access technique used, a balloon dilation method on the small bowel for the relief of SBO was then attempted. We report on our initial experience and refinement of these methods, as well as short-term incidences of recurrence and major complications. We hypothesized that these methods would be feasible to relieve SBO in select patients. To our knowledge, this is the first report of successful percutaneous access approaches to the bowel for the relief of SBO.

Methods

We adopted a modified IDEAL methodology for reporting early technique development and refinement (IDEAL stage 1 and early stage 2a). The IDEAL framework is a five-stage conceptual model of surgical innovation development. Adherence to the framework is intended to limit waste in surgical research by improving study design. IDEAL stage 1 interventions are typically proof-of-concept studies. In IDEAL stage 2a, a new idea is refined and iteratively improved until a stable approach is reached. We simultaneously tested three novel approaches using image guidance, with iterative refinement to the technique within each approach. These three methods (purely endoscopic, percutaneous, and direct cut-down) were developed independently of each other but with the common purpose to seek alternatives to standard surgical adhesiolysis or excision. These approaches are used as access methods for lesions of the small bowel. Regardless of the initial approach, once the bowel was accessed, the lesion was then treated with wire-guided balloon dilation or formation of a diverting stoma.

Participant selection

Retrospective case series analysis was performed of all adults undergoing attempted combined endoscopic, radiological imaging, and/or percutaneous procedures for SBO at a single tertiary referral center. Procedures were performed by a single experienced surgeon and endoscopist (JM), in conjunction with an attending interventional radiologist (BP), in most cases.

All patients had SBO, and a range of underlying pathologies was included. Patients were included whether their obstruction was due to adhesions, malignancy or radiation enteritis. SBO due to a pathology typically treated with resection or other abdominal operation was excluded, such as those with SBO as a result of hernia, volvulus, gallstone ileus, etc. Ideal candidates for a novel approach presented with a chronic, unifocal obstruction in a location amenable to one of the access methods, particularly in the distal ileum. Short-segment stenoses are likely to be the most amenable to this approach.

All patients were included from the start of when these procedures began to be offered in October 2020 until September 2021. The risks, uncertain benefits, and established alternatives of these procedures were discussed with all patients, and in instances where we could not use these methods, we ‘converted’ to more traditional surgical methods.

Choice of access technique

One of the three access methods (purely endoscopic, percutaneous or cut-down) was chosen depending on patient factors, and the anatomy was identified through preoperative imaging (figures 1 and 2). Our preference was for a purely endoscopic approach whenever possible. Lesions of the terminal ileum within 20–30 cm of the ileocecal valve are often amenable to this approach (see further). If endoscopy was insufficient to gain access to a strictured area within the distal ileum, we employed percutaneous access for additional assistance. An open cut-down technique to a segment of small bowel adjacent to the obstruction was employed in cases where the target lesion was proximal in the small bowel, if there were concerns for poor wound healing or if there was an anticipated need for proximal intestinal diversion. In all cases, we performed the procedures using a single plane hybrid or fluoroscopic unit (GE Discovery IGS 740, Milwaukee, Wisconsin, USA). Water-soluble contrast (Omnipaque 300) was used endoluminally.

Request permission

Figure 1

Three approaches to SBO treatment: (A) purely endoscopic, (B) percutaneous, and (C) cut-down. DEIP, double-balloon endoluminal intervention platform.

Request permission

Figure 2

Representative intraoperative images. The endoscopic approach: (A) via an ileostomy using DEIP and (B) wire-guided balloon dilation. Percutaneous approach: (C) T-tack placement, (D) vascular sheath introduction, and (E) intraluminal bowel manipulation with 2.8 mm laparoscopic instrument. Cut-down approach: (F) sheath placement after a small incision has been made, (G) fluoroscopy showing a stricture (black arrowhead) and (H) successful dilation (red arrowheads). DEIP, double-balloon endoluminal intervention platform.

Further treatment

If a stenotic area was refractory to balloon dilation attempts, further treatment using standard surgical methods were used to correct the obstruction. The patient provided consent and was prepared for this in the preoperative discussion and planning.

Results

In total, 12 patients were identified who met the inclusion criteria (table 1). The median age of included patients was 68 years (range 42–81 years old); two-thirds were women; and median American Society of Anesthesiologists class was 3. The age and gender of the participants were redacted to maintain confidentiality. The median body mass index was 22.5. Diagnoses leading to SBO included Crohn’s disease, endometrial or colorectal cancer, adhesions from prior surgery and radiation enteritis. Most patients had a history of prior abdominal surgery, and many had a lengthy history of multiple prior SBO episodes. Four patients had a history of abdominal malignancy.

Five cases were performed purely endoscopically, three by percutaneous access and four via cut-down. No major complications occurred, even in patients who failed management via one of the novel methods. Exclusively endoscopic access was used for strictures close to the ileocecal valve, at a prior ileocolic anastomosis or upstream from a stoma site. Percutaneous access alone was performed in cases where strictures were located either close to the ileocecal valve or in the terminal ileum but could not be accessed endoscopically alone. Cut-down access was required for cases in the mid-small bowel. A variety of technologies and techniques were used within the three access techniques (table 2).

Technical success was achieved in 10 of the 12 cases. The causes of technical failure were unique in the two unsuccessful cases. One case (case 3) began with laparoscopic adhesiolysis, at which time a clear stenotic area was not identified. Percutaneous small bowel access was then obtained and fluoroscopy with intraluminal contrast was performed, but again a clear stenosis could not be identified. Balloon dilation was not performed, and therefore, criteria for technical success were not met. The patient recovered uneventfully after an open enterolysis. A second case (case 6) began with a cut-down, and a chain of multiple long strictures close together was identified. The length of these narrowed areas was too extensive for balloon dilation, so an open small bowel resection was performed. Although balloon dilation was not attempted, radiographic detail from the endoluminal contrast was used to define the extent of resection. Case 11, with advanced carcinomatosis, was converted from a cut-down approach to ileostomy creation. Balloon dilation was performed successfully, but the patient had an additional area of distal stenosis. Given the patient’s frailty and numerous comorbid conditions, an ileostomy was judged to be the safest solution. Intraluminal contrast was used to confirm that the location and orientation of the loop ileostomy and that the small bowel was completely open. Radiographic confirmation that the loop was upstream of the obstruction obviated the need for extensive adhesiolysis or a larger incision.

Procedure times were highly variable. The median procedure time was 135 min, with a range from 27 min to 416 min. Percutaneous access cases tended to be the lengthiest. Several cases were particularly long, most commonly due to dense adhesions or difficulty in passing the wire or instruments via the ileocecal valve. Unsuccessful cases or those that required conversion to another method tended to also have prolonged operative times.

All patients were discharged home tolerating at least a liquid diet. The median postoperative length of hospital stay was 4 days (IQR 2.0–6.25 days). The longest stay of 16 days was observed in a patient who failed management via the cut-down approach. A diverting stoma was created; however, the patient had numerous comorbid conditions which required additional management before discharge.

The median follow-up was 10 months (IQR 7–11 months). At the latest follow-up, two patients had recurrent SBO. Case 2 had recurrence 6 months after percutaneous balloon dilation of radiation-induced strictures. This was treated with repeat percutaneous dilation; however, the patient developed a second recurrence 13.5 months from the initial procedure. This was treated with open ileocolic resection and anastomosis. Extensive radiation enteritis and colitis were noted at the time of laparotomy. Case 4 underwent cut-down balloon dilation and developed recurrence at 3 months, treated with repeat cut-down procedure. The patient had a second recurrence 4.5 months after the index procedure that was treated with mini-laparotomy, adhesiolysis, and repeat cut-down balloon dilation. Extensive adhesions were noted at each of the three procedures. No recurrence was seen in these patients after the aforementioned treatments with a follow-up of now over 5 months.

Discussion

Surgical methods (laparoscopic or open) are the gold standard therapy for treating most cases of symptomatic SBOs but are associated with substantial morbidity and a high risk of recurrence. Endoluminal balloon dilation may be a promising alternative but previously has not been feasible for most patients due to the limited access routes available to the small bowel with traditional endoluminal approaches. We have developed endoscopic, percutaneous, and PA/cut-down methods combined with radiological imaging in the operating theatre that permit expanded application of balloon dilation for forms of SBO wherein a narrowing of the lumen has been diagnosed in the preoperative period. These access methods blend endoscopic, radiological, and traditional surgical techniques to guide access to the small bowel. In this study, we report on the characteristics of patients selected for these approaches as well as intraoperative and short-term outcomes. Technical success was achieved in 10 of 12 patients, avoiding the need for surgery to treat SBO. No major complications occurred.

A wealth of information is gained with the extensive cross-sectional imaging that nearly all patients undergo prior to abdominal surgery. Currently, in most cases, this knowledge is used for diagnostic purposes preoperatively but then not actively used in the operating room. In contrast, cardiologists, interventional radiologists and gastroenterologists have more actively used preoperative imaging during their procedures. We are learning that with preoperative imaging, combined with intraoperative imaging using fluoroscopy and cone beam CT, along with endoscopic and percutaneous methods under development, we were able to avoid major surgery in the majority of patients in this cohort.

The anatomical location of the obstruction is an important factor when choosing an approach. In all cases, we recommend intraoperative radiological correlation in concert with one of the access methods. Endoscopic access is our preferred approach whenever feasible as it is the least invasive. Percutaneous access alone can permit intervention on lesions not amenable to purely endoscopic access, but they must nonetheless be relatively close to the ileocecal valve. The PA/cut-down method is largely free from these constraints and can theoretically be performed anywhere along the small bowel. In practice however, the PA/cut-down approach is limited by anatomical or structural factors. Extensive intra-abdominal scarring and intervening structures such as distended proximal bowel, liver, or transverse colon can limit the areas amenable to percutaneous access. Theoretical disadvantages of the PA/cut-down approach are morbidity from the small abdominal wound and risk of leak from the closed enterotomy. In this small series, we did not identify any wound-related or other complications from either the cut-down or other percutaneous devices such as T-tacks.

Device selection for percutaneous access evolved during the study period. Selection of wires, sheaths, and balloons was based on individual preferences at the start of the study period. Throughout the study period, devices were evaluated and their use evolved. See online supplemental tables 2 and 3 for a listing of devices and interactive changes in their use with each case.

Procedure times were longest with percutaneous access cases. The reasons for these long procedure times were unique to each individual case. In one protracted case, there was considerable difficulty in cannulating the ileocecal valve due to scarring, even with assistance from a percutaneous grasper. Percutaneous cases also required additional steps at the conclusion of the procedure, such as endoscopic clip placement, which add additional time.

In the current configuration, percutaneous access requires initial endoscopic evaluation with an over-the-scope device, followed by percutaneous puncture into the bowel. Insufflation from the endoscope is used to help dilate the bowel up against the anterior abdominal wall for the Seldinger technique sheath insertion, as in percutaneous gastrostomy tube placement. If endoscopic insufflation could be foregone in favor of a purely image-guided approach, this could eliminate some of the procedure time and expand the range of potential therapeutic targets. A variety of equipment and techniques were used across the three approaches (table 2), indicating that there is still considerable evolution of these approaches ongoing. It is likely that future iterative improvements to these approaches will occur.

Even in cases with technical failure of one of the novel methods, the approach still provided valuable information which informed the next steps of the operation. In only one case (case 3) did the percutaneous method not significantly alter the surgical plan. If endoluminal methods are insufficient to relieve the stenosis, the information gained in endoluminal access can limit the extent of further operation required. Radiographic methods, in particular, intraluminal contrast with fluoroscopy or cone beam CT (leading to 3-D images), can verify the orientation of a stoma, confirming that the planned diversion is proximal to the area of concern. This information can greatly limit the dissection necessary for diversion, preventing the need to perform a laparotomy and extensive adhesiolysis.

The recurrence rate after endoscopic balloon dilation of intestinal strictures, when within reach of an endoscope, has historically been relatively high. Small series reporting balloon dilation of colorectal anastomotic strictures typically show excellent efficacy, but many patients require two or more dilation procedures.13 In the small bowel, other series have reported stricture recurrence rates in other disease processes and locations (eg, Crohn’s enteritis) as ranging from 10% to 50%.14 15 In the short follow-up period reported here, we identified two recurrences. Given the small number of patients involved, there were no clear patterns identifiable. Undoubtedly, stricture location, length and the underlying disease process are important determinants of the risk of recurrence. Further work should clarify the preoperative characteristics of patients prone to recurrence.

This study had several important limitations. All procedures were performed at a single institution by one team, and cases were evaluated retrospectively, limiting the generalizability of our findings. No data were available on patients who were offered one of these novel methods but declined and opted for a different approach. Iterative refinements are ongoing, and these techniques are likely to evolve. Although there were no major complications in the present series, this is an important consideration for future work. Further study with a larger number of patients will be required to examine the safety and efficacy of these approaches, clearly in their early stage of development. The optimal patient population for these treatments remains to be defined.

In a heterogenous cohort of patients with SBO, we demonstrated the feasibility of novel access methods for the relief of obstruction. Just as other ‘open’ invasive procedures have become uncommon with advancing technology, we believe open surgical access to the small bowel for the treatment of SBO can evolve to far less invasive approaches with the use of a combination of advanced imaging, endoscopy and new tools. The three approaches we have described—purely endoscopic, percutaneous and cut-down, with use of advanced intraoperative imaging methods in the operating theater—challenge the notion that open or laparoscopic surgical methods are the only means of treating SBO. Our approaches seem safe, with no major complications in this small initial series. Short-term clinical success was achieved in 10 of 12 patients. Recurrence/relapse was seen in two patients, and long-term follow-up will be an important future consideration. Our next task, building on this work, is to develop a prospective study using an advanced IDEAL 2a methodology.