Comparison of Modified Graham’s Patch Repair with and without Anchoring Suture in Patients with Duodenal Ulcer Perforation

By Muhammad Abrar¹, Muhammad Shahid Farooq², Usman Mustafa¹, Ahtesham Ahmed¹, Muhammad Waqar Inam³, Ahmed Raza¹

Affiliations

1. South Surgical Ward, Mayo Hospital, Lahore, Pakistan
2. Department of General Surgery, King Edward Medical University, Mayo Hospital, Lahore, Pakistan
3. East Surgical Ward, Mayo Hospital, Lahore, Pakistan

doi: 10.29271/jcpsppg.2025.01.127

ABSTRACT
Objective: To compare the outcomes of modified Graham's patch repair with and without anchoring suture in patients with duodenal ulcer perforation.
Study Design: A quasi-experimental study.
Place and Duration of the Study: Department of General Surgery, King Edward Medical University, Mayo Hospital, Lahore, Pakistan, from October 2024 to April 2025.
Methodology: A total of 100 patients with perforated duodenal ulcer were assigned to two groups: Group A, modified Graham's patch repair with anchoring suture, and Group B, only modified Graham's patch repair. Postoperative outcomes were assessed in terms of complications (surgical site infection [SSI], bile leakage, and sepsis) and duration of hospital stay. Chi-square and Mann-Whitney U tests were applied for comparing categorical and numerical variables, respectively. A p-value of ≤0.05 was considered statistically significant.
Results: Complication rates with and without anchoring sutures were: Bile leakage, 0% vs. 2%; SSI, 10% vs. 18%; and sepsis, 4% vs. 6% (p >0.05). Median (IQR) hospital stay with and without anchoring suture was 9 (3.25) days vs. 13 (2) days (p <0.001).
Conclusion: These findings suggest that incorporating an anchoring suture in combination with the modified Graham's patch may enhance recovery without significantly affecting complication rates compared to the modified Graham's patch alone.

Key Words: Anchoring suture, Duodenal ulcer perforation, Modified Graham's patch repair.

INTRODUCTION

Duodenal ulcer (DU) perforation leads to the leakage of gastric and duodenal contents into the peritoneal cavity, initially causing chemical peritonitis, which can progress to bacterial contamination if the leakage persists.¹ Globally, peptic ulcer (PU) affects four million people each year, with an incidence rate of 1.5% to 3%.² Although its incidence has declined in recent years, approximately 5% individuals with PU disease (PUD) experience perforation during their lifetime.³

The advancement of effective medical therapies has significantly reduced the need for surgical intervention in PUD.⁴ However, surgery remains the primary treatment for most perforated  ulcer  cases.

 While multiple techniques exist for perforation closure, there is an ongoing debate in the literature regarding the most effective surgical approach for managing these cases.⁵

Graham’s omentopexy, first described by Cellan-Jones in 1929 and later popularised by Graham in 1937, involves simple primary closure of perforation with interrupted sutures, followed by the placement of omentum over the repaired site, which is secured with additional sutures for reinforcement.⁶ However, complications such as reperforation, sepsis, and burst abdomen are common, leading to significant morbidity and mortality.⁷ Modified Graham’s repair follows the same principle but uses a pedicled omental flap instead of a free omental patch, ensuring better vascularity and stability. Some modifications also involve placing the omentum between two layers of sutures (sandwich technique) to reduce the risk of leakage and reperforation.⁸ Other alternative closure techniques include primary closure with interrupted sutures, primary closure reinforced with pedicle omentoplasty, omental plugging, and figure-of-eight closure.⁹ Anchoring suture technique is a more recent modification that secures the edges of perforation to underlying healthy duodenal tissue with continuous or interrupted sutures, anchoring the omental patch firmly without excessive tension. This technique minimises tissue ischaemia, reduces operative time, and has been associated with earlier resumption of oral feeding and shorter hospital stay. It was suggested that while modified Graham’s technique was widely used, it was associated with higher rates of infection and bile leakage. In contrast, the anchoring suture technique demonstrated improved outcomes, with a lower incidence of complications, reduced surgical time, earlier commencement of oral feeding, and shorter hospital stay.¹⁰

DU perforation is a common surgical emergency, and modified Graham’s patch repair remains the standard treatment. However, variations in technique, such as the addition of an anchoring suture, may influence patient outcomes. Existing evidence on this modification is scarce, and its potential benefits in reducing postoperative complications and hospital stay remain unclear.

This study aimed to determine the role of anchoring suture in improving surgical outcomes, ultimately guiding the adoption of a more effective technique in a local setting by comparing outcomes of modified Graham’s patch repair with and without anchoring suture in patients with DU perforation.

METHODOLOGY

This quasi-experimental study was conducted at the Department of General Surgery, King Edward Medical University, Mayo Hospital, Lahore, Pakistan, from October 2024 to April 2025. Ethical approval was obtained from the Institutional Review Board of the King Edward Medical University, Lahore, Pakistan (No: 231/RC/KEMU; dated: 25 March 2025). A sample size of 100 cases was estimated using an 80% power of the study, a 95% confidence interval, and an expected hospital stay of 13.44 ± 2.00 days in the Graham’s technique and 9.60 ± 2.3 days in the Graham’s technique with an anchoring suture.¹⁰

Eligible participants were adults between 20 and 70 years of age, classified as American Society of Anesthesiologists (ASA) I or II, and diagnosed with anterior wall DU perforation in the first or second part of duodenum, diagnosed clinically and radiologically (as evidenced by the presence of air under diaphragm). Patients with perforations due to trauma, neoplastic conditions, sealed perforations, multiple perforations, or those with a prior history of duodenal surgery were excluded.

After obtaining informed consent, patients were divided into two equal groups according to the assigned treatment. Group A underwent modified Graham’s patch repair with an anchoring suture, where an additional seromuscular stitch secured the omental patch to pyloric area or the first part of the duodenum. Group B underwent conventional modified Graham’s patch repair without an anchoring suture. All surgeries were performed under general anaesthesia by an experienced surgical team, ensuring uniformity in procedural techniques. Postoperatively, patients were monitored for surgical site infection (SSI), sepsis, and postoperative leakage for 15 days. SSI was defined as the presence of pus discharge at the wound site occurring between 5 and 15 days after surgery. Sepsis was defined as the presence of fever and leucocytosis with documented positive blood culture during postoperative period, between 3 and 15 days after surgery. Postoperative leakage was defined as leakage of gastric contents or pus from the wound site or drain occurring between 3 and 15 days after surgery. The duration of postoperative hospitalisation was also documented. Patients were discharged from the hospital when fully mobilised, passing flatus and stool, and fever-free for 48 hours.

Statistical analysis was conducted using the SPSS version 25. The Shapiro-Wilk test was applied to assess data normality. Continuous variables were expressed as median (IQR), and categorical variables were summarised as frequencies and percentages. The Mann-Whitney U test was used to compare operative time between the groups, while the Chi-square test was applied for postoperative complications, and a p-value of ≤0.05 was considered statistically significant. Data were stratified based on age, gender, body mass index (BMI), ulcer duration, perforation size, smoking history, diabetes, and hypertension. Post-stratification, hospital stay was analysed using the Mann-Whitney U test, and postoperative complications were compared within each stratum using the Chi-square test.

RESULTS

As shown in Table I, the median (IQR) of age calculated was 43 (22) years for Group A and 43.50 (17) years for Group B (p = 0.983). Males and females were 88% and 12% in Group A and 94% and 6% in Group B (p = 0.485). Median BMI was nearly identical in both groups, 23.25 (5) kg/m² vs. 23 (2d.8) kg/m² (p = 0.983), and the duration of ulcer was 8.50 (3) months and 8.50 (4) months, respectively (p = 0.870). Smoking was common in both groups, with 66% and 60%, respectively (p = 0.534). The presence of diabetes was similar between the groups (18% vs. 22%, p = 0.617), as was the hypertension, identical in both groups (28% each). Perforation size distribution was comparable between the groups, with 58% vs. 62% patients having a perforation size of 0.5-1.0 cm, 22% vs. 26% having a size of 1.0-1.5 cm, 6% vs. 4% having a size of 1.5-2.0 cm, and 14% vs. 8% having a size greater than 2.0 cm. There was no significant difference in perforation size distribution between the groups (p = 0.741). The operative time was slightly lower in Group A, 77.50 (30) minutes, compared to Group B, 85 (20) minutes; however, this difference was not statistically significant (p = 0.235). Median hospital stay was significantly shorter in the Group A, 9 (3.25) days, compared to Group B, 13 (2) days, and this difference was statistically significant (p <0.001).

As shown in Figure 1, the complication rate was lower in the Group A compared to the Group B, bile leakage occurring in 0% vs. 2% (p = 0.500), SSI in 10% vs. 18% (p = 0.249), and sepsis in 4% vs. 6% (p = 0.500).

Table I: Patients' characteristics in both study groups.

Variables	Modified Graham’s patch		p-values
	Group A	Group B
n	50	50
Age (years), [median (IQR)]	43 (22)	43.50 (17)	0.983*
Gender			0.485**
Male	44 (88%)	47 (94%)
Female	6 (12%)	3 (6%)
BMI (kg/m2), [median (IQR)]	23.25 (5)	23 (2.8)	0.983*
Duration of ulcer (months), [median (IQR)]	8.50 (3)	8.50 (4)	0.870*
Smoking	33 (66%)	30 (60%)	0.534***
Diabetes	9 (18%)	11 (22%)	0.617***
Hypertension	14 (28%)	14 (28%)	-
Size of perforation
0.5-1.0 cm	29 (58%)	31 (62%)	0.74***
1.0-1.5 cm	11 (22%)	13 (26%)
1.5-2.0 cm	3 (6%)	2 (4%)
>2.0 cm	7 (14%)	4 (8%)
Operative time (minutes), [median (IQR)]	77.50 (30)	85 (20)	0.235*
Hospital stay (days), [median (IQR)]	9 (3.25)	13 (2)	<0.00*
*Mann-Whitney U test, **Fisher's exact test, ***Chi-square test.

Table II: Comparison of postoperative complications stratified in relation to various patient factors.

Variables	Categories	Bile leakagea		Surgical site infectionb		Sepsisc		p-values
		A	B	A	B	A	B	a	b	c
Age (years)	25-40	0%	0%	16.70%	21.10%	8.30%	15.80%	-	-	0.640*
	41-55	0%	0%	8.30%	21.10%	0%	0%	-	0.624*	-
	55-70	0%	8.30%	0%	8.30%	0%	0%	0.462*	0.462*	-
Gender	Female	0%	0%	0%	66.70%	0%	0%	-	0.083*	-
	Male	0%	2.10%	11.40%	14.90%	4.50%	6.40%	-	0.619*	-
BMI	Normal	0.00%	4.30%	16.70%	13.00%	4.20%	4.30%	0.489*	-	-
	Overweight	0%	0%	3.80%	22.20%	3.80%	7.40%	-	0.100**	-
Ulcer size	0.5-1.0	0%	0%	10.30%	12.90%	3.40%	3.20%	-	0.538*	0.737*
	1.0-1.5	0%	0%	9.10%	30.80%	0%	15.40%	-	0.327*	0.428*
	1.5-2.0	0%	0%	0%	0%	0%	0%	-	-	-
	>2	0%	25%	14.30%	25%	14.30%	0%	0.364f	0.618*	0.636*
Ulcer duration	5-9 months	0%	3.00%	12.10%	21.20%	3.00%	9.10%	0.500f	0.322*	0.613*
	10-13 months	0%	0%	5.90%	11.80%	5.90%	0.00%	-	0.500*	0.500*
Diabetes	Yes	0%	0%	0%	18.20%	0%	0%	-	0.479*	-
	No	0%	2.60%	12.20%	17.90%	4.90%	7.70%	0.487f	0.417**	0.671*
Hypertension	Yes	0%	0%	7.10%	14.30%	7.10%	14.30%	-	0.500*	0.500*
	No	0%	2.80%	11.10%	19.40%	2.80%	2.80%	0.500f	0.362**	0.754*
Smoking	Yes	0%	0%	12.10%	23.30%	6.10%	6.70%	-	0.242**	0.657*
	No	0%	5%	5.90%	10%	0%	5%	0.541*	0.562*	0.541*
*Fisher’s exact test, **Chi-square test.

Table III: Hospital stay of patients in relation to various patient factors.

Variables		Group A	Group B	p-values*
		Median (IQR)	Median (IQR)
Age (years)	25-40	9 (2.75)	13 (2)	<0.001**
	41-55	8 (3.75)	13 (3)	<0.001**
	55-70	10 (2.25)	13 (2.75)	<0.001**
Gender	Male	9 (4)	13 (2)	<0.001**
	Female	9.50 (1.50)	13 (-)	0.026**
BMI	Normal	9 (2.75)	14 (3)	<0.001**
	Overweight	9 (3.25)	13 (2)	<0.001**
Ulcer duration	5-9 months	9 (3.50)	13 (2)	<0.001**
	10-13 months	10 (3)	13 (2.5)	<0.001**
Perforation size (cm)	0.5-1.0	10 (2.50)	13 (3)	<0.001**
	1.0-1.5	8 (3)	13 (2.50)	<0.001**
	1.5-2.0	8 (-)	12 (0)	0.200
	>2.0	7 (0)	13.50 (1.75)	0.006**
Smoking	Yes	9 (4)	13 (3)	<0.001**
	No	9 (2)	13 (2)	<0.001**
Diabetic	Yes	10 (2)	14 (2)	<0.001**
	No	9 (3)	13 (2)	<0.001**
Hypertension	Yes	9 (3.25)	13 (2.50)	<0.001**
	No	9.50 (3.50)	13 (2)	<0.001*
*Mann-Whitney U test, **Statistically significant.

Figure 1: Postoperative complications in both study groups.

As shown in Table II, no stratified groups had bile leakage in Group A. However, in Group B, bile leakage was observed more in patients aged >55 years (8.3%), those with ulcer sizes >2 cm (25%), individuals with normal BMI (4.3%), males (2.6%), patients with ulcer duration of 5–9 months (3.0%), smokers (0%), hypertensive patients (0%), and diabetic patients (0%). SSI was more common in Group A among patients with normal BMI (16.7%) and those with ulcer sizes >2 cm (14.3%), whereas in Group B, it was higher in females (66.7%), overweight individuals (22.2%), and those with ulcer sizes 1.0–1.5 cm (30.8%) and >2 cm (25%). It was also more frequent in patients with longer ulcer durations of 5–9 months (21.2%) and in diabetics (18.2%). Sepsis in Group A was noted more in normal weight individuals (4.2%) and those with ulcer sizes >2 cm (14.3%), while in Group B, it was observed in overweight individuals (7.4%) and those with ulcers lasting 5–9 months (9.1%). No statistically significant association was found between postoperative complications and patient-related factors (p >0.05).

As shown in Table III, data stratification with respect to age, gender, BMI, duration of the ulcer, perforation size, diabetes, hypertension, and smoking showed a statistically significant difference in terms of prolonged hospital stay observed in Group B as compared to Group A (p <0.001). However, for the 1.5-2.0 cm group perforation size, the difference was not statistically significant (p = 0.200).

DISCUSSION

The demographic details of the patient population in the present study are similar to the trial by Kumar et al., where the mean age was reported as 46.80 ± 13.9 years and 48.60 ± 14.04 years, with a male predominance of 83.33% and 80%.¹¹ Additionally, Khan et al. also found a male-to-female ratio of 2.1:1 in their study.¹² In the current study, the most common comorbid condition found was smoking in both groups (66% and 60%), followed by hypertension (28% in each group) and diabetes (18% and 22%). Similarly, previous research has reported smoking in 65% of patients with PUs.¹³ Although the direct association of hypertension and diabetes with the incidence of DUs remains uncertain in existing research. However, it is acknowledged that these comorbids may add to the morbidity and mortality of patients with ulcers.¹⁴

In the current study, the complication rate was lower in the group with anchoring sutures compared to the group without anchoring sutures. Mean hospital stay was also significantly shorter in the anchoring suture group compared to without an anchoring suture (p <0.001). Similarly, Rudraiah and Kalke found that patients who underwent suture reinforcement in addition to the modified Graham's repair had superior outcomes compared to the modified Graham's repair alone, with bile leakage reported 0% vs. 8%, septicaemia 12% vs. 24%, and wound infection 32% vs. 48%. Furthermore, patients in the suture group had earlier oral intake, shorter operative time, and reduced hospital stay (9.60 ± 2.31 days vs. 13.44 ± 2.00 days).¹⁰ Mahmoud et al. documented that among patients with perforation repair, factors significantly associated with leakage included a lack of suture closure (OR 7.0). Further supporting the present study’s results, these findings suggest that sutured closure reduces the risk of leakage compared to patch-only repair.¹⁵ In contrast, Al-Asadi et al. observed that adding anchoring sutures to the modified Graham’s patch repair led to higher rates of postoperative complications, including superficial SSI (21.9%), postoperative ileus (12%), and pneumonia (8%).¹⁶ The use of anchoring sutures in laparoscopic surgery demonstrated more favourable outcomes, including a shorter hospital stay of 9 (3.25) days compared to 13 (2) days without anchoring sutures (p = 0.001), and a faster return to normal activities (p <0.001). However, postoperative complications showed no significant difference between the techniques (p >0.05).¹⁷ While laparo-scopic repair has shown favourable outcomes as reported, the feasibility and technical application of anchoring sutures in a minimally invasive setting may differ from open surgery due to restricted working space and visualisation, and this warrants further evaluation in future studies.

In the current study, bile leakage was observed more frequently in patients aged from 55 to 70 years, those with ulcer sizes >2 cm, individuals with normal BMI, males, and patients with ulcer duration of 5–9 months. Leakage following omentopexy was significantly linked to larger perforation size, delayed presentation, sepsis, immunosuppression, and perioperative shock. Delayed presentation of more than 48 hours increased the risk by 2.5 times, perforation diameter 2.1–3.0 cm eightfold, and the diameter exceeding 3.0 cm heightened the risk 33-fold.¹⁸ Additionally, an increase in the ulcer size by 10 mm has been shown to increase the leak rate by 3.3 times.¹⁹ Dogra et al. reported that low BMI was significantly associated with leak post-omentopexy, whereas in the present study, bile leakage was found to be associated with normal BMI.²⁰

This study has several limitations. The study has not assessed long-term complications or recurrence rates. Additionally, variations in individual surgeon techniques, despite the efforts to standardise procedures, could have influenced outcomes. The study also relied on clinical and radiological diagnosis without intraoperative histopathological confirmation of ulcer aetiology. Lastly, factors such as nutritional status and preoperative resuscitation, which may impact recovery, were not extensively analysed.

CONCLUSION

These findings suggest that incorporating an anchoring suture in combination with the modified Graham's patch may enhance recovery without significantly impacting complication rates compared to the modified Graham's patch alone.

ETHICAL APPROVAL:
Ethical approval was obtained from the Institutional Review Board of King Edward Medical University, Lahore, Pakistan (No: 231/RC/KEMU; dated: 25 March 2025).

PATIENTS’ CONSENT:
Written informed consent was taken from all patients included in the study.

COMPETING INTEREST:
The authors declared no conflict of interest.

AUTHORS’ CONTRIBUTION:
MA: Identify study gap, writing synopsis, data collection, and surgical intervention.
MSF, MWI: Data analysis and proofreading.
UM, AA, AR: Data collection, design of work, and writing of the article.
All authors approved the final version of the manuscript to be published.

REFERENCES

1.  Malik M, Magsi AM, Parveen S, Khan MI, Iqbal M. Duodenal ulcer perforation and its consequences. J Pak Med Assoc 2023; 73(7):1506-10. doi: 10.47391/JPMA.4768.
2. Salari N, Darvishi N, Shohaimi S, Bartina Y, Ahmadipanah M, Salari HR, et al. The global prevalence of peptic ulcer in the world: A systematic review and meta-analysis. Ind J Surg 2022; 2021(9337):1-9. doi: 10.1007/s12262-021-03189-z.
3. Maksymchuk D. Duodenal ulcer and its complications: Bleeding, perforation, penetration. Int J Health Sci 2021; 5(3). doi: 10.53730/ijhs.v5n3.2014.
4. Kamada T, Satoh K, Itoh T, Ito M, Iwamoto J, Okimoto T, et al. Evidence-based clinical practice guidelines for peptic ulcer disease 2020. J Gastroenterol 2021; 56(4):303-22. doi: 10.1007/s00535-021-01769-0.
5. Salman MA, Issa M, Salman A, Tourky M, Elewa A, Alrahawy M, et al. Surgical management of perforated peptic ulcer: A comparative meta-analysis of laparoscopic versus open surgery. Surg Laparosc Endosc Percutan Tech 2022; 32(5): 586-94. doi: 10.1097/SLE.0000000000001086.
6. Egeli T, Cavdaroglu O, Agalar C, Derici S, Aksoy S, Yilmaz I, et al. How to manage difficult duodenal defects? Single center experience. Turk J Surg 2024; 40(2):161-7. doi: 10.47717 turkjsurg.2024.6476.
7. Zhu C, Badach J, Lin A, Mathur N, McHugh S, Saracco B, et al. Omental patch versus gastric resection for perforated gastric ulcer: Systematic review and meta-analysis for an unresolved debate. Am J Surg 2021; 221(5):935-41. doi: 10.1016/j.amj surg.2020.07.039.
8. Salah El-Din AM, Batikhe MY, Ali MM, Redwan AA. Laparoscopic repair of perforated peptic ulcer: Simple closure versus omentopexy. Egypt J Hosp Med 2021; 85(1):3080-6. doi: 10. 21608/ejhm.2021.194060.
9. Shen Q, Liu T, Wang S, Wang L, Wang D. Experience in diagnosis and treatment of duodenal ulcer perforation in children. BMC Pediatr 2023; 23(1):144. doi: 10.1186/s12887- 023-03957-8.
10. Rudraiah H, Kalke SV. A stitch in time saves nine-a novel treatment of duodenal perforation closure. Int J Contemp Med Res 2019; 6(3):C9-13. doi: 10.21276/ijcmr.2019.6.3.33.
11. Kumar R, Walia JS, Attri A, Singh S. Closure of peptic ulcer perforation by Grahm’s patch with omentopexy versus plugging with omentopexy. Int J Surg Med 2020; 6(6):1. doi: 10.5455/ijsm.2020-02-07.
12. Khan SR, Shahinda MM, Khan L, Rahman NU, Bibi R. Frequency of wound infection after duodenal ulcer perforation repair. Pak J Med Health Sci 2023; 17(02):25. doi: 10.53350/pjmhs 202317225.
13. Bushra T, Mondal S, Nazmin F, Arnab KH, Tisa AH, Roy A, et al. Burden of peptic ulcer disease among smoking and non-smoking healthcare providers. A comparative cross-sectional study in Gazipur, Dhaka. Asia Pac J Surg Adv 2024; 1(2):44-50. doi: 10.70818/apjsa.2024.v01i02.07.
14. Shreya A, Sahla S, Gurushankari B, Shivakumar M, Rifai, Kate V, et al. Spectrum of perforated peptic ulcer disease in a tertiary care hospital in South India: Predictors of morbidity and mortality. ANZ J Surg 2024; 94(3):366-70. doi: 10.1111/ ans.18831.
15. Mahmoud K, Abou Chaar MK, Stephens D, Zietlow JM, Heller SF, Turay D, et al. Perforated peptic ulcer: To close or patch - a century-old controversy. J Gastrointest Surg 2025; 29(4): 101993. doi: 10.1016/j.gassur.2025.101993.
16. Al-Asadi OG, Baqir QK, Alnajjar IM. Novel technique for the placement of omental patch during repair of perforated duodenal ulcer. A retrospective study of prospectively collected data. Surg Gastroenterol Oncol 2024; 29(2). doi: 10.21614/sgo-663.
17. Celik Y, Erbil O. Laparoscopic repair versus open repair for perforated peptic ulcers is figure of eight suture safe? Int J Med Sci Dent Rep 2020; 3(2):1-6.
18. Ogbuanya AU, Eni UE, Umezurike DA, Obasi AA, Ikpeze S. Associated factors of leaked repair following omentopexy for perforated peptic ulcer disease; a cross-sectional study. Arch Acad Emerg Med 2024; 12(1):e18. doi: 10.22037/aaem. v12i1.2169.
19. Wang YL, Chan XW, Chan KS, Shelat VG. Omental patch repair of large perforated peptic ulcers ≥25 mm is associated with higher leak rate. J Clin Transl Res 2021; 7(6):759-66. doi: 10.18053/jctres.07.202106.014.
20. Dogra P, Kaushik R, Singh S, Bhardwaj S. Risk factors for leak after omentopexy for duodenal ulcer perforations. Eur J Trauma Emerg Surg 2023; 49(2):1163-7. doi: 10.1007/s000 68-022-02058-y.