Journal of Surgery

Introduction

Intraoperative bleeding is not a negligible complication of colorectal surgery. Haemorrhage can be significant and difficult to manage with conventional methods of haemostasis. Severe and not rapidly controlled cases of haemorrhage may result in patient fatality [1-5]. Therefore, appropriate preoperative planning, a meticulous surgical technique and the utilization of haemostatic instruments and devices are considered mandatory to minimize the risk of severe blood loss. Haemostatic control is strongly recommended in colorectal surgery, in which postoperative bleeding ranging between 1% and 14% [6-8]. Although postoperative bleeding is rarely fatal, it could represent the beginning of related complications which significantly influence the rate of postoperative morbidity and mortality, the length of postoperative hospital stay and, consequently, the health care costs [9-11]. In this context, the intraoperative achievement of haemostasis represents a fundamental step. Metallic surgical clips are widely used for this purpose but have disadvantages, including a propensity to slip or act as a nidus for infection [12-15]. This led to the more recent introduction of non-metallic clips with a locking mechanism with tactile and auditory feedback, such as Hem-oLok^® and Absolok^® clips, which represent the two most widely used non-metallic clips. Currently, the two most widely available no-metallic products, are the Hem-o-Lok® and the Absolok^® clips. Hem-o-Lok^® clips are composed of a nonabsorbable, inert, nonconductive and radiolucent polymer that has no interference during X-ray, computed tomography or magnetic resonance imaging. Hem-o-Lok clips have a lock engagement feature, as well as teeth in the jaws that provide good security. Although the use of Hem-o-Lok® clips close to the upper urinary tract (renal collecting system/ureter) is traditionally considered safe, in urological surgery rare cases of complications due to clip migration into other hollow organs have been reported [16-19]. Furthermore, the appearance of a fibrous reaction around the nonabsorbable clip has been described in rare case reports as a radiological finding that may mimic recurrence or granuloma; Even if the true incidence of this phenomenon is currently unknown, it could represent a serious diagnostic challenge in a case previously operated on for colon cancer with a Hem-o-Lok^® clip [20-21]. Absolok^® clips are constructed from bioabsorbable polymers that degrade by hydrolysis over a period of approximately [6-7] months and are characterized by a latch-closure mechanism [22] comparable to that adopted for Hem-o-Lok clips. In the literature, only one nonrandomized study comparing these two different types of non-metallic clips has been published, but it was performed in the context of urological surgery [23], where the clips were used as a preloading device to anchor monofilament or multifilament sutures; thus, the clip was not used directly as a tool to close the vessels. No comparative study between Absolok^® and Hem-oLok^® clips exists in terms of their proper use and in the context of colorectal surgery. The aim of this prospective monocentric randomized study was to perform a comparative analysis of the safety and haemostatic power of Absolok^® and Hem-o-Lok^® clips during colorectal surgical procedures. The secondary aim was to compare Absolok^® and Hem-o-Lok^® clips in terms of operative time and intraoperative (other than intraoperative bleeding) and postoperative outcomes.

Materials and Methods

The study was a monocentric, prospective, randomized, openlabel and pilot clinical trial. All patients who underwent colorectal resection for benign or malignant disease at Isola TiberinaGemelli Isola Hospital during the study period were included in this study. The integral version of the protocol is reported. All patients who underwent colorectal resection for benign or malignant disease surgery from September 26, 2023, to April 11, 2025, at Isola Tiberina-Gemelli Isola Hospital were included in the study. The exclusion criteria were as follows: patients aged < 18 years; inability to give informed consent; emergent surgery; previous colorectal surgery; uncontrollable diabetes mellitus that needs continuous intravenously administered insulin; a history of myocardial infarction or unstable angina pectoris within 6 months; cardiac failure; New York Heart Association (NYHA) III degree; anticoagulant therapy; liver cirrhosis; Child‒Pugh class C; active hepatitis; and chronic renal failure requiring haemodialysis. Exclusion criteria were chosen to minimize major confounding factors for bleeding and postoperative morbidity in a pilot trial. Conditions associated with markedly increased haemorrhagic risk or impaired healing (e.g. anticoagulant therapy, advanced liver diseases, haemodialysis, severe cardiac failure) were excluded to improve internal validity and patient safety. Patients enrolled in the study, without exclusion criteria, were randomized into two groups (1:1 ratio): the experimental group, in which the Absolok^® clip was used, and the control group, in which the Hem-o-Lok^® clip was used. The allocation of the subjects to either the experimental or the control arm used a random sorting randomization algorithm. The allocation sequence was generated by the software PASS 2019. The sub Investigator (VT) and two members of the team (GR and GL) controlled the random allocation sequence. This study was designed as a pilot randomized trial because no comparative data between the two types of clips were available in colorectal surgery, and the expected rate of clinically relevant clip-related bleeding is low. The main objectives of the study were to evaluate the feasibility, safety and estimation of event rates to perform a future adequately powered non-inferiority trial. Therefore, a formal calculation of the sample size was not performed a priori: we planned to enrol a total of 200 consecutive colorectal resections as a pragmatic sample to obtain precise estimates of bleeding and complication rates and their confidence interval for future power calculations. All patients who underwent colorectal resections (categorized as right hemicolectomies, left hemicolectomies and rectal resections) were included in the study, independent of the type of approach to be performed (open, laparoscopic, or robot assisted). All surgical procedures were performed by the same group of colorectal surgeons (Unit of Digestive and Colorectal Unit at Isola Tiberina Gemelli Isola Hospital) experienced in minimally invasive surgery. All procedures were standardized, adopting, in all cases, the same steps, independent of group assignment. The details of surgical procedures performed are reported. During a right hemicolectomy 4 major vessels were clipped: ileo-colic vein and artery, right branch of middle colic vein and artery. During a left hemicolectomy and during rectal resection 2 major vessels were clipped: inferior mesenteric artery and vein.

The primary endpoint of the study was haemostatic performance at the transection site for each clipped vessel, assessed immediately after vessels division using Siegel scale [24]. The comparative analysis between Absolok^® and Hem-o-Lok^® in terms of haemostatic efficiency was performed evaluating the percentage of adverse bleeding events for each vessel clipped and dissected, according to Siegel’s score, which classified the entity of bleeding into four categories [24] as follows:

• Grade 1: no bleeding at the transection site.
• Grade 2: minor bleeding at the transection site; no intervention was needed.
• Grade 3: minor bleeding at transection site, mild intervention needed
• Grade 4: significant bleeding requiring major intervention such as extensive coagulation, ligation or additional haemostatic measures.

The secondary endpoint of the study was the evaluation of whether the clip type could influence the occurrence of intraoperative and postoperative complications. For this purpose, the following outcome variables were evaluated:

• Operative time (min)
• Intraoperative complications
• Time (postoperative day) to flatus
• Postoperative complications within 30 days of surgery were graded according to Clavien’s classification [25].
• Major postoperative complications (if the grade was ≥3 according to Clavien’s classification [25]).
• Length of hospital stay (days)
• Postoperative bleeding within 30 days (any clinical bleeding requiring transfusion, endoscopic/radiologic/surgical intervention or causing haemodynamic instability
• Reoperation within 30 days
• Mortality within 30 days from surgery

Moreover, all adverse events related to the use of each type of clip observed during the study were recorded. Adverse device events were defined as any device malfunction or deficiency, including failure to lock, clip breakage, clip dislodgement/slippage, need for additional clips due to inadequate vessel control, or any intraoperative/postoperative complication judged as directly related to the clip. This trial was approved by the Ethical Committee of Lazio 3 (ID 6090) and was funded by Ethicon (UK). Ethicon (UK) had no role in study design, data collection, analysis, manuscript writing and decision to submit for publication. All the authors declared no conflicts of interest. Informed consent was obtained from all individual participants included in the study. This study is reported in accordance with the CONSORT 2010 guidelines. The trial was registered on the ISRCTN Registry (The UK’s Clinical Study Registry) as ISRCTN79285407 on 05 February 2026 (retrospectively registered). All the variables are presented as descriptive statistics. In depth, the data are reported as absolute and percentage frequencies for qualitative variables which include: sex (male vs female), ASA class, CCI category, indication for surgery (benign vs malignant), surgical approach (open vs miniinvasive), conversion rate, intraoperative complications (Y vs N), postoperative complications (Y vs N, both overall than major), anastomotic leakage (Y vs N), reoperation (Y vs N) and adverse events (Y vs N). Quantitative data distribution was assessed using the Shapiro‒Wilk test. Hence, the data are expressed as the mean, standard deviation (SD) or median and interquartile range (IQR: 25^th–75^th percentile), as appropriate. For the primary endpoint, the difference between Absolok^® and Hem-o-Lok^® in terms of the achievement rate of Grade ≤3 haemostasis for each vessel clipped and dissected (dichotomous variable) was assessed by Fisher’s exact test. The single grading scale was further reported. In the case of qualitative variables, between-group differences were assessed by Fisher’s exact test and the Chi-square test, with Yates correction, as appropriate. Quantitative data were assessed by either Student’s t test or the nonparametric Mann‒Whitney U test, as appropriate. With respect to the secondary endpoints, the potential associations between the type of clip employed and the intraoperative and postoperative courses were computed by using the same test, as previously mentioned. To evaluate differences between the two groups in terms of operation time, logarithmic transformation was applied to reduce the variability of the data distribution, and a t test was applied. A p value <0.05 was considered to indicate statistical significance. All analyses were performed by using R version 4.4.3. The anonymized dataset that supports the findings of this study is available from the corresponding author and principal investigator and sub investigator on reasonable request.

Results

During the study period, 258 patients were assessed for eligibility, and 200 patients (102 women, 51%) were enrolled in the study. The flow diagram of the study is shown in Figure 1. Fifty-eight patients were excluded from the study because they did not meet the inclusion criteria (38; 19%), and 20 declined to participate (10%). Based on the randomization scheme, patients were allocated: 100 in the experimental group using Absolok^® clips (Absolok^® group) and 100 in the control group using Hem-o-Lok^® clips (Hem-oLok^® group). The baseline characteristics of the patients included in the study are reported in Table 1. No differences existed between the two groups regarding sex distribution, age at surgery BMI Charlson Comorbidity Index (CCI) and ASA (Table 1). One hundred forty-nine patients (75%) underwent colorectal resection because of malignant neoplasms (Absolok^® group 76% vs. Hem-oLok^® group 72%; p=0.630), and the most frequent site of colorectal cancer was the right colon (50 patients, 25%). In the context of rectal cancer, 36 patients (18%) had extraperitoneal rectal cancer, and 31 (15.5%) had previously been treated with chemoradiation therapy because of extraperitoneal locally advanced rectal cancer.

Figure 1: Flow diagram of the progress through the phases of a randomised trial of two groups (that is,enrolment, intervention allocation, follow-up, and data analysis).

Fifty-two patients (26%) underwent colorectal resection for benign disease (complicated diverticular disease or inflammatory bowel disease). All surgical procedures (100%) were performed with a minimally invasive approach, and the conversion (minimally invasive to open) rate was 2% (4 patients). In particular, were performed 59 right hemicolectomies (28 in Absolok^® group vs 31 in Hem-o-Lok^® group), 87 left hemicolectomies (45 in Absolok^® group vs 42 in Hem-o-Lok^® group) and 54 rectal resections (27 in Absolok^® group vs 27 in Hem-o-Lok^® group). Overall, 256 vessels were clipped by Absolok^® clips, and 262 vessels were clipped by Hem-o-Lok^® clips. No differences existed between the 2 groups regarding the indications for surgery (p=0.580), type of surgical approach (p=0.999) or conversion rate (p=0.620) (Table 1). With respect to the primary endpoint of the study, 100% of patients in both groups achieved Grade 1 or no bleeding at the transection site (Table 2). The median operative time was 218 minutes (IQR: 178–264.75 minutes) and did not differ between the two groups (219 min, IQR: 127–430 min; in the Hem-o-Lok^® group vs. 216 min, IQR: 120–353 min; in the Absolok^® group; p=0.418). In 7 patients (3.5%), intraoperative complications occurred: in 3 patients, laceration of the colonic serosa occurred with secondary laparoscopic closure of the defect; in 2 patients massive bleeding occurred: 1 of these (Absolok^® group) patients was managed with laparoscopic haemostasis and the other (Hemolok^® group), due to spleen damage, was managed after conversion in open surgery. In both cases, the bleeding was not attributable to clip malfunction, slippage or transection of a major vessel inadequately secured by the study device. In 1 patient, a bladder wall defect was accidentally perforated and was sutured laparoscopically; and in 1 case, after left hemicolectomy, right colon mobilization and distal transverse colon resection were necessary because of ischaemia of the residual colon. No significant differences in the rate of intraoperative complications were detected between the two groups (Table 2).

	Hem-o-lok®	Absolok®	p value
	N=100	N=100
Age, median[25th-75th percentile]	71.0 [37.9-86.6]	69.0 [39.4-86.5]	0.583a
Sex:			0.888b
M, n (%)	50 (50.0%)	52 (52.0%)
F, n (%)	50 (50.0%)	48 (48.0%)
BMI, median[25th-75th percentile]	25.0 [21.0-31.0]	25.0 [19.0-34.6]	0.836a
Comorbidity:			0.165b
No, n (%)	25 (25.0%)	35 (35.0%)
Yes, n (%)	75 (75.0%)	65 (65.0%)
CCI, median[25th-75th percentile]	4.00 [1.00-9.52]	4.00 [1.00-11.0]	0.827a
ASA:			0.085b
≤2, n (%)	84 (84.0%)	73 (73.0%)
>2, n (%)	16 (16.0%)	27 (27.0%)
Indications for colorectal surgery:			0.63
Malignant disease	72 (72.0%)	76 (76.0%)
Bening disease	28 (28.0%)	24 (24.0%)
Site of colorectal cancer			0.58
Right Colon	26 (26.0%)	24 (24.0%)
Left Colon	19 (19.0%)	26 (26.0%)
Rectum	27 (27.0%)	26 (26.0%)
NAD CRT (only for rectal cancer)	15 (15.0%)	16 (16.0%)	0.87
Minimally Invasive Surgery	100 (100%)	100 (100%)	0.999
Types of Surgical Procedures	100	100
RH
LH	31	28
RAR	42	45	0.88
	27	27
N. of clipped vessels	262	256	-
Conversion Rate	1 (1.0%)	3 (3.0%)	0.62

a: non-parametric Mann Whitney U test; b: non-parametric Chi-square test.

Table 1: Baseline characteristics

	Hem-o-lok®	Absolok®	p value
	N=100	N=100
N. of clipped vessels	262	256	-
Adverse Bleeding Events	0 (0%)	0 (0%)	0.999
Adverse Device Events	0 (0%)	0 (0%)	0.999
Operative time, minmedian [25th75th percentile]	219 [127-430]	216 [120-353]	0.418a
Intra-operative complications:			0.999b
No, n (%)	97 (97.0%)	96 (96.0%)
Yes, n (%)	3 (3.00%)	4 (4.00%)
First Flatus Time, daysmedian [25th- 75th percentile]	2 [1-3]	2 [1-4]	0.708c
Postoperative complications:			0.198b
No, n (%)	78 (78.0%)	86 (86.0%)
Yes, n (%)	22 (22.0%)	14 (14.0%)
Major postoperative complications (grade 3 or major according to Clavien class)	5 (5%)	3 (3%)	0.721
Anastomotic Leak			0.999
No, n (%)	91/93 (97.8%)	86/88 (97.7%)
Yes, n (%)	2/93 (2.15%)	2/88 (2.27%)
Postoperative bleeding within 30 days	0 (0%)	1 (1%)	1
Reoperation within 30 days			0.716
No, n (%)	93 (96.9%)	88 (95.7%)
Yes, n (%)	3 (3.12%)	4 (4.35%)
Length of hospital stay, daysmedian [25th-75th percentile]	5.00 [4.00-12.5]	5.00 [4.00-24.7]	0.118c

a: T test applied to logarithmic transformation of data; b: non-parametric Chi-square test; c: non-parametric Mann Whitney U test.

Table 2: Primary and secondary endpoints.

Overall, 36 (18%) patients experienced complications within 30 days of surgery, but only 8 (4%) patients experienced complications with a grade ≥3 according to the Clavien classification. The 30-day overall postoperative complication and major postoperative complication rates did not differ between the Absolok^® group and the Hem-o-Lok^® group (14% vs. 22%; p=0.198). The overall rate of anastomotic leakage was 2.1% (2.3% in the Absolok^® group vs. 2.2% in the Hem-o-Lok^® group; p=0.999), and the rate of reoperation within 30 days to treat postoperative complications was 4% (4.3% in the Absolok^® group vs. 3.1% in the Hem-o-Lok^® group; p=0.716). The overall median time to first flatus was 2 (IQR: 1–3) days and did not differ between the Absolok^® (median=2 days; IQR: 1–4 days) group and the Hem-o-Lok^® group (median=2 days; IQR: 1–3 days; p=0.708). The median postoperative length of stay (overall: 5 days; IQR: 5–6 days) did not differ between the two groups (Absolok^® group: 5 days; IQR: 4–24.7 days vs. the Hem-o-Lok^® group: 5 days; IQR: 4–12.5 days; p=0.118). None of the analysed patients in the study died within 30 days of surgery (postoperative mortality rate: 0%). Fourteen patients (7%) experienced adverse events, with an equal distribution between the two groups (7% in both the Absolok^® group and the Hem-oLok^® group; p=0.999) (Table 2). Surgeons noted no differences in the handling of the clip applicators, whether for open or laparoscopic techniques; no specific application issues were noted between the two clips.