Perforated peptic ulcer

A 58-year-old man on ibuprofen for his knees arrives at 03:00 with six hours of the worst epigastric pain of his life and a rigid, silent abdomen. His erect chest film shows a crescent of air under the right hemidiaphragm. Every decision from here — how fast to theatre, laparoscopic or open, suture or patch, and what must happen before discharge — is mapped by guideline and cohort data, and every number on this page is quoted from the source.^[1]

Definition and the distinctions that matter

A perforated peptic ulcer is a full-thickness defect of a gastric or duodenal ulcer that allows luminal contents into the peritoneal cavity. Peptic ulcer disease itself is common — lifetime prevalence 5–10%, incidence 0.1–0.3% per year — and despite a sharp reduction in incidence, admissions and mortality over 30 years, complications still occur in 10–20% of patients.^[1]

Three distinctions drive management. First, free versus sealed perforation: a small perforation can seal by omental adhesions, which is the entire rationale for non-operative management — and why WSES permits NOM only when a water-soluble contrast study confirms the perforation has sealed.^[1] Second, duodenal versus gastric: in surgical series the perforation is duodenal, juxta-pyloric or prepyloric in the large majority (prepyloric region 74% in Lohsiriwat's series; duodenal 38.3%, juxta-pyloric 35.6%, gastric 19.1% and pyloric 6.8% in Grassi's 166-patient radiological series), and a perforated gastric ulcer carries the malignancy question that dominates its own section below.^[10][8] Third, size: WSES splits procedures at 2 cm — primary repair below, a tailored approach above — and leak rates up to 12% have been reported when large ulcers are closed with an omental patch.^[1]

Epidemiology — an older, sicker disease

^{[2] [5] [30]}

The incidence has not fallen the way uncomplicated ulcer disease has: Lau's systematic review of complicated peptic ulcer found annual incidence estimates of 19.4–57.0 per 100,000 for haemorrhage and 3.8–14 per 100,000 for perforation, with an average 30-day mortality after perforation of 23.5% (95% CI 15.5–31.0) — three times the mortality of bleeding.^[2] In a defined Norwegian population over ten years, the adjusted incidence was 6.5 per 100,000 per year with a total 30-day mortality of 16.3% and a standardised mortality ratio of 5.7; for patients of 60 and over, incidence rose more than ten-fold and mortality more than fifty-fold compared with younger patients.^[3]

The patient has changed even when the operation has not. Over four decades at one Norwegian hospital, the median age at perforation rose from 63 to 72 years, aspirin use among perforated patients rose from 5% to 18%, and the proportion with ASA 4–5 rose from 5% to 22% — declining incidence, but older patients with more comorbidity.^[4] Møller and colleagues made the same point from Denmark: despite H2-blockers, proton pump inhibitors and the discovery of H. pylori, both the incidence of emergency surgery for PPU and the mortality of those operated on increased, because the population became older and sicker.^[31]

Risk factors — the attributable fractions worth quoting

About one in four ulcer perforations can be attributed to non-steroidal anti-inflammatory drugs, a risk factor of particular importance in the elderly; most perforations in patients under 75 can be attributed to smoking.^[5] NSAIDs cause foregut symptoms, peptic ulceration and small-bowel enteropathy, and that iatrogenic injury is complicated by bleeding and perforation; limiting NSAID use or co-prescribing a PPI reduces ulcer disease and its complications.^[6] The gastroprotectant meta-analysis of 849 randomised trials (142,485 participants) gives the flip side: gastroprotectants — PPIs above all — reduced endoscopic ulcers (OR 0.27), symptomatic ulcers (OR 0.25) and upper GI bleeding (OR 0.40), though not mortality.^[7] Lau's review completes the triad: risk factors for ulcer complications and their recurrence are NSAID and/or aspirin use, H. pylori infection and an ulcer 1 cm or larger.^[2]

Pathophysiology — chemical burn to polymicrobial sepsis

An anterior duodenal or prepyloric ulcer erodes through the wall and vents acid, bile, food and bacteria straight into the peritoneal cavity; posterior lesions follow a different fate — most duodenal ulcers that require surgery for persistent bleeding are large posterior lesions bleeding from the gastroduodenal artery — which is why free perforation is overwhelmingly an anterior, duodenal and juxta-pyloric event in the radiological site series.^[1][8] The early picture is chemical peritonitis; the late picture is infective. WSES is explicit that perforated peptic ulcer peritonitis is by definition polymicrobial — gram-negative and gram-positive bacteria, anaerobes and yeasts are all isolated from peritoneal fluid — which is why the empiric regimen below covers that spectrum and why peritoneal fluid is cultured at operation.^[1]

What converts contamination into death is time and host reserve. The WSES-cited systematic review of 50 studies and 29,782 patients found strong evidence associating older age, comorbidity, and NSAID or steroid use with mortality, with shock on admission, preoperative metabolic acidosis, tachycardia, acute renal failure, low albumin, high ASA score and preoperative delay beyond 24 hours also predicting poor outcome.^[1] Much of the risk profile is simply the septic state of the patient on arrival.^[31]

Clinical presentation

The clinical presentation of gastroduodenal perforation is usually sudden-onset abdominal pain; localised or generalised peritonitis is typical — but WSES's caveat is the one that matters at 03:00: peritonitis may be present in only two-thirds of patients, so equivocal signs do not exclude the diagnosis, and imaging is not optional.^[1]

At the other extreme is the patient who arrives in septic shock. In the GRACE global snapshot — 1,874 surgical patients across 159 centres in 52 countries — age over 50, female sex, shock on admission and acute kidney injury were each independently associated with both 30-day morbidity and mortality; delayed presentation beyond 24 hours was associated with morbidity but not mortality.^[30] Shock on admission is also a Boey factor and a PULP variable, so it earns its place in three separate risk instruments on this page.^[9][11]

Scoring systems — Boey, PULP and the albumin caveat

Boey (1987) — say the three factors, then the four numbers

The Boey score counts three risk factors: major medical illness, preoperative shock, and longstanding perforation (more than 24 hours). In Boey's prospective validation of 259 consecutive patients undergoing simple closure or definitive surgery for perforated duodenal ulcer, the three factors correctly predicted outcome in 93.8% of patients, and all 16 deaths (6.2%) were identified with no false-negative error. The mortality bands are the quote: 0%, 10%, 45.5% and 100% for zero, one, two and three risk factors.^[9] The management corollary came from the same paper: simple closure is preferable in patients with uncomplicated perforations when any risk factor is present, while definitive surgery can be done safely in good-risk patients — and non-operative treatment "deserves re-evaluation" in the three-factor patient because operative outcome is uniformly dismal.^[9]

The score survives external validation. In Lohsiriwat's 152-patient series, mortality climbed 1%, 8%, 33% and 38% across Boey 0–3, with morbidity rising 11%, 47%, 75% and 77%; Boey and ASA outperformed the Mannheim Peritonitis Index.^[10] WSES notes the score's accuracy varies across studies but confirms it remains the most used, followed by ASA and PULP.^[1][12]

PULP (2012) — the Danish successor

The Peptic Ulcer Perforation score was derived from a Danish nationwide cohort of 2,668 surgically treated patients, of whom 708 (27%) died within 30 days. Eight variables: age over 65; active malignant disease or AIDS; liver cirrhosis; steroid use; time from perforation to admission over 24 hours; preoperative shock; serum creatinine over 130 µmol/L; and ASA class 2–5. It predicted mortality with an AUC of 0.83 — considerably better than Boey (0.70) and better than ASA alone (0.78).^[11]

Head-to-head data since then are kinder to PULP than to Boey but never decisive. In Thorsen's 172-patient comparison, PULP carried an odds ratio of 18.6 and ASA 11.6 (both AUC 0.79) against Boey's OR 5.0 and AUC 0.75 — and hypoalbuminaemia alone (37 g/L or below) achieved an OR of 8.7 and AUC 0.78, the strongest single predictor of mortality, a finding WSES repeats.^[13][1] In Saafan's morbidity analysis, PULP had the largest AUC (72%) and was the only score significantly predicting 30-day morbidity.^[14] The 2026 systematic review and meta-analysis (36 studies) found PULP numerically ahead on sensitivity (80% vs 78%), specificity (85% vs 78%) and diagnostic odds ratio (20.67 vs 12.69) for mortality, but concluded the two scores show comparable accuracy and either can be used.^[15] Thorsen's review of ten scoring systems is the honest summary: mortality in this disease ranges 3–40%, reported AUCs swing from 0.63 to 0.98, and no score has displaced bedside judgement.^[12] WSES suggests adopting scoring systems — Boey, PULP and ASA — for risk stratification and outcome prediction (2C).^[1]

Investigations and imaging — CT first, and what "normal" means

WSES recommends routine laboratory studies and arterial blood gas analysis in suspected gastroduodenal perforation (1D) — the creatinine feeds PULP, the albumin is the strongest single mortality predictor, and the blood gas quantifies the metabolic acidosis of sepsis.^[1][13]

For imaging, the guideline order is clear: CT is the recommended examination in suspected perforated peptic ulcer (1C); chest/abdominal X-ray is the initial routine assessment only when CT is not promptly available (1C); and when free air is absent but suspicion persists, add water-soluble contrast orally or via nasogastric tube (2D).^[1] The plain-film trap: reported rates of visible free air vary between 30% and 85% of perforations across studies, and a negative X-ray does not rule out perforation — though free air on film with a clear history and peritonitis is sufficient to justify surgical exploration.^[1] In Grassi's 166 consecutive gastroduodenal perforations, plain film made the diagnosis in 85.5% of those with direct findings, and 12 patients had no free air on any modality.^[8]

CT's own ceiling must be quoted with its strengths. Suspicious findings are unexplained intraperitoneal fluid, pneumoperitoneum, bowel-wall thickening, mesenteric fat streaking and extraluminal water-soluble contrast; CT characterises the site and size of perforation and excludes alternative causes. But up to 12% of patients with perforations have a normal CT — in that scenario, water-soluble contrast (oral or nasogastric) with a triple-contrast CT improves sensitivity and specificity.^[1]

Differential diagnosis

The differential of the acute rigid abdomen with free gas is short: another perforated hollow viscus (colonic diverticular or malignant perforation, small-bowel perforation) and the mimics that produce pain and rigidity without free gas (acute pancreatitis, mesenteric ischaemia, ruptured aortic aneurysm). CT earns its 1C recommendation precisely here: beyond detecting free air, it characterises the site and size of perforation and excludes other causes.^[1] The differential that changes an operation is perforated gastric cancer — four of 44 perforated gastric ulcers (8.8%) in the Edinburgh series were tumours, one of them declared only at follow-up endoscopy — which is why the gastric section below is compulsory reading.^[24]

Management

Resuscitation and the tyranny of the clock

WSES recommends prompt evaluation and early recognition of the patient with PPU-associated sepsis to prevent organ failure and reduce mortality (1B), with scoring systems such as SOFA and qSOFA to grade severity (2C) and haemodynamic monitoring to individualise fluids and vasopressors (1C).^[1] Empiric broad-spectrum antibiotics against gram-negative, gram-positive and anaerobic bacteria start as soon as possible, ideally after peritoneal fluid is collected (1C).^[1]

Then move. The Danish nationwide cohort (2,668 patients; 26.5% 30-day mortality) found every hour of delay from admission to surgery associated with an adjusted 2.4% decrease in survival probability versus the previous hour (adjusted RR 1.024, 95% CI 1.011–1.037).^[16] England's National Emergency Laparotomy Audit analysis (3,809 patients operated within 24 hours; 90-day mortality 10.6%) found adjusted odds of death rising 4% per hour overall and — in the 334 physiologically shocked patients — 6% per hour of delay (OR 1.06, 95% CI 1.01–1.11), concluding source control should occur as soon as possible regardless of time of day.^[17] WSES accordingly recommends surgery as soon as possible, especially in delayed presentation and in patients over 70 (1B).^[1]

Non-operative management — one door, clearly marked

WSES suggests against routine non-operative management; NOM may be considered only in extremely selected cases where the perforation has sealed, confirmed on a water-soluble contrast study (2C).^[1] The evidence behind the caution is the Crofts 1989 randomised trial of 83 patients: 28% of the NOM arm showed no clinical improvement at 12 hours and required operation, mortality was identical (two deaths, 5%, in each arm), and hospital stay was 35% longer in the conservatively treated group.^[1] A later best-evidence review found both strategies can give similar outcomes — but only with correct patient selection, which is precisely the guideline's point.^[18]

If NOM is chosen, its components are defined: nil by mouth, intravenous hydration, nasogastric decompression, antisecretory PPI therapy, intravenous antibiotics and follow-up endoscopy at 4–6 weeks — with mortality rising with every hour of delay to surgery, so selection must be careful.^[1] Two groups are explicitly dangerous: the elderly, who may experience paradoxically higher mortality if NOM fails (caution advised over 70), and any patient with peritonitis or sepsis, for whom free contrast leak or systemic signs mandate operation.^[1] WSES also suggests avoiding endoscopic treatment — clipping, fibrin glue, stenting — of perforated peptic ulcer (2C): fibrotic ulcer tissue loses compliance and clips may not hold.^[1]

The operation — laparoscopic or open

WSES: in the stable patient, a laparoscopic approach is suggested, with open repair recommended when laparoscopic skills and equipment are absent (2B); in the unstable patient, open surgery is recommended (1D).^[1]

The trial evidence is small, old and contested — say so before quoting it. Lau's 1996 Annals of Surgery RCT randomised 103 patients to laparoscopic or open repair, sutured or sutureless: laparoscopy took significantly longer (94.3 versus 53.7 minutes) but needed significantly less postoperative analgesia (median 1 versus 3 doses), with no difference in morbidity, reoperation or mortality; sutureless repair proved as safe as sutured repair and faster.^[19] Tan's 2016 meta-analysis of five RCTs (549 patients) found no significant difference in overall complications, mortality or reoperation, but a lower surgical-site infection rate, shorter nasogastric tube duration and less pain with laparoscopy.^[20] The 2025 meta-analysis with trial sequential analysis (nine RCTs, 670 patients) went further — laparoscopic repair significantly reduced mortality (RR 0.37), total complications (RR 0.57), ileus (RR 0.43), wound complications (RR 0.36) and stay (−2.37 days), with no significant difference in leak (RR 2.00, CI 0.74–5.41), and concluded laparoscopy should be the management of choice where expertise exists.^[21]

The counterweight is the BJS Open scoping review (2025): nine trials, 880 patients, generally delivered by high-volume laparoscopic surgeons, generally at high risk of bias, with the intervention variably defined and co-interventions such as nutrition and H. pylori eradication unreported — the authors call for a well-designed pragmatic trial rather than closure of the question.^[22] And the real world, per GRACE, still opens 80% of these abdomens.^[30] The synthesis an examiner wants: laparoscopy in the stable patient with expertise, no hesitation about opening the unstable one, and honesty about evidence quality either way.^[1][21][22]

The repair — suture, Graham patch, and the 2 cm line

For perforations smaller than 2 cm, WSES suggests primary repair, and — the update that surprises candidates trained on the patch — no recommendation can be made that an omental patch adds protection (2C). The historical "standard" of routine patch is now debated: multiple series show the patch adds operating time without improving leak rates over simple suture, so WSES cannot suggest its routine application, while accepting it as a viable option in selected cases — many authors still advise it for large ulcers with friable edges where sutures might cut through.^[1] The eponym remains examinable regardless: the Graham patch, three interrupted sutures holding an omental graft over the defect, demonstrated by EAST in both open and laparoscopic forms as a core technique of complicated ulcer surgery.^[23][32]

For perforations 2 cm or larger, WSES suggests a tailored approach by location: suspicious large gastric ulcers are resected with intra-operative frozen section whenever possible; large duodenal ulcers may need resection or repair with pyloric exclusion and external bile drainage; duodenostomy is reserved for extreme circumstances (2D).^[1] The bailout catalogue for the large duodenal defect — jejunal serosal patch, Roux-en-Y duodenojejunostomy, pyloric exclusion, omental plug variants — exists because attempted patch closure of large ulcers has reported leak rates up to 12%, and these patients often arrive in septic shock, which should itself steer the surgeon away from heroic resection.^[1]

The perforated gastric ulcer — a different operation

Gastric perforation is more lethal than duodenal (Svanes' trend data) and carries the cancer question.^[5] Leeman's Edinburgh series (44 laparotomies for perforated gastric ulcer): 91% had omental patch repair, only two needed distal gastrectomy, and four (8.8%) were perforated gastric tumours — two suspected intra-operatively, one found on unexpected histology, and one missed even on biopsy and declared at follow-up endoscopy. All four were managed without initial resection.^[24] The biopsy debate is live: Koca's 135-patient series found excisional biopsy of the perforation edge significantly increased postoperative morbidity and Clavien-Dindo III+ complications without changing mortality, concluding that closure followed by early endoscopic biopsy may be the superior way to exclude malignancy.^[25] The examinable synthesis: patch or close the gastric perforation, biopsy if it can be done without enlarging the defect, resect with frozen section when malignancy is overt, and scope every patient afterwards — the cancer can hide behind benign histology.^[24][25][1]

Antibiotics, cultures and the antifungal trap

Start empiric broad-spectrum antibiotics covering gram-negatives, gram-positives and anaerobes as soon as possible, ideally after peritoneal fluid sampling (1C); culture fluid for bacteria and fungi in every surgical patient and adjust therapy to results (1C); and keep the course short — 3–5 days or until inflammatory markers normalise (2C), consistent with the source-control-era evidence that fixed short courses are enough once the defect is closed.^[1] Do not give empiric antifungals as standard: WSES suggests against routine antifungal therapy in PPU, reserving it for high-risk patients (immunocompromised, advanced age, prolonged ICU stay, unresolved abdominal infection), because positive fungal cultures, while prognostically bad, did not translate into a survival benefit from empiric treatment in the cited 133-patient analysis.^[1]

H. pylori — the step that prevents the second perforation

The organism is common in this population: 84.8% of El-Nakeeb's perforated duodenal ulcer patients tested positive (biopsies taken through the perforation at laparotomy), and Wong's meta-analysis confirmed a high prevalence across the randomised trials.^[28][27] Eradication after simple closure is one of the best-supported interventions on this page. Tomtitchong's meta-analysis: one-year ulcer recurrence 5.2% with eradication versus 35.2% with antisecretory therapy alone — pooled relative risk 0.15 (95% CI 0.06–0.37) — with the conclusion that all duodenal perforation patients should be tested and all infected patients eradicated.^[26] Wong's meta-analysis of five RCTs (401 patients) confirmed significantly reduced recurrence at 8 weeks and 1 year.^[27] The single trials are as striking: El-Nakeeb, 6.1% versus 29.6% recurrence at one year; Bose, 18.6% recurrence in the eradicated versus 70% when eradication failed.^[28][29]

Special populations

The elderly. This is now the typical patient: median age at perforation has risen into the seventies, incidence rises ten-fold and mortality more than fifty-fold over 60, and aspirin exposure climbs with each decade.^[4][3] WSES singles out the over-70s twice: surgery should be as soon as possible precisely because they tolerate delay worst, and NOM is hazardous because failure carries paradoxically higher mortality.^[1]

The shocked, comorbid, three-Boey-factor patient. Boey's own conclusion still stands: with all three risk factors, operative mortality was 100% in the 1987 validation, and non-operative treatment "deserves re-evaluation" in exactly these patients — while GRACE confirms shock and acute kidney injury as the modern markers of the same physiology.^[9][30] These patients need the fastest possible source control, the most senior hands, and honest prognostication with the family.^[16][17]

The NSAID- and steroid-dependent patient. Both drug classes appear in the mortality risk factor lists, steroids are a PULP variable, and NSAIDs account for about a quarter of perforations — the perioperative plan must include what happens to the drug afterwards: stop it, substitute it, or co-prescribe protection, because PPIs prevent ulcer complications in chronic users.^{[1][11][5][7]}

Complications and prognosis

The early complications are leak at the repair site, intra-abdominal abscess, ileus, wound complications and persistent sepsis. The comparative numbers come from the 2025 meta-analysis: laparoscopy beat open on total complications, ileus, wound complications and stay, with no significant difference in postoperative leak, abdominal collection, sepsis or reoperation — a leak after laparoscopic repair is not an argument for having opened, and vice versa.^[21] Large-ulcer closures carry the highest leak risk (up to 12% with patch closure of large defects), and any postoperative deterioration mandates cross-sectional imaging for a drainable collection.^[1]

Prognosis in one paragraph: mortality ranges 3–40% across series because case-mix ranges; the pooled average 30-day mortality after perforation is 23.5%; the modern global surgical snapshot reports 9.3% 30-day mortality with 48.5% morbidity; and the Boey bands (0/10/45.5/100%) remain the fastest bedside statement of risk.^{[12][2][30][9]} The modifiable prognostic factors are delay, delay and delay — the only variables on the mortality list the admitting team controls.^[16][17] Long term, the uneradicated, unstopped-NSAID patient carries a one-in-three one-year recurrence; the eradicated, protected patient carries 5%.^[26]

Revision summary

• Definition: full-thickness gastroduodenal defect; free versus sealed, duodenal versus gastric, under versus over 2 cm — each distinction maps to a different WSES branch.^[1]
• Epidemiology: 3.8–14 per 100,000/yr; average 30-day mortality 23.5%; a quarter attributable to NSAIDs; the population is older and sicker every decade.^[2][5][4]
• Boey verbatim: major medical illness, preoperative shock, perforation over 24 hours — mortality 0%, 10%, 45.5%, 100%. PULP: eight variables, AUC 0.83; hypoalbuminaemia is the strongest single predictor.^[9][11][13]
• Imaging: CT first (1C); plain film only when CT unavailable; free air visible in 30–85%; up to 12% of CTs normal — water-soluble contrast when suspicion persists.^[1]
• Delay kills by the hour: 2.4% per hour (Denmark), 6% per hour when shocked (NELA); surgery as soon as possible, especially over 70 (1B).^[16][17][1]
• NOM only for the contrast-proven sealed perforation without peritonitis; Crofts failed 28% at 12 hours; avoid endoscopic clipping or stenting (2C).^[1]
• Stable — laparoscopic (2B); unstable — open (1D); under 2 cm primary repair with the patch now optional; over 2 cm tailored, with the bailout list and 12% leak rate in mind.^[1]
• Gastric perforation: 8.8% are cancer; patch or resect with frozen section, biopsy judiciously (excisional biopsy adds morbidity), scope everyone afterwards.^[24][25]
• Antibiotics broad-spectrum and short (3–5 days); culture the fluid; no routine antifungals.^[1]
• H. pylori: test everyone, eradicate the positive — recurrence 5.2% versus 35.2% at one year (RR 0.15).^[26]

References32ShowHide

1. [1]Tarasconi A, Coccolini F, Biffl WL, Tomasoni M, et al. Perforated and bleeding peptic ulcer: WSES guidelines. World J Emerg Surg, 2020.PMID 31921329
2. [2]Lau JY, Sung J, Hill C, Henderson C, et al. Systematic review of the epidemiology of complicated peptic ulcer disease: incidence, recurrence, risk factors and mortality. Digestion, 2011.PMID 21494041
3. [3]Thorsen K, Søreide JA, Kvaløy JT, Glomsaker T, et al. Epidemiology of perforated peptic ulcer: age- and gender-adjusted analysis of incidence and mortality. World J Gastroenterol, 2013.PMID 23372356
4. [4]Dadfar A, Edna TH Epidemiology of perforating peptic ulcer: A population-based retrospective study over 40 years. World J Gastroenterol, 2020.PMID 32994689
5. [5]Svanes C Trends in perforated peptic ulcer: incidence, etiology, treatment, and prognosis. World J Surg, 2000.PMID 10658061
6. [6]Tai FWD, McAlindon ME Non-steroidal anti-inflammatory drugs and the gastrointestinal tract. Clin Med (Lond), 2021.PMID 33762373
7. [7]Scally B, Emberson JR, Spata E, Reith C, et al. Effects of gastroprotectant drugs for the prevention and treatment of peptic ulcer disease and its complications: a meta-analysis of randomised trials. Lancet Gastroenterol Hepatol, 2018.PMID 29475806
8. [8]Grassi R, Romano S, Pinto A, Romano L Gastro-duodenal perforations: conventional plain film, US and CT findings in 166 consecutive patients. Eur J Radiol, 2004.PMID 15093233
9. [9]Boey J, Choi SK, Poon A, Alagaratnam TT Risk stratification in perforated duodenal ulcers. A prospective validation of predictive factors. Ann Surg, 1987.PMID 3800459
10. [10]Lohsiriwat V, Prapasrivorakul S, Lohsiriwat D Perforated peptic ulcer: clinical presentation, surgical outcomes, and the accuracy of the Boey scoring system in predicting postoperative morbidity and mortality. World J Surg, 2009.PMID 18958520
11. [11]Møller MH, Engebjerg MC, Adamsen S, Bendix J, et al. The Peptic Ulcer Perforation (PULP) score: a predictor of mortality following peptic ulcer perforation. A cohort study. Acta Anaesthesiol Scand, 2012.PMID 22191386
12. [12]Thorsen K, Søreide JA, Søreide K Scoring systems for outcome prediction in patients with perforated peptic ulcer. Scand J Trauma Resusc Emerg Med, 2013.PMID 23574922
13. [13]Thorsen K, Søreide JA, Søreide K What is the best predictor of mortality in perforated peptic ulcer disease? A population-based, multivariable regression analysis including three clinical scoring systems. J Gastrointest Surg, 2014.PMID 24610235
14. [14]Saafan T, El Ansari W, Al-Yahri O, Eleter A, et al. Assessment of PULP score in predicting 30-day perforated duodenal ulcer morbidity, and comparison of its performance with Boey and ASA, a retrospective study. Ann Med Surg (Lond), 2019.PMID 31193430
15. [15]Christanto AGR, Adrianto AA Comparative accuracy of Boey score and PULP score for predicting prognosis in patients with perforated peptic ulcer: A systematic review and meta-analysis. J Visc Surg, 2026.PMID 42031639
16. [16]Buck DL, Vester-Andersen M, Møller MH Surgical delay is a critical determinant of survival in perforated peptic ulcer. Br J Surg, 2013.PMID 23754645
17. [17]Boyd-Carson H, Doleman B, Cromwell D, Lockwood S, et al. Delay in Source Control in Perforated Peptic Ulcer Leads to 6% Increased Risk of Death Per Hour: A Nationwide Cohort Study. World J Surg, 2020.PMID 31664496
18. [18]Gupta S, Alawad AA, Dacosta K, Mahmoud A, et al. Operative versus non-operative management for perforated peptic ulcer disease. Ann Med Surg (Lond), 2022.PMID 36268326
19. [19]Lau WY, Leung KL, Kwong KH, Davey IC, et al. A randomized study comparing laparoscopic versus open repair of perforated peptic ulcer using suture or sutureless technique. Ann Surg, 1996.PMID 8757375
20. [20]Tan S, Wu G, Zhuang Q, Xi Q, et al. Laparoscopic versus open repair for perforated peptic ulcer: A meta analysis of randomized controlled trials. Int J Surg, 2016.PMID 27504848
21. [21]Sokhal BS, Mohamedahmed A, Zaman S, Wuheb AA, et al. Laparoscopic versus open repair for peptic ulcer perforation: a systematic review, meta-analysis and trial sequential analysis of randomised controlled trials. Time to conclude! Ann R Coll Surg Engl, 2025.PMID 39361132
22. [22]Chalmers KA, Lee MJ, Cousins SE, Peckham Cooper A, et al. Laparoscopic versus open repair of perforated peptic ulcer: systematic scoping review and in-depth evaluation of existing evidence. BJS Open, 2025.PMID 40045705
23. [23]Istl AC, Gray DK Roscoe R. Graham: An enduring legacy in the 21st century. J Trauma Acute Care Surg, 2017.PMID 27806000
24. [24]Leeman MF, Skouras C, Paterson-Brown S The management of perforated gastric ulcers. Int J Surg, 2013.PMID 23454244
25. [25]Koca F, Koch C, Schulze F, Pession U, et al. Excisional biopsy of perforated gastric ulcer: mandatory or potentially harmful? Langenbecks Arch Surg, 2024.PMID 38963438
26. [26]Tomtitchong P, Siribumrungwong B, Vilaichone RK, Kasetsuwan P, et al. Systematic review and meta-analysis: Helicobacter pylori eradication therapy after simple closure of perforated duodenal ulcer. Helicobacter, 2012.PMID 22404446
27. [27]Wong CS, Chia CF, Lee HC, Wei PL, et al. Eradication of Helicobacter pylori for prevention of ulcer recurrence after simple closure of perforated peptic ulcer: a meta-analysis of randomized controlled trials. J Surg Res, 2013.PMID 23158404
28. [28]El-Nakeeb A, Fikry A, Abd El-Hamed TM, Fouda el Y, et al. Effect of Helicobacter pylori eradication on ulcer recurrence after simple closure of perforated duodenal ulcer. Int J Surg, 2009.PMID 19138577
29. [29]Bose AC, Kate V, Ananthakrishnan N, Parija SC Helicobacter pylori eradication prevents recurrence after simple closure of perforated duodenal ulcer. J Gastroenterol Hepatol, 2007.PMID 17295765
30. [30]Abouelazayem M, Jain R, Wilson MSJ, Martinino A, et al. Global 30-day morbidity and mortality of surgery for perforated peptic ulcer: GRACE study. Surg Endosc, 2024.PMID 38886232
31. [31]Møller MH, Adamsen S, Wøjdemann M, Møller AM Perforated peptic ulcer: how to improve outcome? Scand J Gastroenterol, 2009.PMID 18752147
32. [32]Hudnall A, Bardes JM, Coleman K, Stout C, et al. The surgical management of complicated peptic ulcer disease: An EAST video presentation. J Trauma Acute Care Surg, 2022.PMID 35358158