DOI: https://doi.org/10.21203/rs.3.rs-1943148/v1
Background: There is insufficient evidence on whether indocyanine green (ICG) fluorescence angiography can reduce the incidence of anastomotic leakage (AL). This retrospective cohort study aimed to evaluate the effect of ICG fluorescence angiography on AL rates in laparoscopic rectal cancer surgery at a single institution.
Methods: Patients who underwent laparoscopic low anterior resection or intersphincteric resection with ICG fluorescence angiography (ICG group; n=73) and patients who underwent a similar surgical procedure for rectal cancer without ICG fluorescence (non-ICG group; n=114) were enrolled consecutively in this study. ICG fluorescence angiography was performed prior to transection of the proximal colon, and anastomosis was performed with sufficient perfusion using ICG fluorescence imaging. AL incidence was compared between both groups, and the risk factors for AL were analyzed.
Results: AL occurred in 3 (4.1%) and 14 (12.3%) patients in the ICG and non-ICG groups, respectively. In the ICG group, the median perfusion time from ICG injection was 34 s, and 5 patients (6.8%) required revision of the proximal transection line. None of the patients requiring revision of the proximal transection line developed AL. In the univariate analysis, longer operating time (odds ratio: 2.758; 95% confidence interval: 1.023–7.624) and no implementation of ICG fluorescence angiography (odds ratio: 3.266; 95% confidence interval: 1.038–11.793) were significant factors associated with AL incidence, although the creation of a diverting stoma or insertion of a transanal tube was insignificant.
Conclusion: ICG fluorescence angiography was associated with a significant reduction in AL during laparoscopic rectal cancer surgery. Changes in the surgical plan due to ICG fluorescence visibility may help improve the short-term outcomes of patients with rectal cancer.
Anastomotic leakage (AL) is a critical complication of colorectal surgery. Previous studies have reported that AL in colorectal surgery occurs in 2–28% of patients [1–7]. AL can worsen short-term and long-term outcomes, such as local recurrence rates [8]. The risk factors of AL have been reported to be male sex, preoperative chemoradiotherapy, and lower anastomosis [9, 10]. In contrast, anastomotic tension, incomplete anastomosis, and vascular perfusion in anastomosis are considered surgery-related factors affecting AL incidence [11–13].
Regarding vascular perfusion in colorectal anastomosis, near-infrared fluorescence imaging using indocyanine green (ICG) has recently been applied to assess intestinal blood flow [14]. The results of the PILLAR II trial showed that the rate of AL was 1.4% in left-sided colorectal resections. Several studies have shown that ICG fluorescence angiography could achieve a low AL prevalence of 2.8 − 4.7% in rectal surgery [15–18]. To our knowledge, only two multicenter randomized controlled trials (RCTs) have investigated the effectiveness of ICG fluorescence angiography on anastomotic leakage after rectal surgery [19, 20]. However, neither trial showed the effectiveness of ICG fluorescence angiography compared with the control group. Therefore, there is insufficient evidence on whether ICG fluorescence can reduce AL incidence.
This retrospective cohort study aimed to evaluate the effect of ICG fluorescence angiography on AL rates in laparoscopic rectal cancer surgery at a single institution.
This retrospective cohort study was conducted at Kagawa University Hospital in Kagawa, Japan. The study was approved by the Institutional Review Board (IRB) of Kagawa University, Kagawa, Japan (approval number: 2021–032). The study and the manuscript adhered to the STROBE guidelines for observational studies. The requirement for informed consent was waived with permission from the IRB due to the retrospective nature of the study.
A total of 79 patients who underwent elective laparoscopic anterior resection or intersphincteric resection (ISR) with lymphadenectomy for rectal cancer using ICG fluorescence angiography between May 2019 and December 2021 were included. A total of 124 patients who underwent elective laparoscopic anterior resection or ISR with lymphadenectomy for rectal cancer without ICG fluorescence angiography between January 2014 and April 2019 were included in the control group. Therefore, 203 patients were included in this study. In this total study periods, surgical procedures and postoperative care were identical before and after ICG use. Patients who underwent multivisceral organ resection due to tumor invasion (n = 14) or had a history of rectal cancer surgery (n = 4) were excluded from this study. Consequently, 187 patients who underwent laparoscopic anterior resection or ISR for rectal cancer were included in the analysis.
The laparoscopic surgical technique was standardized at our institution, and all procedures were performed by experienced colorectal surgeons. After pneumoperitoneum and placement of the five ports, medial to lateral dissection and lymphadenectomy were performed. After dissection of the mesorectum, the rectum was transacted using a linear stapler. Anastomosis for anterior resection was performed using a double-stapling technique with a circular stapler. When performing ISR, a hand-sewn anastomosis was performed.
After rectal transection, the specimen was extracted through the umbilical port, which was extended to approximately 3–5 cm. A bolus of ICG (10.0 mg) was injected intravenously after dissection of the mesocolon and immediately before the transection of the proximal colon. The visualization of ICG fluorescence at the level of the planned transection line, where surgeons choose under macroscopic inspection, was assessed in a completely dark operating room using an Olympus Medical Imaging Video System and Laparoscope (Olympus, Leiderdorp, the Netherlands). If the surgeon judged that bowel perfusion was poor via ICG fluorescence imaging, the resection line of the proximal colon was changed to the proximal colon where ICG fluorescence was clearly visible (Fig. 1).
The following variables were collected and analyzed: patient characteristics, comorbidities, any preoperative therapy (chemoradiotherapy or chemotherapy), surgical procedures, operative time, blood loss, changes in the transection line from the initial part, postoperative complications, including AL and mortality. Tumor staging was performed using the Union for International Cancer Control tumor-node-metastasis classification system.
Among patients who underwent laparoscopic anterior resection or ISR (n = 187), variables were compared between the ICG (n = 73) and non-ICG (n = 114) groups. Quantitative data were reported as medians (interquartile ranges). Since the data could not assume a normal distribution, the Kruskal–Wallis test was used to compare continuous variables. The chi-square and Fisher’s exact tests were used to compare categorical variables and proportions. Univariate analysis of all patients (n = 187) was performed to assess the factors associated with AL. All statistical analyses were performed using JMP statistical software (version 15.1; SAS Institute Inc., Cary, NC, USA). P-values < 0.05 were considered statistically significant.
The patient and tumor characteristics are presented in Table 1. Age, sex, body mass index, and the presence of preoperative comorbidities were not significantly different between both groups. Neoadjuvant treatment (chemoradiotherapy or chemotherapy) was administered more frequently in the ICG group (p = 0.01) than in the non-ICG group. Tumor location and preoperative staging were not significantly different between both groups.
| ICG group (n = 73) | Non-ICG group (n = 114) | P-value | |
|---|---|---|---|
| Sex, n (%) | 0.524 | ||
| Male | 46 (63%) | 77 (67.5%) | |
| Female | 27 (37%) | 37 (32.5%) | |
| Age, ya | 67 (57–72) | 68 (59–73) | 0.517 |
| BMIa | 22.7 (20.7, 25.1) | 22.2 (19.9, 24.8) | 0.371 |
| Comorbidities, n (%) | |||
| Hypertension | 22 (30.1%) | 36 (31.6%) | 0.835 |
| Diabetes | 18 (24.6%) | 17 (14.9%) | 0.096 |
| Cardiac dysfunction | 9 (12.3%) | 15 (13.2%) | 0.868 |
| Pulmonary dysfunction | 9 (12.3%) | 10 (8.8%) | 0.432 |
| Preoperative therapy, n (%) | 0.010 | ||
| Chemoradiotherapy | 8 (10.9%) | 3 (2.6%) | |
| Chemotherapy | 5 (6.9%) | 2 (1.8%) | |
| Tumor size, mma | 30 (20–50) | 35 (20–50) | 0.160 |
| Tumor location, n (%) | 0.478 | ||
| Upper rectum | 35 (47.9%) | 47 (41.2%) | |
| Middle rectum | 8 (11%) | 19 (16.7%) | |
| Low rectum | 30 (41.1%) | 48 (42.1%) | |
| Tumor staging, n (%) | |||
| Stage I | 21 (28.7%) | 33 (28.9%) | |
| Stage II | 18 (24.6%) | 27 (23.7%) | |
| Stage III | 25 (34.2%) | 40 (35.1%) | |
| Stage IV | 9 (12.5%) | 14 (12.3%) | |
| ICG: Indocyanine green | |||
| aValues are shown as median (interquartile range) | |||
Surgical outcomes are presented in Table 2. There were no significant differences in the proportion of operative procedures (high anterior resection, low anterior resection, or ISR) and lateral pelvic lymph node dissection between the two groups. Operating time was significantly shorter in the ICG group (median, 280 min) than in the non-ICG group (median, 333 min) (p = 0.008). Although the proportion of diverting stoma creation was not significantly different between both groups, the transanal tube was inserted more often in the ICG group (61%) than in the non-ICG group (27%), which was significantly different (p < 0.001). In the ICG group, 5 of 73 patients (6.8%) had the transection line changed to a more proximal colon because blood perfusion at the initially planned bowel was judged to be poor based on ICG fluorescence (Table 2, Fig. 1). One patient required additional bowel resection of 14 cm and underwent splenic flexural mobilization for safe anastomosis.
| ICG group (n = 73) | Non-ICG group (n = 114) | P-value | |
|---|---|---|---|
| Surgical procedure, n (%) | 0.316 | ||
| HAR | 18 (24.7%) | 33 (28.9%) | |
| LAR | 45 (61.6%) | 73 (64%) | |
| ISR | 10 (13.7%) | 8 (7.1%) | |
| Lateral lymph node dissection, n (%) | 8 (10.9%) | 18 (15.8%) | 0.351 |
| Operative time, mina | 280 (212, 375) | 333 (268, 441) | 0.008 |
| Blood loss, mLa | 11 (0, 81) | 24 (0, 114) | 0.638 |
| Diverting stoma | 31 (42.4%) | 40 (35.1%) | 0.310 |
| Transanal tube | 45 (61.6%) | 31 (27.3%) | < 0.001 |
| The perfusion time after ICG injection, secondb | 34 (11, 82) | − | − |
| Revision of proximal transection line, n (%) | 5 (6.8%) | − | − |
| Distance from the initially planned transection line, cmb | 8 (3, 14) | − | − |
| HAR: High anterior resection, LAR: Low anterior resection, ISR: Intersphincteric resection, ICG: Indocyanine green | |||
| aValues are shown as median (interquartile range) | |||
| bValues are shown as median (range) | |||
Details of the postoperative complications are presented in Table 3. Seventeen patients were diagnosed with AL. The AL rate was significantly lower in the ICG group (4.1%) than in the non-ICG group (12.3%) (p = 0.046). In the ICG group, AL did not occur in cases with a changing transection line due to the ICG fluorescence visibility. The rates of other complications, such as surgical site infection, ileus, urinary dysfunction, and anastomotic bleeding, were not significantly different between both groups. The length of postoperative hospital stay was significantly shorter in the ICG group (median, 11 days) than in the non-ICG group (median, 12 days) (p = 0.015). Characteristics of patients diagnosed with AL are presented in Table 4. The AL rates was higher in male patients than in female patients. Operating times were long in patients diagnosed with AL compared to those for entire cohort. More than half of patients required re-operation with diverting stoma in both groups. Days from initial surgery to AL occurrence were shorter in non-ICG group (median, 3 days) than in ICG group (median, 6 days).
| ICG group (n = 73) | Non-ICG group (n = 114) | P-value | |
|---|---|---|---|
| Anastomotic leakage, n (%) | 3 (4.1%) | 14 (12.3%) | 0.046 |
| Surgical site infection, n (%) | 2 (2.7%) | 4 (3.5%) | 0.771 |
| Ileus, n (%) | 4 (5.5%) | 13 (11.4%) | 0.124 |
| Urinary dysfunction, n (%) | 8 (10.9%) | 13 (11.4%) | 0.124 |
| Anastomotic bleeding, n (%) | 4 (5.5%) | 3 (2.6%) | 0.273 |
| Re-operation within postoperative 30 days, n (%) | 3 (4.1%) | 9 (7.9%) | 0.302 |
| Mortality within postoperative 30 days, n (%) | 0 | 0 | |
| Length of postoperative hospital stay, daysa | 11 (10, 14) | 12 (9, 18) | 0.015 |
| ICG: Indocyanine green | |||
aValues are shown as median (interquartile range)
| ICG group (n = 3) | Non-ICG group (n = 14) | |
|---|---|---|
| Sex, n (%) | ||
| Male | 3 (100%) | 11 (78.5%) |
| Female | 0 | 3 (21.5%) |
| Age, ya | 65 (49–74) | 65 (35–76) |
| Low rectal cancer | 1 (33.3%) | 10 (71.4) |
| Operating time, mina | 367 (225–399) | 497 (269–676) |
| Clavien-Dindo classification ≥ grade IIIb | 2 (66.6%) | 8 (57.1%) |
| Days from initial surgery to AL occurrencee | 6 (5–7) | 3 (1–7) |
| ICG: Indocyanine green, AL: Anastomotic leakage | ||
aValues are shown as median (range)
The univariate analysis of the variables associated with the AL incidence is presented in Table 5. The proportion of AL in male patients was higher than that in female patients (11.4% vs. 4.7%), and the proportion of AL in patients with low rectal cancer was higher than that in patients with upper or middle rectal cancer (14.1% vs. 5.5%); however, these differences were insignificant. There were no significant differences in the rates of AL between patients with and without transanal tube insertion (9.2% vs. 9%) and diverting stoma (10.3% vs. 7%). The non-implementation of ICG fluorescence angiography (odds ratio: 3.266; 95% confidence interval: 1.038–11.793) and a longer operating time (odds ratio: 2.758; 95% confidence interval: 1.023–7.624) were significantly associated with a higher AL incidence.
| AL (+) | P-value | OR (95% CI) | ||
|---|---|---|---|---|
| Sex | Male | 14 / 123 (11.4%) | 0.083 | 2.611 (0.821–9.447) |
| Female | 3 / 64 (4.7%) | |||
| Age, ya | ≥ 70 | 4 / 70 (5.7%) | 0.210 | 0.484 (0.151–1.550) |
| < 70 | 13 / 117 (11.1%) | |||
| BMIa | > 25 | 2 / 27 (7.4%) | 0.394 | 0.506 (0.106–2.419) |
| < 25 | 12 / 88 (13.6%) | |||
| Tumor location | Low | 11 / 78 (14.1%) | 0.051 | 2.818 (0.994–7.983) |
| Upper, Middle | 6 / 109 (5.5%) | |||
| Tumor staging | Stage I, II | 11 / 99 (11.1%) | 0.312 | 1.708 (0.604–4.829) |
| Stage III, IV | 6 / 88 (6.8%) | |||
| Neoadjuvant therapy | + | 2 / 18 (11.1%) | 0.754 | 1.283 (0.268–6.123) |
| − | 15 / 169 (8.9%) | |||
| Surgical Procedure | LAR, ISR | 14 / 136 (10.3%) | 0.330 | 1.836 (0.504–6.675) |
| HAR | 3 / 51 (5.9%) | |||
| Operating time (min) | ≥ 360 | 10 / 68 (14.7%) | 0.048 | 2.758 (1.023–7.624) |
| < 360 | 7 / 119 (5.9%) | |||
| Blood loss (mL) | ≥ 100 | 6 / 52 (11.5%) | 0.472 | 1.470 (-.514–4.204) |
| < 100 | 11 / 135 (8.2%) | |||
| Transanal tube | − | 10 / 111 (9%) | 0.962 | 0.975 (0.354–2.688) |
| + | 7 / 76 (9.2%) | |||
| Diverting stoma | − | 12 / 116 (10.3%) | 0.448 | 1.523 (0.513–4.520) |
| + | 5 / 71 (7%) | |||
| ICG fluorescence imaging | − | 14 / 114 (12.3%) | 0.046 | 3.266 (1.038–11.793) |
| + | 3 / 73 (4.1%) |
BMI: Body mass index, ICG: Indocyanine green, ISR: Intersphincteric resection, ICG: Indocyanine green, HAR: High anterior resection, LAR: Low anterior resection, ISR: Intersphincteric resection, OR: Odds ratio, CI: confidence interval
This retrospective cohort study revealed that ICG fluorescence angiography effectively prevents AL after laparoscopic rectal cancer surgery. Furthermore, our results suggested that ICG fluorescence imaging can reduce AL occurrence by utilizing the results of blood perfusion using ICG fluorescence.
In this study, intestinal perfusion was assessed using ICG fluorescence angiography. Traditionally, the presence of blood flow is evaluated using several clinical signs, such as palpable pulsation, peristaltic movement, or active bleeding from the marginal artery [21]. However, these assessment methods were dependent on the surgeons. Karliczek et al. reported that the clinical judgment of surgeons appeared to have low sensitivity and specificity in predicting anastomotic leakage in colorectal anastomoses [22]. Recently, several techniques, such as oxygen spectrometry, laser speckle imaging, thermography, and handheld vital microscopy, have been developed to evaluate intestinal perfusion [23–26]. However, these techniques are not yet widely used due to their high cost and technical complexity. ICG fluorescence angiography was first reported to be useful in colorectal surgery by Kudszus et al. [27]. Trastulli et al. reported that using ICG fluorescence angiography led to a significant reduction in AL in colorectal surgery in a meta-analysis of 25 studies [28]. Additionally, considering the low cost of ICG dye in Japan (6$), ICG fluorescence angiography is the most convenient and cost-effective method to evaluate intestinal perfusion.
Regarding AL proportions, we observed a significant reduction in the proportion of AL in the ICG group (4.1%) than in the non-ICG group (12.3%). Previous studies showed that the rates of AL when using ICG fluorescence were 0–9% [15–18, 21, 29–31], similar to our result. Although these studies showed the efficacy of this technique in reducing AL incidence, RCTs could not show this [19, 20]. It has been suggested that RCTs might have some flaws, such as sample size and endpoint selection [32]. Several RCTs are currently ongoing to prove the clinical benefit of routine use of ICG fluorescence [33, 34]. In this study, while operating time was significantly associated with AL incidence after laparoscopic rectal cancer surgery, male sex and tumor location tended to be associated with AL. These factors are consistent with previous reports [10, 35, 36]. Our findings suggest that ICG fluorescence contributes to reducing AL risk in rectal cancer surgery.
In this study, the surgical plan for transection of the proximal colon was changed in 6.8% (5/73) of the patients in the ICG group. Although AL did not occur in these patients, if AL had occurred in all of them, the proportion of AL would have increased to 11%, similar to the proportion of AL in the non-ICG group (12.3%). Previous studies have shown that revisions of the proximal transection line were observed in 3.1 − 20.9% [15–19, 29–31, 37]. Of the patients with revision of the proximal transection line, AL occurred in 0–16.7%, as reported in previous studies [15–18, 30, 31, 37]. Although poor intestinal perfusion is not the only cause of AL, anastomosis with sufficient blood perfusion can contribute to reducing AL. Furthermore, considering the characteristics of patients diagnosed with AL in this study, days from initial surgery to AL occurrence were shorter in non-ICG group (median, 3 days) than in ICG group (3 days). Our findings suggest that ICG fluorescence angiography to evaluate intestinal perfusion is useful for identifying areas with poor vascular perfusion, which may result in early onset of AL, and changes in the surgical plan due to ICG fluorescence visibility could contribute to a safe anastomosis.
This study has several limitations. First, blood perfusion in the distal rectum was not investigated, which might have influenced AL incidence. Second, a selection bias might have occurred because this was a retrospective cohort study that was not randomized or controlled. Third, the ICG dose used in this study was 10 mg. Although no standard dosage of ICG for evaluating intestinal perfusion has yet been established, the visibility of ICG fluorescence might be different due to the height or body weight of patients. Therefore, the results of this study do not provide definitive evidence for the effectiveness of ICG fluorescence imaging in reducing AL. Further multi-institutional, randomized, controlled studies, including ongoing studies, should be planned to verify the definitive benefit of ICG fluorescence in reducing the risk of AL incidence in rectal cancer surgery. However, despite these limitations, we believe that our results are still valuable and applicable because consecutive patients who underwent rectal cancer surgery with anastomosis were assessed and a standard surgical procedure was performed by a single colorectal team at a single institution.
In conclusion, ICG fluorescence angiography was associated with a significant reduction in AL during laparoscopic rectal cancer surgery. Also, changes in the surgical plan due to the visibility of ICG fluorescence may help improve the short-term outcomes of patients with rectal cancer. This study provides valuable insights into the efficacy of ICG fluorescence in patients who have undergone rectal cancer surgery by comparing the rates of AL without ICG fluorescence. Further prospective large clinical trials are warranted to validate the definitive efficacy of ICG fluorescence imaging in reducing AL after laparoscopic rectal cancer surgery.
anastomotic leakage
Indocyanine green
Institutional Review Board
intersphincteric resection
randomized controlled trial
Ethics approval and consent to participate
The study was approved by the Institutional Review Board (IRB) of Kagawa University, Kagawa, Japan (approval number:2021–032). The requirement for informed consent was waived with permission from the IRB due to the retrospective nature of the study.
Not applicable.
The datasets used and/or analyzed during the current study are available from the corresponding author upon reasonable request.
The authors declare that they have no competing interests.
This research did not receive any specific grants from funding agencies in the public, commercial, or not-for-profit sectors.
AK and DF collected and interpreted data. AK, KK, and KO wrote the manuscript. All authors made substantial contributions to the conception and design of the study and were involved in drafting the manuscript and revising it critically for important intellectual content. All authors declare that they contributed to this article, approved the final submitted version, and agreed to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.
The authors would like to thank Nobuyuki Miyatake, M.D., Ph.D., at Kagawa University for their advice and help with statistical analyses. We also thank Editage (www.editage.com) for their editing support.