Single-stapled versus non-single-stapled colorectal anastomosis in minimally invasive anterior resections: a systematic review and meta-analysis

Abstract

Background

Anastomotic leak remains a major cause of morbidity after anterior resection for rectal cancer. The single-stapled technique (SST) has been proposed to eliminate crossover of intersecting staple lines in conventional non-single-stapled techniques (NSST). However, evidence comparing the two approaches in minimally invasive anterior resections remains limited.

Methods

A systematic review and meta-analysis was conducted according to PRISMA 2020 guidelines (PROSPERO CRD420251053461). Randomised and observational studies comparing SST versus NSST in adult minimally invasive anterior resections were included. Primary outcomes were anastomotic leak and length of hospital stay. Secondary outcomes included operative time, postoperative morbidity (Clavien–Dindo ≥ III), and mortality. Random-effects models were used to pool effect estimates.

Results

Five observational studies (n = 1647; SST = 587, NSST = 1060) met inclusion criteria. Pooled analysis demonstrated that SST significantly reduced anastomotic leak rates compared with NSST (OR 0.43; 95% CI 0.26–0.70; p = 0.0007; I² = 23%). Length of stay was shorter with SST (SMD − 0.43; 95% CI − 0.75 to − 0.11; p = 0.008; I² = 88%). Operative time, blood transfusion requirements, and major postoperative complications did not differ significantly. Mortality was zero in all reporting studies.

Conclusions

Single-stapled anastomosis in minimally invasive anterior resection is associated with a lower pooled risk of anastomotic leak compared with non-single-stapled techniques. Despite heterogeneity in secondary outcomes, SST appears to be a promising technical refinement; however, randomised trials are required before routine adoption.

Supplementary Information

The online version contains supplementary material available at https://doi.org/10.1007/s10151-026-03380-y.

Highlights

• This is the first systematic review and meta-analysis dedicated solely to comparing single stapled and non-single-stapled colorectal anastomosis in minimally invasive anterior resections.

• Single-stapled anastomosis is associated with an approximate 50% reduction in anastomotic leak compared with conventional techniques.

• Length of hospital stay was shorter with single-stapled anastomosis, while operative time and major complications were comparable between techniques.

• Findings highlight a promising technical refinement that may improve anastomotic integrity in laparoscopic and robotic rectal surgery.

• Results support the need for multicentre randomised trials to confirm benefits and define standardised application of the single-stapled technique.

Supplementary Information

The online version contains supplementary material available at https://doi.org/10.1007/s10151-026-03380-y.

Introduction

Anastomotic leak remains one of the most serious complications following anterior resection for colorectal disease. Despite ongoing refinement in minimally invasive techniques, leak rates continue to range from 3% to 15% in contemporary series, influenced by patient comorbidities, tumour characteristics, neoadjuvant treatment, and operative technique [1–4]. Anastomotic failure contributes to substantial morbidity, reoperation, prolonged hospitalisation, stoma formation, and impaired long-term oncological outcomes [5, 6].

Stapling devices are now integral to colorectal surgery, with linear and circular staplers enabling several anastomotic configurations. The conventional non-single-stapled technique (NSST) divides the rectal stump with a linear stapler before applying a circular stapler, creating intersecting staple lines. These “dog-ears” may compromise local blood flow and increase leak risk [7, 8].

The single-stapled technique (SST) was developed to eliminate staple-line intersections. By closing the dog-ears with a purse-string suture and creating only one final circular staple line, SST may reduce focal ischaemia and promote improved healing. Adoption has increased in minimally invasive anterior resection—both laparoscopic and robotic—particularly for mid- and low-rectal cancers [9–12]. Observational studies have suggested lower leak rates and shorter recovery with SST, although they vary in definitions, inclusion criteria, and perioperative protocols.

Given these differences, a synthesis of contemporary evidence is needed. This systematic review and meta-analysis compared SST with NSST in minimally invasive anterior resection, evaluating anastomotic leak, operative outcomes, postoperative morbidity, and length of hospital stay.

Methods

This review adheres to the guidelines outlined in the Preferred Reporting Items for Systematic Reviews and Meta Analyses (PRISMA) 2020 statement [13]. Furthermore, it has been registered on PROSPERO under the registration number PROSPERO 2025 CRD420251053461.

Studies published in or translated to the English language from their inception until August 1, 2025 were included in the search. The initial criteria for inclusion involved randomised controlled trials (RCTs) and observational studies that compared SST to NSST, for elective adult anterior resections. Systematic reviews were excluded from the primary analysis, but primary studies included within these systematic reviews were considered for inclusion if they met the predefined inclusion criteria. Only studies focusing on adult anterior resections for elective patients and involving a stapled anastomosis were eligible for inclusion in this review. The exclusion criteria included no animal/cadaveric studies, non-English studies, paediatric population, palliative surgery, metastatic rectal cancer disease, faecal incontinence preoperatively, and non-stapled anastomosis.

The primary outcomes of interest encompassed anastomotic leak rates and length of stay in hospital. Secondary perioperative outcomes included the operative times, need for blood transfusions, and 30-day postoperative complication rates with Clavien–Dindo classification score ≥ 3. It is important to note that all these outcomes were assessed at the endpoint of the respective follow-up periods for the included studies, unless otherwise explicitly specified.

Electronic database searches were carried out in Ovid Medline, Embase, Cochrane Central Register of Controlled Trials (CENTRAL), and the Cochrane Database of Systematic Reviews (CDSR). These searches were conducted by a senior information specialist from the library department of the Royal College of Surgeons of England on July 3, 2023. Furthermore, an additional search was undertaken through snowballing, referencing relevant articles. A final review was conducted before concluding the literature search on August 15, 2025.

The primary aim of the study was to assess whether SST, in elective adult anterior resections for rectal cancer surgical patients, confers any benefits over NSST in reducing anastomotic leak rates. To guide our search comprehensively, we employed the Patient-Intervention-Control-Outcome (PICO) framework, as outlined in Supplementary Table 1. The thorough search strategy is depicted in Supplementary Table 2a, b.

Abstract screening and full text review was performed by two independent blinded reviewers [SJ and RA], with conflict resolution by a third senior reviewer [RL], to generate a final list of eligible studies for inclusion in the meta-analysis. Demographic, clinical and outcome data in both treatment arms from individual studies were extracted by two independent researchers [MI and SJ] and cross-checked with another independent researcher for adequacy and accuracy [RA].

Observational studies were categorized following the criteria established by Mathes and Pieper [14]. The assessment of the risk of bias in observational studies utilised the Joanna Briggs Institute (JBI) assessment tool [15]. In this case with the JBI appraisal tool, the overall risk of bias for a specific study was determined by the number of questions answered with “yes”, “no”, or “unclear”. Studies were considered to have a low concern of bias if there was an unfavourable answer to one question or fewer, a moderate concern if there were unfavourable answers to two to three questions, and a high concern if four or more questions received unfavourable answers.

Statistical analysis

Categorical variables were summarised as counts and ratios; continuous data were reported as mean (SD). When studies reported medians with IQRs or ranges, means and SDs were estimated using the method reported by Wan et al. [16].

Meta-analyses for categorical variables used risk ratios (RR) or odds ratios (OR) with 95% CIs. Continuous outcomes were synthesised using standardised mean difference (SMD). Heterogeneity was evaluated using the I² statistic, with I² > 50% considered significant. Random-effects models (Mantel–Haenszel) were used when heterogeneity was present [17].

Handling of confounding factors

Patients and disease characteristics were highlighted and compared in SST and NSST comparisons.

Operational definitions

SST

Linear stapler division of the rectum, followed by purse-string closure of dog-ears and creation of a single circular staple line. Alternatively a combined transabdominal and transperineal approach could be done where suitable for a TTSS approach. However, the mechanics of both are different. The type of suture used to invert the dog-ears was not specifically looked at as they all yielded the same type of configuration, achieving a circular staple line only without any intersections.

NSST

One or more linear stapler firings with subsequent circular stapling, leaving intersecting staple lines.

Results

Study selection and characteristics

The study selection process is summarised in the PRISMA diagram (Fig. 1). Eighty-seven records were identified on the initial search after excluding eight duplicates. Out of 12 studies that were eligible for full text review, six studies were excluded for either not fulfilling the study question criteria, no comparison performed, incorrect study intervention, and incorrect publication type to produce a final list of six studies.

Fig. 1.

PRISMA chart

Six studies met the inclusion criteria all of which were observational studies. Two case–control [7, 8] and four cohort studies [18–21] were included. A total of 1647 participants at the study level were included with 587 in the interventional group and 1060 in the control group. The included study characteristics are summarised in Table 1.

Table 1.

Study characteristics

Category	Study	Raju 2025	Foppa 2023	Harji 2023	Spinelli 2021	Saurabh 2017	Kim 2013
Study type		Retrospective cohort study	Retrospective cohort study	Prospective cohort study	Prospective cohort study	Retrospective case-matched study	Retrospective case–control study
Inclusion criteria		Adults 18–80 years, benign disease and malignant, NOSE LAR	Adults > 18 years, MRI-defined low rectal cancer, stapled anastomosis < 5 cm from AV done MI-TME	Adults > 18 years, TME for rectal cancer, anastomosis within 6 cm of anal verge	Consecutive mid-low rectal cancer (< 10 cm from anal verge) undergoing elective laparoscopic TME with stapled anastomosis and protective stoma, single surgeon adopting either DS, TaTME or TTSS techniques	BMI ≤ 30, ASA I–III, elective LAR, tumour 10–40 cm from AV, ≤ 5 cm diameter, T1–T3NxM0	Patients with rectal cancer undergoing laparoscopic low anterior resection with TME
Exclusion criteria		Prior major pelvic surgery, inability to close rectum with linear stapler	Nonrestorative procedures, manual anastomosis, open conversions, total neoadjuvant (to reduce intergroup heterogeneity)	Intersphincteric or delayed CAA	Concurrent or previous pelvic malignancies, stage 4 tumours, any type of manual anastomosis	Tumour > 5 cm in diameter, BMI > 30, T4 or metastatic disease, emergency surgery, ASA IV	BMI > 30, T4 tumours, bulky specimens
Comparisons		SSA vs DSA	SST (transanal rectal transection and double purse-string SS anastomosis) vs DST (TA MI linear stapled rectal transection and DS anastomosis)	TTSS vs DST and TTSS vs CAA	TTSS vs TaTME vs DS	SST with NOSE vs DST with conventional extraction	SST vs DST
Total population		179 (40 SSA, 139 DSA)	643 (185 SST, 458 DST)	275 (70 TTSS, 110 DST, 95 CAA)	277 (127 DS, 100 TaTME, 50 TTSS)	188 (82 SST-NOSE, 106 DST)	180 (60 SST, 120 DST)
Study period		2011–2022	January 2010–January 2022	January 2016–April 2021	January 2015–August 2020	January 2012–April 2015	July 2007–April 2010
Follow-up period		Immediate postoperative outcomes	90-day postoperative outcomes	12-month LARS scores and 30-day leakage	Long-term outcomes	Immediate postoperative outcomes	Short-term outcomes only
Outcomes	Primary endpoint	SSA had no leaks but higher intra-op redo rate; DSA had 6 leaks; similar LOS & time	90-day AL (International study group of Rectal Cancer definitions and severity grading system)—Clinical and radiological leaks. SST had significantly lower leak rate (6.48% vs 15.28%), shorter stay, fewer complications	30-day anastomotic leak (ISGRC). TTSS had lower leak rate vs DST (8.6% vs 20.9%), better function vs CAA, no safety issues	Rate of anastomotic leak (International study group of Rectal Cancer)—both clinical and radiological. Postoperative ileus defined as postoperative prolonged > 3 days nil per oral or need for nasogastric tube insertion	Anastomotic leaks, anastomotic bleed rates, LOS. SST-NOSE had less pain, faster recovery, similar leak rate (2.4% vs 2.8%), fewer infections	SST had longer distal resection margin, less pain, similar leakage and morbidity vs DST
	Secondary endpoints		CD classifications (highest rank recorded)	Postoperative complications (30-day CD), grade of complications, reoperation rate (within 30 days of index operations), LOS and functional outcome (LARS score)	Postoperative complications as per Clavien–Dindo classification. (30-day CD was not different, but 90-day CD was significantly higher in DS group.) LOS was not different among groups	Operative time, first bowel movement, readmission rates, postoperative complications
Limitations		Extended period of data collection, outcomes may have been influenced over time by experience and skills, variability in surgical indications, confounding factors not standardised	Temporal gap between 2 techniques, experience of surgeons leading to interoperative variability, improvement of perioperative protocols, single-institution	Selection bias in the choice of anastomotic technique based on surgeon preference and patient characteristics	Series from an expert colorectal surgeon and data about reproducibility is not yet available. Monocentric and retrospective design creates a potential selection bias. All DS cases were performed laparoscopically. Learning curve remains to be assessed but should be shorter for an expert rectal surgeon if coming from TaTME	Single-centre study, recruitment bias as 3 senior surgeons participated, level of evidence is not as high as RCTs, no evaluation of patient satisfaction or cosmetic aspect, exclusion criteria as possible contraindications for SST-NOSE-LAR procedure	Single surgeon, retrospective study, selection bias, aggressive T stage tumour with DST, excluded bulky tumours, lack of estimation of anal sphincter function
Strengths		–	Strict inclusion criteria leading to homogenous population, clear internationally recognised definitions of AL (ISGRC) and RC, use of MRI-defined criteria for LOREC	Robust evidence reporting system, well defined and robust definitions, standardised definitions	Strict criteria for defining AL and the data was collected from a prospectively maintained database
Low RC definition			Entirely or partially located below the origin of the levator ani muscle on the pelvic bone (Pelican cancer foundation in LOREC program)		Mid-low RC was defined as < 10 cm from the anal verge
Neoadjuvant CRTx			Locally advanced stage II (T3/4N0M0) or stage III (any TN + M0)—50 Gy + capecitabine	Exact neoadjuvant treatment was not mentioned	Stage II: T3/4N0M0 or Stage III: any TN + M0—50 Gy + capecitabine		T3/4N + or extraperitoneal disease—50 Gy in #25 for 5 weeks
Restaging			8 weeks	Not mentioned	Restaging about 8 weeks after treatment and surgery scheduled between 10 and 12 weeks		Surgery performed 6–8 weeks after LC-CRT
Recommendations			Prospective multicentre trial—to confirm short-term outcomes, long-term and functional outcomes	Future trial using the IDEAL framework from the Blencowe framework for standardising the technique. Stage 3 RCT	Prospective multicentre trial to scrutinize the functional outcomes		Further studies on larger scale to assess this technique and validate the approach. Active evaluation and survey of anal function needed to avoid possibility of anal sphincter injury

AL anastomotic leak, ASA American Society of Anesthesiologists, AV anal verge, BMI body mass index, CAA coloanal anastomosis, CD Clavien–Dindo classification, CRTx chemoradiotherapy, DS double-stapling (technique), DSA double-stapled anastomosis, DST double-stapling technique, Gy gray, IDEAL Idea, Development, Exploration, Assessment and Long-term Study framework, ISGRC International Study Group of Rectal Cancer, LAR low anterior resection, LARS low anterior resection syndrome, LC-CRT long-course chemoradiotherapy, LOREC Low Rectal Cancer Development Programme, LOS length of stay, MI minimally invasive, MRI magnetic resonance imaging, NOSE natural orifice specimen extraction, NxM0 regional lymph nodes cannot be assessed, no distant metastasis, RCT randomised controlled trial, RC rectal cancer, SSA single-stapled anastomosis, SST single-stapling technique, TaTME transanal total mesorectal excision, TME total mesorectal excision, TTSS transanal transection and single-stapling technique

The principal patient populations were adults undergoing minimally invasive anterior resection (laparoscopic or robotic) for benign and malignant disease. Across studies the majority of patients were male with no significant differences between the studies and groups (p = 0.48), and mean ages were different between groups (p = 0.03) (no clinically important differences were noted in the study-level extracts). No significant differences were noted between the two groups in terms of the body mass index (BMI) of the patients as well (p = 0.26). Four studies [7, 8, 19, 21] included patients with ASA ≥ 3 and recorded a history of previous abdominal surgeries showing no significant differences (p = 0.74 and 0.09 respectively). See Table 2 and Supplementary Figs. 1, 2 for more details.

Table 2.

Patient population characteristics

		Raju 2025			Foppa 2023			Harji 2023
		SST	NSST	p-value	SST	NSST	p-value	SST	NSST	p-value
Population		40	139		185	458		70	110
Age, mean ± SD (range)		60.9 (34–80)	59.5 (20–88)	0.53	62.77 ± 12.36	64.67 ± 11.86	0.069	64 (36–87)	65 (30–86)	0.369
Male gender (%)		15 (8.4%)	57 (31.8%)	–	123 (67)	267 (58)	0.066	43 (61%)	73 (66%)	0.306
BMI, mean (range) kg/m2		27.9 (22–38)	29.1 (16.5–46.7)	0.16	24.80 ± 3.52	25.05 ± 4.26	0.475	–	–	–
ASA	I	–	–	–	65 (35%)	119 (26%)	0.068	–	–	–
	II	–	–		99 (53%)	286 (62%)		–	–
	III	–	–		18 (10%)	50 (11%)		–	–
	IV	–	–		3 (2%)	3 (1%)		–	–
Smoking history	Active	–	–	-	32	85	0.03	–	–	–
	Ex-smokers	–	–		27	88		–	–
	Non-smokers	–	–		126	285		–	–
Previous abdominal surgery		–	–	–	74	208	0.21	–	–	–

		Spinelli 2021			Saurabh 2017			Kim 2013
		SST (n = 150)	NSST (n = 127)	p-value	SST	NSST	p-value	SST	NSST	p-value
Population		150	127		82	106		60	120
Age, mean ± SD (range)		63 ± 12	64.65 ± 11.47	0.606	63.3 ± 13.9	64.7 ± 10.9	–	59.7 (25–84)	62.9 (38–88)	0.115
Male gender (%)		70.6	53.5		47 (61.3%)	65 (57.3%)	–	26 (43.3%)	57 (47.5%)	0.636
BMI, mean (range) kg/m2		25 ± 3	25.34 ± 4.38		24.4 ± 4.2	24.4 ± 3.2		23.4 (17.6–29.7)	23.9 (16.8–31.9)	0.3
ASA	I	45	30	0.494	59	75	–	29 (48.3%)	53 (44.2%)	0.535
	II	86	77					28 (46.7%)	64 (53.3%)
	III	19	20		23	31		3 (5%)	3 (2.5%)
	IV	Excluded	Excluded		–	–		–	–
Smoking history	Active	25	26	0.513	–	–	–	–	–	–
	Ex-smokers	22	26		–	–		–	–
	Non-smokers	103	75		–	–		–	–
Previous abdominal surgery		64	65	0.239	9	12	–	11	20	0.835

Fig. 2.

a Anastomotic leaks. b Operative times. c Length of hospital stay

Operative approaches and surgeon experience varied between series. The case mix was variable with the majority of the procedures done as a minimally invasive surgery with only one study by Harji et al. [20] showing the need for an open surgery in 11 patients (SST n = 3 vs NSST n = 8). Most of the patients in the reported studies underwent an anterior resection for malignancy except the study by Raju et al. [18] which does not clearly mention the number of resections for cancer. Defunctioning ileostomy formation was reported clearly in only four studies [7, 19–21] with only two patients in the NSST group in the study by Kim et al. [7] and zero in the SST group, whereas all patients in the study by Spinelli et al. had a defunctioning ileostomy created. For more details, see Table 3.

Table 3.

Tumour and operative characteristics

		Raju 2025			Foppa 2023			Harji 2023			Spinelli 2021			Saurabh 2017			Kim 2013
		SST (n = 40)	NSST (n = 139)	p value	SST (n = 185)	NSST (n = 458)	p value	SST (n = 70)	NSST (n = 110)	p value	SST (n = 150)	NSST (n = 127)	p value	SST (n = 82)	NSST (n = 106)	p value	SST (n = 60)	NSST (n = 120)	p value
T	0	–	–	–	31	54	–	–	–	0.358	18	19	0.329	7	11	–	–	–	0.421
	1	–	–		36	66		11	18		29	25		7	6		16	30
	2	–	–		46	99					41	25		35	41		17	24
	3	–	–		73	213		59	92		58	49		40	59		27	64
	4	–	–		5	26					4	9		–	–		0	2
N	0	–	–	–	127	312	–	15	28	0.901	103	88	0.629	29	36	–	–	–	–
	1	–	–		45	106		52	79		46	24		41	52		–	–
	2	–	–		13	40					11	15		11	18		–	–
M	0	–	–	–	175	412	–	57	99	0.105	–	–	–	82	106	–	–	–	–
	1	–	–		10	30		13	11		–	–	–	–	–		–	–
Surgical approach	Robotic	–	–	–	–	–	–	13 (19%)	40 (36%)	–	–	–	–	–	–	–	37 (61.7%)	4 (3.3%)	–
	Laparoscopic	40	139		185	458		54 (77%)	62 (56%)		–	–	–	82	106		23 (38.3%)	116 (96.7%)
Defunctioning ileostomy		0	2	–	178	442	–	54	79	0.001	127	150	–	–	–	–	0	2	0.553
Neoadjuvant therapy		–	–	–	140	335	0.508	51	57	0.003	95	76	0.748	–	–	–	5	12	0.724
CRM positive		–	–	–	4	6	0.484	–	–	–	–	–	–	–	–	–	59	116	0.666
Distance from AV (cm), median (range)		–	–	–	5 (3–6)	5 (4–6)	0.983	7 (3–12)	9 (6–13)	< 0.001	5.5 ± 2.3	6.2 ± 1.86	0.598	–	–	–	7.8 (3–12)	7.9 (4–12)	0.734
Type of anterior resection	HAR	29	102	–	–	–	–	–	–	–	–	–	–	–	–	–	–	–	–
	LAR	10	35		185	458		–	–		–	–	–	82	106		–	–

AV anal verge, CRM circumferential resection margin, HAR high anterior resection, LAR low anterior resection, NSST non-single-stapled technique, SST single-stapled technique

Risk of bias/study quality

Most studies were non-randomised and therefore prone to confounding by indication and selection bias; reporting of blinding, consecutive case inclusion and follow-up was variable across reports. The overall risk of bias of the assessment was either moderate or high and hence a high-quality RCT would be beneficial in future. For more detailed assessment using the JBI tool, see Supplementary Tables 3a, b.

Outcomes

Outcomes are summarised in Table 4.

Table 4.

Outcomes measured

			Raju 2025			Foppa 2023			Harji 2023			Spinelli 2021			Saurabh 2017			Kim 2013
			SST (n = 40)	NSST (n = 139)	p value	SST (n = 185)	NSST (n = 458)	p value	SST (n = 70)	NSST (n = 110)	p value	SST (n = 150)	NSST (n = 127)	p value	SST (n = 82)	NSST (n = 106)	p value	SST (n = 60)	NSST (n = 120)	p value
Operative time			228 ± 52.87	275 ± 74.63	0.68	283.4 ± 73.97	281.29 ± 87.82	0.782	–	–	–	293 ± 69	277 ± 78		227.9 ± 55.8	218.1 ± 57	0.304	203.9 (120–400)	167.6 (90–300)	< 0.001
Length of stay			5.25 ± 1.16	21.5 ± 13	0.29	11 ± 17.9	52.7 ± 110.1	< 0.0001	13.5 ± 7.6	17.25 ± 9.7	0.695	5	5	0.791	4.8 ± 3.4	5.9 ± 2.4	0.025	7.7 (4–27)	8 (5–33)	0.622
Total anastomotic leaks			0	6	0.34	12	70	0.002	6	23	0.028	7	22	0.002	2	3		6	10	0.711
Complications	Total		0	6	–	42	178	–	42	43	0.79	40	53		6	13	–	9	23	0.31
	Intraoperative		6	5	–	2	5	0.99	–	–	–	2	1	0.778	–	–	–	–	–	–
	30 day	Total	–	–	–	34	136	0.003	42	43	–	18	21	0.112	–	–	–	–	–	–
		CD1	–	–		5	23		31	26		1	1		–	–		–	–
		CD2	–	–		16	56					11	10		–	–		–	–
		CD3a	–	–		5	14		11	17		2	3		–	–		–	–
		CD3b	–	–		7	42					3	7		–	–		–	–
		CD4	–	–		1	1					0	0		–	–		–	–
LARS score	No LARS (< 20)		–	–	–	–	–	–	35	71	< 0.001	–	–	–	–	–	–	–	–	–
	Minor (21–29)		–	–		–	–		4	2		–	–		–	–		–	–
	Major (30–42)		–	–		–	–		2	5		–	–		–	–		–	–
Blood-transfusions	Intraoperative		–	–	–	1	1	0.493	–	–	–	1	0	0.978	1	0	–	0	1	1
	Postoperative		–	–	–	1	11	0.194	–	–	–	0	2		–	–	–	–	–	–
Complications	Paralytic ileus		–	–	–	3	11	–	–	–	–	–	–	–	1	3	–	0	2	0.315
	Anastomotic bleed		–	–	–	1	5	–	–	–	–	–	–	–	1	1	–	–	–	–
	Pneumonia		–	–	–	1	1	–	–	–	–	–	–	–	0	1	–	–	–	–
	Urinary retention		–	–	–	3	7	–	–	–	–	–	–	–	–	–	–	0	1	0.478
	Anaemia requiring transfusion		–	–	–	1	11	–	–	–	–	–	–	–	–	–	–	–	–	–
	Wound complications		–	–	–	3	15	–	–	–	–	–	–	–	0	4	–	0	6	0.078
Mortality			–	–	–	–	–	–	–	–	–	–	–	–	0	0	–	0	0	–

LARS low anterior resection syndrome, NSST non-single-stapled technique, SST single-stapled technique

Primary outcomes

Anastomotic leak

A random-effects meta-analysis of all the included studies [7, 8, 18–21] in the meta-analysis produced the pooled estimate that odds ratio (OR) for anastomotic leak (SST vs NSST) was 0.43 (95% CI 0.26–0.70; p = 0.0007) meaning that SST was associated with an approximately 43% relative reduction in reported anastomotic leak (Fig. 2a); and low statistical heterogeneity I² = 23% indicating consistency of direction of effect across the included series. A subset analysis of the techniques to yield a single complete circular stapled anastomosis without TTSS was also done which showed a pooled leak rate reduction of 50% (OR 0.50; 95% CI 0.30–0.82; p = 0.006).

Kim et al. [7] had reported a higher leak rate with SST than seen in any of the other studies included here which is likely due to two reasons: first, it was the first study population and the surgeons were still learning how to master this new technique and the rates started to decrease over a period of time; second, they had a higher risk due to higher proportion of patients with previous abdominal surgeries, therefore increasing the difficulty of the surgeries as well.

Ninety-six percent of patients in both groups in the study by Foppa et al. [19] had a defunctioning ileostomy created despite which the anastomotic leak rates were significantly higher in this study than the other studies, reinforcing that an ileostomy is not protective. However, the malignancies in this study were more advanced in both groups compared to other studies which could be a potential contributor due to higher neoadjuvant rates and local reaction of the cancer itself. Raju et al. [18], on the other hand, showed zero anastomotic leaks in the SST group which is due to the rather higher number of pathologies being benign and more refined experience of the surgeons.

Secondary and other outcomes

Operative time

The overall operative time (mean ± SD) was available for five studies [7, 8, 18, 19, 21] and it showed no significant differences between the two groups on pooled analysis using standard mean differences (SMD 0.05; 95% CI − 0.23 to 0.33; p = 0.75) and a high heterogeneity was noted I² = 83% (Fig. 2b). The operative times in the study by Kim et al. [7] showed that the new interventional technique SST took longer, while the operative times were lower in the SST group in the study by Raju et al. [18]. These differences are again attributable to the fact that the experience of the surgeons over a period led to faster operative times and the number of patients with previous abdominal surgeries was lower.

Length of hospital stay

Length of hospital stay (days) was recorded in all studies [7, 8, 18–21] and SST showed a significantly shorter length of hospital stay compared to NSST (SMD − 0.43; 95% CI − 0.75 to − 0.11; p = 0.008) (Fig. 2c), although the spread of the data was very heterogeneous with I² = 88%. The length of hospital stay was significantly heterogeneous in the study by Foppa et al. [19] with potential outliers as well skewing the data which is likely because there were more anastomotic leaks also reported in the NSST group in the study. Therefore, the findings of this particular outcome will have to be treated cautiously.

Mortality

Mortality was recorded only by two studies [7, 8] which both showed zero mortalities. The other studies did not record any data on mortalities.

Other reported outcomes

Some other outcomes of note reported by a small subset of studies but not enough data for performing a pooled analysis are described below:

• Anastomotic bleeds: This outcome was reported by only two studies [8, 19] with Foppa et al. having a higher rate of bleeding in the NSST group (n = 5) compared to n = 1 in the SST group and it also showed the need for a higher number of patients needing blood transfusions either intra- or postoperatively.

• Paralytic ileus: Was reported by three studies [7, 8, 19] which showed that NSST had a an almost fourfold higher incidence of paralytic ileus compared to SST (n = 16 vs 4) and would also contribute to longer length of hospital stays with NSST as seen above.

• Other respiratory and wound problems were again reported variably but consistently showed higher incidences with NSST compared to SST.

• Only the study by Foppa et al. [19] reported on the LARS score which was significantly lower in the SST group.

Sensitivity and subgroup considerations

• Sensitivity analysis removing the smallest study with zero events in SST [18] did not materially alter the pooled OR for leak (direction remained protective for SST), though confidence intervals widened slightly.

• Heterogeneity was low for the leak outcome but very high for continuous outcomes (operative time, LOS) and for total complications; this likely reflects differences in case-mix (benign vs malignant indications), perioperative care pathways (ERAS utilisation), drain/stoma policies and definitions/timing of outcome ascertainment.

Discussion

Main findings

This systematic review and meta-analysis synthesising data from five observational cohorts suggests that SST is associated with a significantly lower rate of anastomotic leak observed compared with NSST in minimally invasive anterior resections. The pooled odds ratio indicated nearly a 43% relative risk reduction (OR 0.43; 95% CI 0.26–0.70), with minimal heterogeneity (I² = 23%). A subset analysis done excluding the TTSS technique showed that the OR was different and the mechanics of the two techniques are variable. However, this current available data does not allow us to directly make a comparative conclusion. Secondary analyses revealed a shorter hospital stay in SST cohorts, although findings for operative time and overall morbidity were heterogeneous.

Comparison with existing evidence

Anastomotic leak rates after low anterior resection remain as high as 15% despite advances in technique and perioperative optimization [15]. The theoretical rationale for SST—eliminating intersecting staple lines and associated dog-ear ischaemia—has been supported by experimental perfusion studies demonstrating improved microvascular flow at the staple line compared with double-stapled configurations [22, 23]. Recent clinical reports by Foppa et al. (2023) [19] and Harji et al. (2023) [20] similarly found lower leak rates with SST in laparoscopic and robotic anterior resections, corroborating the present pooled effect.

Earlier comparative work in open surgery yielded mixed results [7, 8], likely reflecting older stapling technology, variable definitions of leak, and absence of enhanced recovery protocols. The present findings therefore extend current knowledge by demonstrating a consistent protective association in the minimally invasive era. Importantly, the observed low heterogeneity strengthens the credibility of the leak outcome despite the non-randomised design of included studies.

Potential mechanisms

By producing a single staple line, SST avoids crossing linear and circular staple intersections, potentially preserving circumferential perfusion and reducing focal stress concentration [18]. This technical advantage could translate into improved anastomotic integrity. Moreover, SST facilitates precise tissue alignment and minimizes manipulation in the narrow pelvic space of minimally invasive rectal surgery [24]. However, the success of SST remains operator-dependent; inadequate purse-string closure or poor visualization during transanal stapler introduction may offset these benefits.

Clinical implications

The present data provide a signal that SST may reduce anastomotic leak after minimally invasive anterior resection. However, given the observational nature of the evidence and the heterogeneity in secondary outcomes, the findings should be considered hypothesis-generating rather than definitive. Should these results be validated, SST could represent a practical refinement of minimally invasive anterior resection, potentially lowering leak-related morbidity and associated reoperation rates. Nonetheless, patient selection remains essential—particularly in irradiated, obese, or ultra-low rectal cases where technical exposure may limit applicability. They should continue to adhere to core anastomotic principles and use diverting stomas selectively. SST should therefore be viewed as an adjunct to, not a substitute for, meticulous surgical technique, adequate perfusion, and tension-free anastomosis.

Technical considerations and standardisation

The practical application of the single staple line (SSL) technique is heavily influenced by anatomical constraints and patient factors, which must be considered to ensure successful adoption. Applying a purse-string or a bunching stitch to the distal rectal stump after specimen extraction is straightforward for upper rectal cancers where the anastomosis lies at or above the peritoneal reflection, particularly in open surgery. However, technical difficulty increases significantly in the distal third of the rectum, especially in the context of a narrow male pelvis where the sacral curvature and the rigid prostate create a challenging angle of approach. In such cases, the prolapse of seminal vesicles or pelvic side wall fat into the operative field after excision can further hinder articulation of the circular stapler. Conversely, the distal rectum is often more accessible in women and in patients with a low BMI. While some surgeons advocate reinforcing the linear staple line with interrupted or continuous sutures, this manoeuvre is often impractical in the lowest third of the rectum where pelvic access is most constrained. Emerging approaches such as transanal total mesorectal excision (TaTME) and transanal transabdominal surgery (TTSS) can facilitate anastomotic construction in this region, but they require additional perineal setup and a dedicated skillset.

Standardisation of the SSL technique and a structured approach to the learning curve are essential for safe dissemination. In our experience, the SSL technique—specifically the placement of a purse-string or omega stitch to incorporate the lateral dog-ears of the linear staple line—can be reliably performed via an abdominal approach in both laparoscopic and robotic low rectal cancer surgery, provided the surgeon has adequate familiarity with minimally invasive pelvic dissection. We have previously demonstrated the technical details of this manoeuvre to facilitate single-staple anastomosis [25]. For surgeons adopting this method, we recommend initiating practice with high anterior resections where pelvic access is less restrictive. In our institutional experience, proficiency in applying the stitch in the low rectum was achieved within three cases, suggesting a short learning curve for surgeons already competent in advanced rectal cancer surgery. This progression aligns with the IDEAL framework for surgical innovation, where technical refinement and operator proficiency are critical precursors to widespread implementation and before evaluating outcomes in larger comparative studies.

Strengths and limitations

This is the first synthesis exclusively comparing SST and NSST in minimally invasive anterior resections.

Strengths

• Adherence to PRISMA 2020 standards, prospective registration, and consistent effect direction across studies.

• The present synthesis used detailed, study-level extractions to calculate pooled effect estimates and to present both quantitative and narrative syntheses.

• For the primary outcome (leak) the pooled estimate was robust and heterogeneity was low.

Limitations

• Non-randomised evidence: All included studies are observational (prospective and retrospective cohorts, case–control/matched designs) with moderate to high risk of bias in several reports; residual confounding and selection bias are therefore likely. The small number of studies also limited subgroup and publication bias analyses.

• Heterogeneous definitions and reporting: Differences in the definition and timing of “anastomotic leak”, variable use of imaging and selective reporting complicate pooling.

• The manuscript’s central contribution is to highlight the shared principle—a single circular staple line without crossover—as a promising direction that warrants further technique-specific prospective study.

• Surgeon and centre effects: Technique, case selection (tumour height, neoadjuvant radiotherapy), use of diverting stomas and perioperative pathways (ERAS) differ between studies and may explain some observed effects—particularly for continuous outcomes such as length of stay.

• Sparse data for some outcomes: Important clinical endpoints such as long-term functional outcomes, quality of life and permanent stoma rates were rarely reported and could not be reliably synthesised.

Future research potential

• Well-designed confirmatory multicentre RCTs comparing SST versus standard non-SST techniques in anterior resection with standardised definitions (clinical and radiological leak), consistent perioperative protocols (ERAS), and pre-specified functional/oncological endpoints.

• Prospective registries with granular data on anastomotic technique, stapler size, reinforcement strategies, diverting stoma policy and imaging criteria to allow adjusted observational comparisons and instrumental variable analyses.

• Parallel mechanistic studies (e.g. microvascular perfusion mapping at the staple line, ex vivo modelling) to elucidate the biological basis for any observed differences.

Conclusion

Using the study-level data, our analysis shows that single-stapled anastomosis after minimally invasive anterior resection is associated with an observed lower pooled risk of anastomotic leak compared with non-single-stapled techniques. Secondary outcomes are heterogenous and less certain. The body of evidence is largely observational and variably at risk of bias. Definitive adoption of SST as a standard technique will require confirmatory randomised evidence; currently, on the basis of observational study analysis, SST should be adopted with caution in the interim. Meanwhile, SST may be considered a valid option in the armamentarium for colorectal surgeons where anatomically appropriate following the core principles of a good anastomosis.

Supplementary Information

Below is the link to the electronic supplementary material.

Authors contributions

Shubham Jain: Conception of the study design, data collection, analysis and validation, writing of the manuscript, revision and approval of the manuscript. Radi Abweni: Conception of the study design, data collection and analysis, revision and approval of the manuscript. Maria Iacob: Data collection and analysis, revision and approval of the manuscript. Roland Labinoti: Conception of the study design, data collection and analysis, revision and approval of the manuscript. Arshad Malik: Conception of the study design, data validation, revision and approval of the manuscript.

Funding

None.

Data availability

Data available on request. Extracted from the included studies and presented as a meta-analysis.

Declarations

Conflict of interest

The authors declare that they have no financial or other interest. All authors read and approved the final manuscript.

Ethical approval and consent to participate

As this is a study-level meta-analysis no ethics approval was needed from the local institution body and neither did we need any patient consent.

Consent for publication

All authors consented to the publication of this manuscript. No patient related data is included in this study as it is a study-level meta-analysis and hence no consent was needed.