a b s t r a c t

Background: Although surgical intervention is the favorable Treatment modality for Perforated peptic ulcer, non- surgical treatment is another option. The aim of this study is to analyze the results of conservative treatment for perforated peptic ulcer.

Methods: Between 2003 and 2014, 403 patients were admitted to our hospital for perforated peptic ulcer, and 383 patients underwent surgery, whereas 20 were allocated to conservative treatment. The results of nonsurgical in- tervention in these patients were analyzed retrospectively.

Results: The overall mortality rate of conservative treatment was 40%. Eleven patients remained hospitalized less than 2 weeks; among them, patients with a high (>=IV) American Society of Anesthesiologists class at admission had higher mortality than those with a low (b IV) American Society of Anesthesiologists class (83.3% vs 0%, P =

.015). However, when patients remained hospitalized longer than 2 weeks, the mortality rates did not differ be- tween patients with the low and high American Society of Anesthesiologists classes. Eight patients presented with a high American Society of Anesthesiologists class, of which 3 received early enteral feeding, and all of them survived. In contrast, the survival of patients without early enteral feeding was 0%, suggesting that early en- teral feeding improved survival of patients with the high American Society of Anesthesiologists class (P = .018). Conclusions: A higher American Society of Anesthesiologists class correlated with mortality in patients undergo- ing conservative treatment during the first 2 weeks of hospitalization. Early enteral feeding might improve the outcome of conservative treatment in patients with high American Society of Anesthesiologists class.

(C) 2016

1. Introduction

In the past century, Peptic ulcer disease was a common health prob- lem. Since histamine-2 receptor (H-2) blockers and proton pump inhib- itors (PPIs) were introduced in the 1970s [1,2], these antisecretory drugs have played an important role in the treatment of peptic ulcer dis- ease, according to the principle of “no acid, no ulcer” [3]. In addition, Marshall and Warren [4] and discovered Helicobacter pylori (H. pylori) in 1982 and proved the crucial involvement of this pathogen in the de- velopment of peptic ulcers. Furthermore, the eradication of H. pylori re- duces the recurrence of peptic ulcer [5]. Because of the aforementioned findings and advancements, the incidence of uncomplicated peptic

? This study was supported by the Research Fund of Tri-Service General Hospital (TSGH- C103-067).

* Corresponding author at: Division of gastroenterology, Tri-Service General Hospital, No.325, Sec 2, Cheng-Gong Rd., Neihu, Taipei 114, Taiwan. Tel.: +886 287927409;

fax: +886 287927139.

E-mail address: [email protected] (H.-H. Lin).

ulcer has declined [6,7]. However, the incidence of PErforated peptic ulcer (PPU) has remained unchanged in the past decades [7-9]. It may be that the increased use of nonsteroidal anti-inflammatory drugs or as- pirin in elderly patients increased the risk of PPU [10,11].

Perforated peptic ulcer is an emergent condition, and surgical inter- vention is the preferred Therapeutic treatment modality [12,13]. At first, broad-spectrum antibiotics should be administered intravenously, and then simple closure, omental patch repair, or laparoscopic treatment is performed in most patients, followed by antisecretory treatment and

H. pylori eradication, if indicated [13-15]. The mortality rate ranges be- tween 4% and 30% [7]. When patients are unsuitable for Surgical repair, nonsurgical treatment involving fasting, nasogastric tube suction, intra- venous broad-spectrum antibiotics, and antisecretory therapy is anoth- er option for PPU [16-20]. In 1946, Taylor [20] first reported the results of conservative treatment for PPU, which yielded promising results, with 11% mortality rate after conservative treatment for perforated du- odenal ulcers. In addition, in 1989, Crofts et al [21] reported a random- ized trial in which similar outcomes were reported for nonsurgical treatment and emergency surgery, and the mortality rate was 5% in both groups. Gul et al [17] reported an overall mortality rate of 3% in

http://dx.doi.org/10.1016/j.ajem.2016.05.045

0735-6757/(C) 2016

patients with perforated Duodenal ulcer managed conservatively. How- ever, these results are not widely accepted. In 1971, Cohen et al [22] re- ported their experience in the management of 852 patients with PPU, in which 87 patients received conservative treatment only, and their mor- tality rate was 100%, which was significantly higher than the 9% in the operative treatment group in the same study. The reason for this huge difference in mortality rates between conservative and operative treat- ments in previous studies may be selection bias, and in such studies, only patients with a low risk were recruited. Further, in those studies, patients managed conservatively would be switched to surgical treat- ment immediately if the former treatment was unsuccessful.

In 1987, Boey et al [13,23] reported that the major medical illness, preoperative shock, and prolonged perforation (over 24 hours) are risk factors for patients with perforated duodenal ulcer and can predict the outcome of surgical treatment accurately. Kocer et al also showed that old age, delayed surgery, presence of shock, high American Society of Anesthesiologists (ASA) class, and definitive surgery are poor prog- nostic factors for patients undergoing emergency surgery for PPU [7]. Larkin et al [24] performed a retrospective study of patients undergoing conservative treatment and reported that the mortality rates of patients with perforated duodenal ulcers were lower in the group with ASA clas- ses I-III than in the group with ASA classes IV-V (0% vs 52.9%). The above reports suggest that prognostic factors are crucial for the outcome of both surgical and nonsurgical treatment of patients with PPU.

In this study, we aimed to retrospectively analyze the results of con- servative treatment in patients undergoing nonsurgical treatment for PPU in a teaching hospital. The clinical characteristics of our patients

3. Results

Fig. 1. Plain abdominal radiograph showing the enteral feeding.

were first examined, and subsequently, we analyzed the putative prog- nostic factors and determined whether these factors were important for the entire course of conservative treatment.

Materials and methods

Patients

In this retrospective study, medical records of patients who present- ed to Tri-Service General Hospital with PPU, during a 10-year period be- tween January 2003 and February 2014, were reviewed. The diagnosis of PPU was based on radiological (chest radiography or computed to- mography scans), endoscopic, or operative findings. This study focused on patients who did not undergo surgical intervention for PPU. Patient age, sex, ASA class, Acute Physiology and Chronic Health Evaluation II (APACHE II) score, clinical presentation, management mode, mortality, and duration of hospital stay of these patients were analyzed. Nonsurgi- cal treatment of these patients with PPU consisted of fasting, nasogastric tube suction, intravenous fluids, intravenous broad-spectrum antibi- otics, and antisecretory therapy with PPIs. Some patients underwent en- doscopic placement of enteral feeding tubes, which bypassed the perforated site [25-27] and received early enteral feeding before the PPU healed (Fig. 1). Patients undergoing conservative treatment were categorized into 2 groups according to the duration of hospital stay (>= or b 15 days). The mortality rate of these patients in the 2 groups was calculated. The study was approved by the institutional review board of Tri-Service General Hospital.

The clinical features of PPU in patients who did not undergo surgical intervention

During 2003-2014, 403 patients were admitted to the Tri-Service General Hospital for PPU. Three hundred eighty-three patients underwent surgery, whereas 20 patients (median age, 74 years; range, 31-99) received conservative treatment because they were unsuitable or unwilling to undergo surgery. Five patients were men, and 15 were women. Fourteen patients had Shock Index (heart rate/systolic blood pressure) b 1 at admission. Median ASA class was III (range, I-V). Median APACHE II score was 10.5 (range, 5-46). Median duration of hospital stay was 14 days (range, 1-78). Of 20 patients, 8 died of sepsis with multiple-organ dysfunction, and the overall mortality rate of conserva- tive treatment was 40%. Patients were divided into 2 groups according to the duration of hospital stay (>= or b 15 days), and 9 patients remained hospitalized longer than 2 weeks (Table 1). There was no difference in age, ASA class, and APACHE II score between these 2 groups. There was no significant difference in the percentage of patients with clinical improvement after conservative treatment for 12 hours between these 2 groups. Female sex predominated in both groups. The propor- tion of shock index 1 or higher at admission was higher in patients with lengths of hospital stays shorter than 2 weeks than in patients with lengths of hospital stay longer than 2 weeks (45.5% vs 11.1%); however, the difference in these values was not significant (P = .16). Similarly, mortality rates were higher in patients with shorter hospital

Table 1

Clinical characteristics of patients with PPU according to the duration of hospital stay

Hospital stay

b 15 d

Hospital stay P

>= 15 d

Statistical analysis

All data were presented as median and range for continuous vari- ables or number and percentage for categorical variables. Statistical analysis was performed using SPSS statistics software, version 18 (IBM Co, Somers, New York). Continuous variables were compared using Mann-Whitney U tests, and categorical variables were compared using Fisher exact test. All reported P were 2-tailed, and P b .05 was consid- ered significant.

No. 11 9

Median age (range) 74 (48-99) 74 (31-97) .94

Gender (male/female) 2/9 3/6 .62 Shock index >= 1 at admission, no. (%) 5 (45.5) 1 (11.1) .16 Median ASA class (range) III (II-V) III (I-IV) .19 Median APACHE II score (range) 14 (5-46) 10 (7-23) .37

With clinical improvement in 12 h, no. (%) 5 (45.5) 3 (33.3) .67

Mortality, no. (%) 6 (54.5) 2 (22.2) .19

Median hospital stay (range) 8 (1-14) 34 (15-78) b.001 Abbreviations: Shock index, heart rate (min)/systolic blood pressure (mm Hg).

stays than in patients with longer hospital stays (54.5% vs 22.2%), but the difference was not significant (P = .19).

Table 3

Outcome of patients with poor prognostic factors after early enteral feeding

Survival (%) P

Factors related to survival rate of patients with different durations of hospital stays

In patients with hospital stays shorter than 2 weeks (Table 2), the survival rate was significantly higher in patients who had a shock index less than 1 at admission compared with those with a shock index of 1 or higher at admission (83.3% vs 0%, P = .015). Patients who experienced clinical improvement after 12 hours of starting con- servative treatment had a higher survival rate compared with patients without clinical improvement after 12 hours (100% vs 0%, P = .002). Similarly, the survival rate was significantly higher in patients with ASA class I-III than that of patients with ASA class IV-V (83.3% vs 0%, P = .015). Furthermore, the survival rate was significantly higher in pa- tients with APACHE II score less than10 than in patients with APACHE II score of 10 or higher (100% vs 14.3%, P = .015).

In patients with hospital stay longer than 2 weeks (Table 2), there was no significant difference in the survival rate between patients with shock index less than 1 and 1 or higher at admission (75% vs 100%, P = 1.0). The survival rate was not significantly different between the patients with and without clinical improvement after 12 hours of starting conservative treatment (66.7% vs 83.3%, P = 1.0). There was also no statistically significant difference in the survival rates between patients with ASA score I-III and those with ASA score IV-V (66.7% vs 100%, P = .5). In addition, there was no significant difference in survival rates between patients with APACHE II score less than 10 and those with APACHE scores of 10 or higher (75% vs 80%, P = 1.0).

Early enteral feeding might improve the outcome of patients with high ASA class or APACHE II score

Three patients underwent endoscopic placement of an enteral feed- ing tube and received early enteral feeding. They all presented with a high ASA class or high APACHE II score upon admission and did not present clinical improvement within 12 hours of starting conservative treatment. Among the patients with ASA class IV-V, the survival rate of patients with and without early enteral feeding was 100% and 0%, re- spectively (P = .018; Table 3). Similarly, among the patients with APACHE II score of 10 or higher, the survival of patients with and with- out early enteral feeding was 100% and 22.2%, respectively (P = .045; Table 3). These results suggested that the early enteral feeding

Table 2 The correlation of survival rate with the shock index, ASA class, and APACHE II score in pa- tients with a Short duration of hospital stays

Survival (%) P

Hospital stay b 15 d

Shock index >= 1 at admission	0/5 (0)	.015
Shock index b1 at admission	5/6 (83.3)
With clinical improvement after 12 h	5/5 (100)	.002
Lack of clinical improvement after 12 h	0/6 (0)
ASA class I-III	5/6 (83.3)	.015
ASA class IV-V	0/5 (0)
APACHE II score >= 10	1/7 (14.3)	.015
APACHE II score b10 Hospital stay >= 15 d	4/4 (100)
Shock index >= 1 at admission	1/1 (100)	1.0
Shock index b1 at admission	6/8 (75)
With clinical improvement after 12 h	2/3 (66.7)	1.0
Lack of clinical improvement after 12 h	5/6 (83.3)
ASA class I-III	4/6 (66.7)	.5
ASA class IV-V	3/3 (100)
APACHE II score >= 10	4/5 (80)	1.0
APACHE II score b10	3/4 (75)

Abbreviations: Shock index, heart rate (min)/systolic blood pressure (mm Hg).

ASA class IV-V

With early enteral feeding	3/3 (100)	.018
Without early enteral feeding	0/5 (0)
APACHE II score >=10 With early enteral feeding	3/3 (100)	.045
Without early enteral feeding	2/9 (22.2)
Lack of clinical improvement after 12 h With early enteral feeding	3/3 (100)	.045
Without early enteral feeding	2/9 (22.2)

improved survival of patients with a high ASA class or high APACHE II score. Among patients who lacked clinical improvement after 12 hours of starting conservative treatment, the survival rate was higher for patients with early enteral feeding than that for patients without early enteral feeding (100% vs 22.2%, P = .045).

4. Discussion

Perforated peptic ulcer is not a rare complication of peptic ulcer [3], and an emergent surgical repair is the preferred intervention for PPU [12,13]. Conservative treatment is another option for PPU [16-21] based on the observation that perforations may be gradually sealed by the surrounding viscera [15,28]. However, conservative treatment has not been widely accepted. Therefore, the surgical intervention is the preferred choice for these patients currently [12,13], and nonsurgical treatment is preserved for these patients unsuitable or unwilling to un- dergo operation [15]. Previous retrospective studies have revealed that most patients with PPU underwent surgery and that only 2% to 10% of these patients did not undergo surgical intervention [9,19,22,29]. In our study, 4.9% patients with PPU underwent only nonsurgical treat- ment during the study period.

In 1967, a retrospective study reviewed 402 patients with PPU and showed that the mortality rates associated with nonsurgical and surgi- cal treatment for PPU were 72.5% and 7.5%, respectively [29]. Similarly, in 1971, Cohen et al [22] also reported that the nonsurgical mortality was significant higher than the operative mortality (100% vs 9%). Both retrospective studies suggested that nonsurgical intervention was asso- ciated with a high mortality rate. However, antisecretory agents, such as H-2 blockers and PPIs, were not introduced until the 1970s [1,2]. There- fore, no effective antisecretory agent was available during the study pe- riod of the above studies. Bucher et al [19] retrospectively reviewed patients with PPU between 1978 and 2004, and the mortality rate of conservative treatment was 30%, which was much lower than those re- ported in previous retrospective studies. In a subgroup analysis, they found that the patients who received H-2 blockers as antisecretory drugs had higher mortality rates than those who received PPIs (64% vs 11%, P = .008), suggesting that advances in antisecretory therapy im- proved the outcome of conservative treatment for PPU [19]. In our study, all patients received PPIs, indicating that these antisecretory drugs were routinely applied to the patients with PPU at our hospital.

In 1989, Crofts et al [21] reported a prospective randomized trial that compared the results of conservative treatment for PPU with that of emergent surgical intervention. In this study, patients who were allocat- ed to receive conservative treatment underwent surgery when their clinical symptoms and signs did not improve after 12 hours. Notably, the mortality rates of nonsurgical and emergent surgical treatment were both 5% [21]. Gul et al [17] also conducted a prospective study to define the role of conservative treatment for perforated duodenal ul- cers. As in previous study, conservative treatment would be switched to surgical treatment when there was no clinical improvement after 12 to 16 hours of treatment with conservative treatment. The overall mortality rate of conservative treatment was 3% in this study [17]. Two prospective studies described above showed a low mortality rate

after conservative treatment; however, in our study, the mortality rate of conservative treatment was 40%, which was higher than that reported in those prospective studies [17,20,21]. The reasons for such difference may be that the patients who were recruited in those pro- spective studies were carefully selected and closely monitored. For ex- ample, the patients who were hemodynamically unstable or had been perforated for more than 24 hours were excluded in those prospective studies [17,21]. It has been proven that shock and delay in treatment are associated with a high mortality in patients with PPU [23,30,31]; therefore, the exclusion of patients with poor prognostic factors may re- sult in a better outcomes with conservative treatment.

As shown in Table 2, mortality rates were higher for the patients with an unstable hemodynamic status, higher ASA class, or higher APACHE II score during the first 2 weeks of hospitalization, suggesting that the poor condition at presentation of a patient with PPU is associat- ed with poor outcome during the first 2 weeks of conservative treat- ment. Therefore, an unstable hemodynamic status, higher ASA class, and higher APACHE II scores are poor predictive factors of patient outcomes of conservative treatment. In addition, the lack of clinical improvement after 12 hours of starting conservative treatment may be also a poor predictive factor. In the above prospective studies, the patients who did not have a clinical improvement after 12 hours with conservative treatment were switched to surgical intervention immedi- ately [17,21]; thus, only the patients without this poor predictive factor remained in the group of conservative treatment. This may be another reason for the low mortality rates after conservative treatment in these prospective studies. In our study, survival rates were significantly higher for patients without this poor predictive factor than for patients with the poor predictive factor during the first 2 weeks of hospitaliza- tion (Table 2). This result confirmed that the lack of clinical improve- ment after 12 hours of starting conservative treatment is indeed a poor predictive factor of patient outcomes. In contrast, the poor predic- tive factors (such as unstable hemodynamic status, high ASA class, high APACHE II score, and lack of clinical improvement in 12 hours) were not associated with the mortality of patients with conservative treatment after 2 weeks of hospitalization (Table 2). These results suggest that the patients’ condition at presentation with PPU is crucial for the surviv- al during the first 2 weeks, but it may not be so important for survival when patients could remain hospitalized longer than 2 weeks with con- servative treatment.

It has been shown that conservative treatment rather than

surgical intervention of PPU is associated with a longer hospital stay [15,32,33]; thus, patients who underwent nonsurgical treatment had a prolonged duration of fasting and parental nutrition period. Moreover, parental nutrition is associated with an increased infection rate, im- paired mucosal function, and increased hospital costs [34-36]. In con- trast, enteral nutrition was shown to reduce morbidity in critically ill patients [34,35]. Therefore, early enteral nutrition should be considered to improve the outcome of patients with PPU who undergo conservative treatment. Our previous reports showed that postpyloric enteral feed- ing could provide reliable enteral nutrition [25,26]. In this study, pa- tients who had poor predictive factors had higher survival rates with early enteral nutrition compared with those patients without early en- teral nutrition (Table 3). This result suggests that early enteral feeding should be considered when patients present poor predictive factors and undergo nonsurgical treatment for PPU. However, this was a retro- spective study, and the number of patients with early enteral feeding was not large enough to make a definite conclusion. Therefore, a ran- domized controlled trial is needed to address this issue.

In conclusion, the poor general condition of patients with PPU at pre- sentation is a poor prognostic factor associated with mortality during the first 2 weeks of conservative treatment. Lack of clinical improve- ment after 12 hours of conservative treatment is also a poor predictive factor for mortality after nonsurgical treatment of PPU. However, the roles of these factors are not crucial when the patients could survive after conservative treatment for 2 weeks. Early enteral feeding may

improve the outcome of patients with poor prognostic factors, but this issue needs to be addressed in a further study.

References

1. Henn RM, Isenberg JI, Maxwell V, et al. Inhibition of gastric acid secretion by cimet- idine in patients with duodenal ulcer. N Engl J Med 1975;293:371-5. http://dx.doi. org/10.1056/NEJM197508212930802.
2. Fellenius E, Berglindh T, Sachs G, et al. Substituted benzimidazoles inhibit gastric acid secretion by blocking (H⁺ + K⁺)ATPase. Nature 1981;290:159-61 [http:// www.nature.com/nature/journal/v290/n5802/abs/290159a0.html].
3. Malfertheiner P, Chan FK, McColl KE. Peptic ulcer disease. Lancet 2009;374:1449-61. http://dx.doi.org/10.1016/S0140-6736(09)60938-7.
4. Marshall BJ, Warren JR. Unidentified curved bacilli on gastric epithelium in active chronic gastritis. Lancet 1983;1:1273-5. http://dx.doi.org/10.1016/S0140-6736(83)92719-8.
5. Hentschel E, Brandstatter G, Dragosics B, et al. Effect of ranitidine and amoxicillin plus metronidazole on the eradication of helicobacter pylori and the recurrence of duodenal ulcer. N Engl J Med 1993;328:308-12. http://dx.doi.org/10.1056/ NEJM199302043280503.
6. Cai S, Garcia Rodriguez LA, Masso-Gonzalez EL, et al. Uncomplicated peptic ulcer in the UK: trends from 1997 to 2005. Aliment Pharmacol Ther 2009;30:1039-48. http://dx.doi.org/10.1111/j.1365-2036.2009.04131.x.
7. Kocer B, Surmeli S, Solak C, et al. Factors affecting mortality and morbidity in pa- tients with Peptic ulcer perforation. J Gastroenterol Hepatol 2007;22:565-70. http://dx.doi.org/10.1111/j.1440-1746.2006.04500.x.
8. Svanes C. Trends in perforated peptic ulcer: incidence, etiology, treatment, and prog- nosis. World J Surg 2000;24:277-83 [http://www.ncbi.nlm. nih.gov/pubmed/ 10658061].
9. Svanes C, Salvesen H, Stangeland L, et al. Perforated peptic ulcer over 56 years. Time trends in patients and disease characteristics. Gut 1993;34:1666-71 [http://www. ncbi.nlm.nih.gov/pubmed/8282252].
10. Higham J, Kang JY, Majeed A. Recent trends in admissions and mortality due to pep- tic ulcer in England: increasing frequency of haemorrhage among older subjects. Gut 2002;50:460-4 [http://www.ncbi.nlm.nih.gov/pubmed/11889062].
11. Gisbert JP, Legido J, Garcia-Sanz I, et al. Helicobacter pylori and perforated peptic ulcer prevalence of the infection and role of Non-steroidal anti-inflammatory drugs. Dig Liver Dis 2004;36:116-20. http://dx.doi.org/10.1016/j.dld.2003.10.011.
12. Behrman SW. Management of complicated peptic ulcer disease. Arch Surg 2005; 140:201-8. http://dx.doi.org/10.1001/archsurg.140.2.201.
13. Lohsiriwat V, Prapasrivorakul S, Lohsiriwat D. Perforated peptic ulcer: clinical pre- sentation, surgical outcomes, and the accuracy of the Boey scoring system in predicting Postoperative morbidity and mortality. World J Surg 2009;33:80-5. http://dx.doi.org/10.1007/s00268-008-9796-1.
14. Moller MH, Adamsen S, Wojdemann M, et al. Perforated peptic ulcer: how to im- prove outcome? Scand J Gastroenterol 2009;44:15-22. http://dx.doi.org/10.1080/ 00365520802307997.
15. Bertleff MJ, Lange JF. Perforated peptic ulcer disease: a review of history and treat- ment. Dig Surg 2010;27:161-9. http://dx.doi.org/10.1159/000264653.
16. Visick AH. Conservative treatment of acute perforated peptic ulcer. Br Med J 1946;2: 941-4 [http://www.ncbi.nlm.nih.gov/pubmed/20341242].
17. Gul YA, Shine MF, Lennon F. Non-operative management of perforated duodenal ulcer. Ir J Med Sci 1999;168:254-6 [http://www.ncbi.nlm.nih.gov/pubmed/ 10624365].
18. Brinkman JM, Oddens JR, Van Royen BJ, et al. Non-operative treatment for perforated gastro-duodenal peptic ulcer in Duchenne muscular dystrophy: a case report. BMC Surg 2004;4:1. http://dx.doi.org/10.1186/1471-2482-4-1.
19. Bucher P, Oulhaci W, Morel P, et al. Results of conservative treatment for perforated gastroduodenal ulcers in patients not eligible for surgical repair. Swiss Med Wkly 2007;137:337-40 http://dx.doi.org/2007/23/smw-11796.
20. Taylor H. Perforated peptic ulcer; treated without operation. Lancet 1946;2:441-4. http://dx.doi.org/10.1016/S0140-6736(46)90971-3.
21. Crofts TJ, Park KG, Steele RJ, et al. A randomized trial of nonoperative treatment for perforated peptic ulcer. N Engl J Med 1989;320:970-3. http://dx.doi.org/10.1056/ NEJM198904133201504.
22. Cohen MM. Treatment and mortality of perforated peptic ulcer: a survey of 852 cases. Can Med Assoc J 1971;105:263-9 [http://www.ncbi.nlm.nih.gov/pubmed/ 5563345].
23. Boey J, Choi SK, Poon A, et al. Risk stratification in perforated duodenal ulcers. A pro- spective validation of predictive factors. Ann Surg 1987;205:22-6 [http://www.ncbi. nlm.nih.gov/pubmed/3800459].
24. Larkin JO, Bourke MG, Muhammed A, et al. Mortality in perforated duodenal ulcer depends upon pre-operative risk: a retrospective 10-year study. Ir J Med Sci 2010; 179:545-9. http://dx.doi.org/10.1007/s11845-010-0515-1.
25. Chang WK, McClave SA, Chao YC. Simplify the technique of nasoenteric feeding tube placement with a modified suture tie. J Clin Gastroenterol 2005;39:47-9 [http:// www.ncbi.nlm.nih.gov/pubmed/15599210].
26. Lin HH, Hsieh TY, Chu HC, et al. Transpyloric tube feeding for 506 consecutive pa- tients with delayed gastric emptying. J Intern Med Taiwan 2006;17:52-60 [http:// www.tsim.org.tw/journal/jour17-2/07.PDF].
27. Ginsberg GG, Kochman ML, Norton ID, et al. Clinical gastrointestinal endoscopy. 2nd ed. Elsevier Health Sciences; 2011.
28. Donovan AJ, Berne TV, Donovan JA. Perforated duodenal ulcer: an alternative thera- peutic plan. Arch Surg 1998;133:1166-71 [http://www.ncbi.nlm.nih.gov/pubmed/ 9820345].
29. Devitt JE, Taylor GA. Perforated peptic ulcer. Can Med Assoc J 1967;96:519-23 [http://www.ncbi.nlm.nih.gov/pubmed/6019350].
30. Saber A, Gad MA, Ellabban GM. Perforated duodenal ulcer in high risk patients: is percutaneous drainage justified? N Am J Med Sci 2012;4:35-9. http://dx.doi.org/ 10.4103/1947-2714.92902.
31. Prabhu V, Shivani A. An overview of history, pathogenesis and treatment of perforat- ed peptic ulcer disease with evaluation of prognostic scoring in adults. Ann Med Health Sci Res 2014;4:22-9. http://dx.doi.org/10.4103/2141-9248.126604.
32. Fujii Y, Asato M, Taniguchi N, et al. Sonographic diagnosis and successful nonopera- tive management of sealed perforated duodenal ulcer. J Clin Ultrasound 2003;31: 55-8. http://dx.doi.org/10.1002/jcu.10125.
33. Zittel TT, Jehle EC, Becker HD. Surgical management of peptic ulcer disease today-indication, technique and outcome. Langenbecks Arch Surg 2000;385: 84-96 [http://www.ncbi.nlm.nih.gov/pubmed/10796046].
34. Nicholas JM, Cornelius MW, Tchorz KM, et al. A two institution experience with 226 endoscopically placed jejunal feeding tubes in critically ill surgical patients. Am J Surg 2003;186:583-90. http://dx.doi.org/10.1016/j.amjsurg.2003.09.005.
35. Marik PE, Zaloga GP. Early enteral nutrition in acutely ill patients: a systematic re- view. Crit Care Med 2001;29:2264-70 [http://www.ncbi.nlm.nih.gov/pubmed/ 11801821].
36. Braun J, Bein T, Wiese CH, et al. Enteral feeding tubes for critically ill patients. Anaesthesist 2011;60:352-65. http://dx.doi.org/10.1007/s00101-010-1800-0.