: To evaluate feasibility of near-infrared indocyanine green (NIR-ICG) imaging for bowel vascularization assessment after full thickness bowel resection for recto-sigmoid endometriosis (RSE).
: This is a prospective, single-center, preliminary study on consecutive symptomatic patients submitted to discoid or segmental resection for RSE and NIR-ICG evaluation for vascular assessment of anastomotic line, from May 2018 to January 2020.
: Tertiary university hospital.
: Thirty-two women with RSE meeting eligibility criteria were included for study analysis.
: NIR-ICG evaluation of anastomotic line vascularization after RSE removal.
Main Outcome Measure(s)
: Fluorescence degree of anastomotic line was assessed with a 0-2 Likert scale, as follow: 0 or “absent” (no fluorescence observed), 1 or “irregular” (not uniform distribution or weak fluorescence), and 2 or “regular” (uniform distribution of fluorescence and similar to proximal colon).
: In all the patients included in the study (100%) NIR-ICG allowed the evaluation of fluorescence degree of the anastomotic line. No adverse reaction related to ICG use was recorded. The protocol did not greatly lengthen operating time [median, 4 (range, 3-5) minutes]. Excellent inter-operator agreement was observed. Most of the patients (31/32, 96.9%) showed regular fluorescence on anastomotic line; in one patient with irregular fluorescence at NIR-ICG after discoid excision, anastomotic suture was reinforced through interrupted stitches. Notably, we had one case of anastomotic leakage after segmental resection with intraoperative good fluorescence at NIR-ICG evaluation.
: NIR-ICG imaging for anastomotic perfusion assessment after discoid or segmental resection for RSE seems to be a feasible, safe and reproducible method.
Bowel endometriosis is defined as the presence of endometrial glands and stroma infiltrating the bowel wall and reaching at least the muscular layer . Rectum and sigmoid colon are the most common involved gastrointestinal sites accounting for about 90% of cases of bowel endometriosis . Rectosigmoid endometriosis (RSE) affects from 8 to 12% of patients with endometriosis and causes a severe impairment of women’s health with progressive pelvic pain, deep dyspareunia, diarrhoea, constipation, tenesmus, dyschezia, or bowel occlusion [3,4]. Surgery for RSE is required in patients with contraindications or poor response to hormonal therapies or in case of complicated endometriosis (bowel or urinary stenosis) .
Despite its proven efficacy on pain symptoms, surgery for RSE is associated with not negligible risk of surgical complications, including bowel perforation/leakage or rectovaginal fistula (2.9% to 10.6%) and pelvic abscess (6.0%) [6,7].
It has been clearly established that good perfusion is essential to prevent bowel complications [8,9].
Near infrared imaging with indocyanine green (NIR-ICG) enables surgeons to visualize real-time tissue vascularization, helping intraoperative evaluation and decision in several fields, including colorectal surgery . In this setting, NIR-ICG use and the related changes in surgical course (revision of the planned transection line or anastomosis) were associated with a decrease in anastomotic leakage and re-intervention .
Regarding endometriosis surgery, Bourdel et al. demonstrated NIR-ICG as a safe and feasible technology that may reduce major bowel complications after shaving technique for RSE. Nevertheless, opening of the bowel lumen during discoid excision and/or segmental resection for RSE represents the main independent predictive factor for rectovaginal fistula and rectosigmoid perforation .
In the present study we aim to evaluate feasibility and safety of NIR-ICG use to intraoperatively assess blood perfusion of the anastomotic line after full thickness bowel resection for RSE.
MATERIAL AND METHODS
This is a prospective, monocentric, preliminary study carried out in a tertiary level referral center for endometriosis. Consecutive symptomatic patients scheduled for minimally-invasive surgical excision of RSE from May 2018 to January 2020 were enrolled. Exclusion criteria were: age less than 18 years old, menopausal status, contraindications to minimally invasive surgery, allergy to iodine, inflammatory bowel diseases or previous colorectal surgery. Women submitted to shaving technique for RSE were excluded from study analyses. The study was approved by the local ethical committee (225/2017/O/Oss). Anamnestic and preoperative data were recorded, including: age, body mass index (BMI), parity, infertility, previous surgery for endometriosis, preoperative hormonal therapy within 3 months before surgery, and severity of pain symptoms according to the 11-point Visual Analogue Scale (VAS). As per routine clinical practice, patients underwent gynecologic examination, transvaginal and transabdominal ultrasound scan, and in selected cases (bowel nodule diameter >3 cm, rectal implants involving the inner muscularis propria at transvaginal ultrasonography, or severe bowel symptoms) magnetic resonance imaging . All patients provided signed informed consent before entering the study; they were also informed about the off-label use of indocyanine green for study purpose.
The day before surgery, monobasic sodium phosphate and a clear liquid diet were provided. The operative procedures were performed by a gynecological surgical staff with extensive experience in the management of deep infiltrating endometriosis. The peritoneal cavity was inspected and all the endometriosis lesions were described according to revised American Society for Reproductive Medicine classification (rASRM) . The choice of the bowel surgical procedure was personalized and made at the time of surgery according to the depth and size of colorectal involvement.
Bowel procedures were performed as previously published developing peri-rectal avascular retroperitoneal spaces with a nerve-sparing technique . Shaving was attempted first and consisted in careful dissection of the endometriotic nodule peeling it off the bowel wall with subsequent suture when needed. In case of persistence of deep bowel wall infiltration, a discoid excision of the rectal wall was performed using a trans-anal circular stapler for bowel implants <3 cm, located on the ventral surface of the rectum, and within 15 cm from the anal verge; otherwise, for lesions ≥3 cm, or those involving >50% of circumference or impacting the sigmoid tract, a segmental resection was performed together with a dedicated colorectal surgeon. In this latter case, the bowel was distally sectioned using linear stapler 1 to 2 cm under the involved tract, it was then exteriorized through a small incision (3 cm) at the point of the suprapubic trocar and it was excised. An end-to-end or end-to-side anastomosis was done using a trans-anal circular stapler.
After completion of the bowel surgery, bowel integrity was always tested by filling the pelvic cavity with warm saline solution and insufflating air in the rectum or by methylene blue injection through the anus (Michelin test). Protective ileostomy was created in case of ultra-low resection (<5 cm from the anal verge) and associated posterior colpotomy . In all patients a pelvic drain was inserted at the end of surgery.
In patients submitted to discoid or segmental bowel resection, a bolus of 0.25 mg per kg of ICG was injected in a peripheral vein. After injection, fluorescence became rapidly visible on the bowel wall through the activation of the NIR camera-head (KARL STORZ SE & Co., KG, Tuttlingen, Germany) with laparoscope placed at the shortest distance required to capture the entire operative area of interest. Bowel perfusion score was assessed using a 0-2 Likert scale as follow: 0 or “absent” (no fluorescence observed), 1 or “irregular” (not uniform distribution or weak fluorescence), and 2 or “regular” (uniform distribution of fluorescence and similar to proximal colon).
The time required for NIR-ICG protocol, defined as the time from ICG intravenous injection to the moment when a fluorescence degree was assessed, was registered in all cases. Slower uptake over a minute during latency period (from ICG injection to first detection of fluorescence along anastomotic line) or segmental defect were also considered as abnormal findings .
If irregular fluorescence (Likert scale 1) was demonstrated up to 2 minutes from ICG injection, the bowel was observed for up to 5 minutes. If absent fluorescence (Likert scale 0) was established up to 2 minutes from ICG injection, an additional ICG bolus was considered.
All surgical videos were recorded and stored with identification numbers using a random number generator. First surgeon who performed all the procedures (RS) assessed degree of fluorescence of NIR-ICG evaluation during the operation. A second surgeon (MM) performed a blinded review from the recorded videos.
Protocol scheme planned revision of anastomosis or change of surgical technique from discoid to segmental resection in case of absent bowel fluorescence; otherwise, additional reinforcing stitches were recommended in case of irregular fluorescence.
During immediate postoperative course, vital parameters including body temperature were recorded three times/day. Blood value of C reactive protein (CRP) was assessed at day 3, 5 and 7. White blood cells (WBCs) count was carried out every 48 hours. When patients presented external temperature >38.0°C, signs of acute abdomen and progressive increase in CRP or WBC, computed tomography (CT) with intravenous plus or minus rectal contrast was performed to rule out postoperative complications.
Criteria for hospital discharge in our clinic include stool canalization, decrease of inflammatory markers and regular vital parameters. For each patient, peri-operative surgical data were collected, including changes to the surgical strategy (additional bowel stitches, change of bowel procedure) and intra- and post-operative complications according to Clavien-Dindo classification . As for usual clinical practice, post-operative visit and ultrasound examination was scheduled after 1 and 3 months.
The primary outcome of this study was protocol completion rate of NIR-ICG evaluation to assess bowel vascularization after surgery for RSE. Secondary outcomes were: time added by the protocol, concordance between two surgeons regarding fluorescence score, peri-operative complications and change of surgical strategy.
Continuous data were expressed as mean ± standard deviation or median (range). Categoric variables were expressed as absolute number and percentages. Cohen’s k coefficient was calculated to determine inter-operator agreement between 2 observers in classifying residual colorectal perfusion using NIR-ICG imaging. A k value of 0.81 to 1.00 indicated very good agreement, 0.61 to 0.80 good agreement, 0.41 to 0.60 moderate agreement, 0.21 to 0.40 fair agreement, and <0.20 poor agreement. Statistical analysis was performed using the Statistical Package for the Social Sciences software (IBM SPSS v. 25, SPSS Inc.).
During study period, 46 women with symptomatic RSE scheduled for minimally invasive complete removal of macroscopic lesions were enrolled. Fourteen patients (14/46,30.4%) submitted to shaving technique were excluded according to protocol criteria.
TABLE 1Demographic and clinical data of the study group (32 women).
TABLE 2Surgical details of the study group (32 women).
TABLE 3NIR-ICG evaluation for vascularization assessment of bowel anastomotic line (32 women).
No intraoperative complications, including those related to ICG use, were recorded. Median latency time (from ICG injection to first visual detection) was 33 (range,6-41) seconds. No additional ICG bolus was performed. Median protocol time was 4 (range, 3-5) minutes.
We observed postoperative complications. We reported 2 (6.2%) grade II Clavien-Dindo complications: two cases of anemia requiring blood transfusion. Three complications (9.4%) belonging to grade III of Clavien-Dindo classification were observed: one patient complained rectorrhagia needing rectoscopy, one showed grade II hydroureteronephrosis requiring double J stent placement, and one had an anastomotic leakage.
This latter underwent recto-sigmoid segmental resection with end-to-side mechanical anastomosis and presented a regular fluorescence score at NIR-ICG evaluation at the end of surgical procedure. Three days after the postoperative course was complicated by abdominal pain, hyperpyrexia and inflammatory markers increase. CT scan showed free air into abdominal cavity, bowel anastomotic defect and pelvic abscess. The patient underwent a second laparoscopy with thorough washing of the abdominal cavity, suture of the anastomotic defect and a temporary ileostomy. No other adverse events occurred during her subsequent postoperative course.
Inter-operator agreement rate in classifying bowel perfusion using NIR-ICG was very good (k, 1.0 [95% confidence interval, 1.0 −1.0]).
Three months after surgery, no patient was lost to follow-up.Twenty-eight women (87.5%) assumed hormonal therapy. All pain symptoms significantly improved after surgery and no late surgical complication was reported.
To the best of our knowledge, this is the first study showing NIR-ICG imaging as a feasible, safe and reproducible method for anastomotic perfusion assessment after discoid or segmental resection for RSE.
Several studies demonstrated the efficacy of bowel surgery for endometriosis based on a low rate of symptomatic and anatomical recurrence . However, colorectal surgery for RSE is associated with a significant morbidity, eventually requiring additional interventional procedures and drastically reducing quality of life of the affected patients .
Among surgical complications, bowel fistula (anastomotic leakage and rectovaginal fistula) represents the most serious and life-threatening complication occurring from 2.9% to 10.6% of operated patients and leading to longer hospitalization, slower recovery and eventual second surgery .
Adequate bowel perfusion after surgical procedure is essential for optimal healing and prevention of surgical morbidity . Intraoperative detection of bowel devascularization and change of surgical plan may reduce the risk of bowel fistula, as demonstrated in colorectal surgery .
Traditionally, surgeons check for an adequate vascularization of the bowel after colorectal procedures at white light through evaluation of bowel color, vessels pulsatility and, in case of segmental resection, bleeding from the edges of anastomotic stumps. However, these highly subjective parameters may under or overestimate the assessment .
NIR-ICG is considered a promising intraoperative tool to visualize real-time tissue vascularization. A systematic review on 5498 patients from twenty studies, including two randomized controlled trials, showed that the risk of anastomotic leak may be significantly reduced through the systematic use of ICG fluorescence imaging in patients undergoing surgery for colorectal cancer .
Over the last few years, NIR-ICG was employed in endometriosis surgery for the detection of occult lesions  and the assessment of ureteral perfusion after ureteral surgery . A few data are available about fluorescence angiography during bowel endometriosis excision. Video case reports described ICG use to assess bowel vascularization before and after discoid or segmental resection for RSE [24-26]. Two case series applied fluorescence-guided laparoscopy to the surgical excision of rectovaginal deep nodules reporting a better visualization of the margins and separation of lesions from healthy surrounding tissue and reducing the risk of rectal and vaginal perforation [27,28].
Bourdel et al.  first investigated NIR-ICG evaluation for bowel vascular assessment in women with RSE, but only after shaving technique. In this preliminary study on 21 patients most of the patients showed a very good fluorescence at the rectal shaving area; only in one case the fluorescence was considered sub-optimal requiring reinforced stitches on rectal serosa. Differently from this study we reported NIR-ICG use in RSE patients submitted to discoid excision and segmental resection, which have been demonstrated as independent predictive factors for bowel fistula .
In line with Bourdel et al.’s results , in our study NIR-ICG imaging was completed allowing the assessment of anastomotic line perfusion in all cases, with rapid protocol time and excellent agreement between two operators in two different settings.
We experienced one case of irregular fluorescence after discoid resection requiring additional interrupted stitches. Noteworthy, one patient submitted to segmental resection with a regular fluorescence at NIR-ICG evaluation presented with a postoperative anastomotic leak. Using computer analysis program and dynamic perfusion cartogram, quantitative method of NIR-ICG evaluation may improve the accuracy of this technology to predict bowel complications [29, 30]. In particular, perfusion time related factors rather than intensity of fluorescence could be predictors of bowel complications, as demonstrated by a recent clinical study evaluating quantitative analysis of colon perfusion after rectal oncological procedures .
Despite the prospective design and the inclusion of RSE patients at high risk of bowel complications, the present study has some drawbacks. First, we did not evaluate ICG fluorescence from the luminal side through trans-anal approach, which may allow vascular assessment of entire colorectum circumference including the mucosal layer . However, tissue penetration of this fluorophore is up to 5 mm and covers the entire depth of colorectal wall ; in addition, anal endoscopy is an invasive method that is not routinely performed after colorectal surgery for RSE.
Although the clinical decision to change surgical plan according to the NIR perfusion assessment was based on meta-analysis including randomized clinical trials from general surgeons performing colorectal resection for several diseases , this approach needs to be validated by further larger, multicentric and controlled studies. However, a formal proof that NIR-ICG for bowel perfusion assessment after discoid or segmental resection for RSE is clinically useful would require a large number of cases.
Furthermore, specific light source and optic device are quite expensive and not yet included in basic endoscopic operating rooms. The cost of this new technology (specific equipment and fluorophore) can obstruct widespread application to resource-limited health-care systems.
In conclusion, NIR-ICG for bowel perfusion assessment seems to be a safe, feasible and reproducible method, that could influence intra-operative decision-making and improve surgical outcomes of RSE patients. Further studies are needed to evaluate the efficacy of NIR-ICG guided bowel surgery for RSE in terms of surgical morbidity reduction and cost-effectiveness analysis.
No financial support was received for this study.
Seracchioli R, Mabrouk M, Guerrini M, et al. Dyschezia and posteror deep infiltrating endometriosis: analysis of 360 cases. Invasive Gynecol. 2008 Nov-Dec;15(6):695-9.
Abrao MS, Petraglia F, Falcone T, et al. Deep endometriosis infiltrating the recto-sigmoid: critical factors to consider before management. Hum Reprod Update. 2015; 21:329–339.
Mabrouk M, Montanari G, Di Donato N, et al. What is the impact on sexual function of laparoscopic treatment and subsequent combined oral contraceptive therapy in women with deep infiltrating endometriosis?. J Sex Med. 2012;9(3):770-778.
Mabrouk M, Frascà C, Geraci E, et al. Combined oral contraceptive therapy in women with posterior deep infiltrating endometriosis. J Minim Invasive Gynecol. 2011;18(4):470-474.
Mabrouk M, Raimondo D, Altieri M, et al. Surgical, Clinical, and Functional Outcomes in Patients with Rectosigmoid Endometriosis in the Gray Zone: 13-Year Long-Term Follow up. J Minim Invasive Gynecol. 2019 Sep-Oct;26(6):1110-1116.
Roman H, Bridoux V, Merlot B, et al. Risk of bowel fistula following surgical management of deep endometriosis of the rectosigmoid: a series of 1102 cases. Hum Reprod. 2020 Jul 1;35(7):1601-1611.
Abo C, Moatassim S, Marty N, et al. Postoperative complications after bowel endometriosis surgery by shaving, disc excision, or segmental resection: a three-arm comparative analysis of 364 consecutive cases. Fertility Steril. 2018 Jan;109(1):172-178.e1.
Spagnolo E, Zannoni L, Raimondo D, et al. Urodynamic evaluation and anorectal manometry pre- and post-operative bowel shaving surgical procedure for posterior deep infiltrating endometriosis: a pilot study. J Minim Invasive Gynecol, 2014;21(6):1080-1085.doi:10.1016/j.jmig.2014.05.012
Frasson M, Flor-Lorente B, Ramos Rodrıguez JL, et al. ANACO Study Group; ANACO Study Group (2015) Risk factors for anastomotic leak after colon resection for cancer: multivariate analysis and nomogram from a multicentric, prospective, national study with 3193 patients. Ann Surg 262(2):321–330
Keller DS, Ishizawa T, Cohen R, et al. Indocyanine green fluorescence imaging in colorectal surgery: overview, applications, and future directions. Lancet GastroenterolHepatol. 2017 Oct;2(10):757-766.
Chan DKH, Lee SKF, Ang JJ. Indocyanine green fluorescence angiography decreases the risk of colorectal anastomotic leakage: Systematic review and meta-analysis. Surgery. 2020 Sep 30:S0039-6060(20)30556-0. doi: 10.1016/j.surg.2020.08.024. Epub ahead of print. PMID: 33010938.
Bourdel N, Jaillet L, Bar-Shavit Y, et al. Indocyanine green in deep infiltrating endometriosis: a preliminary feasibility study to examine vascularization after rectal shaving. FertilSteril. 2020 Jul 7:S0015-0282(20)30329-0.
Revised American Society for Reproductive Medicine classification of endometriosis: 1996. FertilSteril. 1997 May;67(5):817-21.
Raimondo D, Mattioli G, Degli Esposti E, et al.Impact of Temporary Protective Ileostomy on Intestinal Function and Quality of Life after a 2-Year Follow-up in Patients Who Underwent Colorectal Segmental Resection for Endometriosis. J Minim Invasive Gynecol. 2020 Sep-Oct;27(6):1324-1330.
Nguyen JMV, Hogen L, Laframboise S, et al. The use of indocyanine green fluorescence angiography to assess anastomotic perfusion following bowel resection in surgery for gynecologic malignancies – A report of 100 consecutive anastomoses. Gynecol Oncol. 2020 Aug;158(2):402-406.
Dindo D, Demartines N, Clavien PA. Classification of surgical complications: a new proposal with evaluation in a cohort of 6336 patients and results of a survey. AnnSurg. 2004 Aug;240(2):205-13.
Bendifallah S, Vesale E, Daraï E, et al. Recurrence after Surgery for Colorectal Endometriosis: A Systematic Review and Meta-analysis. J Minim Invasive Gynecol. 2020;27(2):441-451.e2.
Turco LC, Scaldaferri F, Chiantera V, et al. Long-term evaluation of quality of life and gastrointestinal well-being after segmental resection for deep infiltrating endometriosis (ENDO-RESECT QoL). Arch Gynecol Obstet. 2020 Jan;301(1):217-228
Zheng Y, Zhang N, Lu W, Zhang L, Gu S, Zhang Y, Yi X, Hua K. Rectovaginal fistula following surgery for deep infiltrating endometriosis: Does lesion size matter? J Int Med Res. 2018 Feb;46(2):852-864.
Kingham TP, Pachter HL (2009) Colonic anastomotic leak: risk factors, diagnosis, and treatment. J Am Coll Surg 208(2):269–278
Karliczek A, Harlaar NJ, Zeebregts CJ, et al. (2009) Surgeons lack predictive accuracy for anastomotic leakage in gastrointestinal surgery. Int J Colorectal Dis 24(5):569–576
Vizzielli G, Cosentino F, Raimondo D, et al. Real three-dimensional approach vs two-dimensional camera with and without real-time near-infrared imaging with indocyanine green for detection of endometriosis: A case-control study [published online ahead of print, 2020 Apr 9]. Acta ObstetGynecolScand. 2020;10.1111/aogs.13866. doi:10.1111/aogs.13866
Raimondo D, Borghese G, Mabrouk M, et al. Use of Indocyanine Green for Intraoperative Perfusion Assessment in Women with ureteral Endometriosis: A preliminary Study. J Minim Invasive Gynecol 2020 Apr 10;S1553-4650(20)30178-3.
Seracchioli R, Raimondo D, Arena A, et al. Clinical use of endovenous indocyanine green during rectosigmoid segmental resection for endometriosis. FertilSteril. 2018 Jun;109(6):1135.
Malzoni M, Iuzzolino D, Rasile M, et al. Surgical Principles of Segmental Rectosigmoid Resection and Reanastomosis for Deep Infiltrating Endometriosis. J Minim Invasive Gynecol. 2020 Feb;27(2):258.
Spagnolo E, Hernández A, Pascual I, et al. Bowel and ureteral assessment by indocyanine green real-time visualization during deep infiltrating endometriosis surgery. Colorectal Dis. 2020 Apr 26.
De Neef A, Cadière GB, Bourgeois P, et al. Fluorescence of Deep Infiltrating Endometriosis During Laparoscopic Surgery: A Preliminary Report on 6 Cases.SurgInnov. 2018 Oct;25(5):450-454.
Siegenthaler F, Knabben L, Mohr S, et al. Visualization of endometriosis with laparoscopy and near-infrared optics with indocyanine green. Acta ObstetGynecol Scand. 2020 May;99(5):591-597.
Son GM, Kwon MS, Kim Y, et al. Quantitative analysis of colon perfusion pattern using indcyanine green (ICG) angiography in laparoscopic colorectal surgery.SurgEndosc. 2019 May;33(5):1640-1649.
Diana M, Agnus V, Halvax P, et al. Intraoperative fluorescence-based enhanced reality laparoscopic real-time imaging to assess bowel perfusion at the anastomotic site in an experimental model. Br J Surg. 2015 Jan;102(2):e169-76.
Amagai H, Miyauchi H, Muto Y, et al. Clinical utility of transanal indocyanine green near-infrared fluorescence imaging for evaluation of colorectal anastomotic perfusion. SurgEndosc. 2019 Dec 9.
Received in revised form:
In Press Journal Pre-Proof
Conflict of Interest: The authors report no conflict of interest.
Precis: NIR-ICG for bowel perfusion assessment seems to be a safe, feasible and reproducible method, that could influence intra-operative decision-making and improve surgical outcomes of RSE patients.
© 2020 Published by Elsevier Inc. on behalf of AAGL.