ABSTRACT

KEY WORDS: Burn disease, Burn injury, Injury severity index, Ulcers

INTRODUCTION
Gastroduodenal complications can occur in 15% to 25% of patients with thermal burn injuries. The most frequent complication of acute gastroduodenal ulcers in patients with severe burn injuries is bleeding.^1-3 After thermal injuries, the mortality rate for bleeding from acute gastroduodenal ulcers remains high and can range from 25% to 75% of these patients.^3,4 Treatment of patients with bleeding from acute gastroduodenal ulcers after severe burns, despite the introduction of new surgical, endoscopic, and drug treatment methods into practice, remains an urgent and complex problem in surgery.^5-7

Treatment of stress gastroduodenal ulcers complicated by bleeding presents significant difficulties. There is no uniform management tactic for patients with this complication. To date, the effectiveness of various methods of endoscopic hemostasis and therapy with antisecretory drugs remains unclear.^7,8

Because clinical signs of the appearance of acute gastroduodenal ulcers and their complications in patients with severe burns are often masked by other symptoms during severe systemic organ dysfunction, most patients receive a late diagnosis.^9,10 There is no doubt about the need for preventive measures to reduce the likelihood of stress ulcers in these patients. However, methods of prevention and categorization of patients remain unclear.^11,12

In this study, we compared methods of treatment of patients with bleeding from acute gastroduodenal ulcers after severe burns.

MATERIALS AND METHODS
In the burn department of the Samarkand City Medical Association, 80 patients with serious burn injuries and bleeding from acute gastroduodenal ulcers have been treated over the past 10 years. The average age of patients was 51 ± 16 years, with 46.4% of patients over 60 years old. Total area of thermal damage to the skin, which indicated the severity of thermal injury, was equal to 36.7 ± 7.7% of the body surface. Acute gastric ulcers were the source of bleeding in 38 patients (47.5%), acute duodenal ulcers were the source in 19 patients (23.7%), and a combined localization of ulcers was the source in 23 patients (28.8%).

Analysis of feces for hidden blood in patients with burns can play an important role in the diagnosis of stress ulcers complicated by bleeding. The benzidine test (Gregersen modification) shows positivity with blood loss of 2 to 5 mL. We performed the Gregersen benzidine reaction in 58 patients. The other 22 patients underwent the Weber-Van Deen reaction, which shows a positive result with blood loss of 30 to 50 mL.

Patients were divided into 2 groups: the main group (n = 42) and the control group (n = 38). The main group received antiulcer therapy for 7 days (consisting of pantoprazole 40 mg/day intravenously, famotidine 20 mg/3 times per day intravenously, amoxiclav 50 mg/kg/day, and tinidazole 50 mg/kg/day). In addition, patients received prokinetics/blockers of dopamine D₂ receptors (domperidone) (suspension of 10 mg through a nasogastric probe or inside), 100 mL intravenous L-arginine, and active surgical tactics (early necrotomy, necrectomy with autodermoplasty), which were used to correct vascular endothelial dysfunction. The main group included 12 patients with a high propensity for gastroduodenal complications; these patients had a history of chronic gastropathy or were constantly taking ulcerogenic drugs.

Patients in the control group were treated only with H₂-receptor blockers for antiulcer therapy and traditional treatment with wait-and-see surgical tactics. In this group, 9 patients had a high propensity for gastroduodenal complications; these patients had a history of chronic gastropathy or were constantly taking ulcerogenic drugs.

Esophagogastroduodenofibroscopy was performed proactively during treatment on days 1, 3, 7, 14, and 21 and, if necessary, at a later date. We used the Forrest endoscopic classification to assess ulcerative bleeding. The following methods were used to perform endoscopic hemostasis: injection, argon plasma coagulation, and radiowave coagulation. Angiogenesis, vascular endothelial growth factor (VEGF) levels, and gastrointestinal tract pathology were also studied. In addition, we also studied the role of Helicobacter pylori on the formation of acute gastroduodenal ulcers/complicated gastroduodenal bleeding in our patients. We used centrifuged blood serum for enzyme immunoassays to study VEGF levels. Twelve healthy individuals were selected for a comparison control.

We used Microsoft Excel 2013 and SPSS version 22.0 for our statistical analyses.

RESULTS
In our 80 patients with severe burns and gastroduodenal bleeding, bleeding was observed in 29 patients (36.2%) in the form of vomiting or coffee grounds from a nasogastric probe. Bleeding in the form of melena was observed in 24 patients (30%) patients. Simultaneous observation of these 2 symptoms was detected in 27 patients (33.8%).

In patients with gastroduodenal bleeding, excitement (in 3 patients), inhibition (in 4 patients), and a feeling of thirst or hunger (in 7 patients) were observed. Tachycardia (115-120 beats/min) was observed in all patients. Hypotension (90-100 mm Hg)/tachypnea (24-28 beats/min) was observed in 31 patients. Clinical and endoscopic data provided the basis to determine diagnostic and therapeutic tactics.

We used the Cook classification to identify latent, explicit, and hemodynamically significant gastroduodenal bleeding (Figure 1). As shown in Figure 1, 29 patients (35.4%) had hidden clinical signs of gastroduodenal bleeding. These patients underwent fecal analysis for latent blood. In 17 patients, the Gregersen benzidine reaction was positive, with Weber-Van Deen reaction positive in 12 patients. As shown in Figure 1, the main group had less obvious and hemodynamically significant indicators than the control group.

The severity of blood loss that developed against the background of gastroduodenal bleeding was assessed based on a comparison of laboratory, clinical, and instrumental data in accordance with the Gorbashko classification method. In the main group, 20 patients (47.6%) had mild bleeding, 13 patients (30.1%) had moderate bleeding, and 9 patients (21.4%) had severe blood loss (Figure 2). In the control group, 11 patients (30.6%) had mild bleeding, 13 patients (34.2%) had moderate bleeding, and 14 patients (36.8%) had severe blood loss (Figure 2). In the main group, there were less patients with moderate and severe bleeding versus that shown in the control group. Thus, bleeding from acute stressful gastroduodenal ulcers worsened in 1 of 3 cases.

According to time of occurrence of acute stressful gastroduodenal ulcers and rate of bleeding complications, most patients were diagnosed in both groups in the first weeks after injury (Figure 3). Bleeding in most patients with severe burns (48.7%) developed as a result of alternative exposure to additional risk factors, ie, when the burn area was 40% and deep burns exceeded 20%.

In our patients, gastroduodenal hemorrhage was observed in 54.3% of patients with a total burn area of more than 40% and in 81.4% of patients with a total burn area of 20% in the first 2 weeks after burn injury. In addition, gastroduodenal bleeding in the main group decreased significantly in the following weeks compared with that shown in the control group, that is, by 5% and 2.6% at week 2 and week 3, respectively. Thus, these patients required preventive and antiulcer treatment from the moment of admission to the hospital.

To determine the effect of Helicobacter pylori on the formation of acute gastroduodenal ulcers, we examined 51 patients for helicobacteriosis. Of these, 28 patients were from the main group and 23 patients were from the control group. To determine the presence of contamination of the mucosa of the gastroduodenal zone of Helicobacter pylori, we used a noninvasive diagnostic method (serological method based on the determination of cytotoxin-associated gene A protein [CagA] antibodies to Helicobacter pylori in blood serum). To detect antibodies to H. pylori, an enzyme-linked immunoassay method was used. Blood sampling was performed on the day of admission of patients, before start of antibacterial therapy. Blood was obtained from the central vein. It is known that H. pylori is activated in patients with chronic gastropathy. Therefore, patients with a history of chronic gastropathy were also compared (Table 1).

Variations of _{H. pylori} infection were not significantly activated on the first day of burns. It was noted that, in patients with a tendency for gastroduodenal complications, the indicators were higher at a ratio of 1.5 to 2.3 times after 7 days and at a ratio of 3.5 to 2.6 times after 14 days among the main and control groups. In patients without a tendency for gastroduodenal complications, the titer of H. pylori-CagA antibodies increased by a ratio of 1.2 to 1.6 times for 7 days compared with healthy individuals (Table 1). Thus, the titer of H. pylori-CagA antibodies in patients with a tendency for gastroduodenal complications was characterized by a gradual increase in the acute period of burn disease. This indicates an increase in the density of H. pylori in the gastric mucosa.

We found that development of vascular endothelial dysfunction depended on the severity and duration of the burn periods. In patients in the control group, VEGF concentrations on day 3 were 10 times higher in patients with a tendency for gastroduodenal complications than in healthy individuals and 6.7 times higher than in patients without a tendency for gastroduodenal complications. On day 7, these indicators in patients with a tendency for gastroduodenal complications were 19 times higher than in healthy individuals and 9.5 times higher patients without a tendency for gastroduodenal complications versus that shown in healthy individuals. On day 14, VEGF concentration was 22.5 times in patients with a tendency for gastroduodenal complications and 14.8 times in patients without a tendency for gastroduodenal complications compared with that shown in healthy individuals.

In the main group of patients, VEGF concentrations remained virtually unchanged for 3 days compared with that shown in the control group of patients with a predisposition for gastroduodenal complications but decreased to 18.5% on day 7 and to 24.2% on day 14. In patients without a tendency for gastroduodenal complications, there was a decrease of 10.9% on day 7 and 15.4% on day 14 compared with patients in the control group. Thus, the use of L-arginine-containing drug in the main group of patients on all sampling days resulted in a decrease of VEGF concentration compared with that shown in the control group (Table 2).

Various methods of therapeutic endoscopy for gastroduodenal bleeding in severely burned patients were used in 52 patients (66.7%). Argon plasma coagulation was the predominant method of hemostasis (Figure 4) and was used in 34 patients (65.4%), with the infiltration method only used in 9 patients (17.3%) and radiowave coagulation in 9 patients (17.3%).

As shown in Table 3, endoscopic manipulations aimed at stopping the ongoing ulcerative bleeding were performed in 31 patients, with hemostasis achieved in 29 patients. Accordingly, the effectiveness of endoscopic hemostasis methods as a means of stopping ulcerative bleeding was 93.5%.

In patients with Forrest Ia bleeding, bleeding was stopped in 84.6% (11 of 13 patients). In patients with Forrest Ib bleeding, hemostasis was achieved in 88.2% (15 of 17 patients). In 1 patient with Forrest Ia bleeding, endoscopic hemostasis was not performed (due to technical difficulties). When endoscopic hemostasis is ineffective, patients could receive urgent surgery. When we compared the effectiveness of techniques of endoscopic hemostasis on continued bleeding, argon plasma coagulation was unsuccessful in 1 patient (with Forrest Ia). However, during infiltration of the periulcerous zone, bleeding was not stopped in 1 patient (with Forrest Ib). With both argon plasma and radiowave coagulation, no patients with Forrest Ib bleeding had treatment failure.

Preventive endoscopic hemostasis, aimed at ensuring stable hemostasis and reducing the likelihood of its recurrence in the Forrest II picture, was performed in 29 patients (58% of all patients with this endoscopic picture). Endoscopic hemostasis was not performed in 21 patients (42%), as signs of stable hemostasis were noted and the risk of possible recurrence was assessed as low (Figure 5).

In patients in the main group (n = 42), after 2 or 3 days of antisecretory treatment, pain in the epigastric region significantly decreased, with feeling of discomfort in the stomach area, heartburn, and belching almost disappearing, and there was no pronounced gastrointestinal paresis. Healing of erosions with a diameter of up to 1 mm was observed on average at 8 to 10 days. Obvious bleeding from acute ulcers was detected in 9 of 42 patients (21.4%), which stopped after endoscopic therapy. One patient had a relapse on day 2 after injury.

In patients in the control group (n = 38), pain syndrome, discomfort in the stomach, and belching persisted for 7 to 10 days; some patients had vomiting with food eaten. Gastrointestinal bleeding was detected in 8 of 38 patients who were stopped endoscopically; however, 3 patients had relapse.

During the study period, 6 patients received surgery for bleeding from acute gastroduodenal ulcers after severe burns, with 3 of these patients (3.8%) needing surgery as a result of continued bleeding and 3 (3.8%) needing surgery as a result of recurrent bleeding. The following types of operations were performed: gastric resection (n = 1), gastrotomy with stitching of a bleeding ulcer (n = 3), and stitching of an ulcer with pyloroplasty and stem vagotomy (n = 2). Two patients (33.3%) died after surgical interventions. For patients who received organ-preserving operations for bleeding from symptomatic gastroduodenal ulcers, the mortality rate was 16.7% (1 of 6 patients died) (P > .05). The operations of choice for continued bleeding from symptomatic gastroduodenal ulcers are organ-preserving surgery, gastrotomy with stitching of an ulcer with a bleeding vessel, and stitching of an ulcer with pyloroplasty and vagotomy.

The total mortality rate was 13.7% (11 of 80 patients died). Acute blood loss and posthemorrhagic anemia were the direct causes of death in 3 patients (27.3%) with bleeding from symptomatic gastroduodenal ulcers. For the other 8 patients (72.7%), the cause of death was a different pathology. These 8 patients died from severe concomitant pathology with healed ulcers or ulcers in the scarring stage.

Thus, the use of modern endoscopic methods to stop bleeding (argon plasma coagulation) and modern antisecretory drugs allows for the successful conservative treatment of patients with bleeding of acute gastroduodenal ulcers, minimizing the likelihood of an endoscopically irreversible recurrence of bleeding (the frequency of recurrence of bleeding was reduced from 15% to 11.7%).

DISCUSSION
In addition to vomiting in the form of coffee grounds, abdominal pain, bleeding in the form of melena, tachycardia, tachypnea, patients with burns and gastroduodenal bleeding can present with agitation, constipation, and hunger. In patients with severe burns, bleeding from stressful gastroduodenal ulcers is characterized by a high tendency for hypotension against the background of burn shock and hypovolemic condition. Therefore, it is necessary to pay attention to both anamnesis and clinical signs.

In our patient group, latent bleeding was observed in 36.2% of patients. We monitored the possibility of gastroduodenal bleeding with the Gregersen benzidine reaction and the Weber-Van Deen reaction for timely detection of latent gastroduodenal bleeding in patients with severe burns. Among patients with severe burns, gastroduodenal bleeding was severe in about one-third of patients. Most of the diagnosed stressful phenomena developed in the first weeks with momentary trauma.

The risk of erosive and ulcerative lesions of gastroduodenal zones, including those complicated by gastroduodenal bleeding, increases with an increase in the total area of the burn lesion, especially with an area of more than 40%. In this regard, it is necessary to perform esophagogastroduodenofibroscopy in patients who have more than 40% burn area upon admission to the hospital; this allows the timely detection of erosive and ulcerative changes from the gastroduodenal zones so that treatment and prevention of gastroduodenal bleeding can be started.

In patients with a tendency for gastroduodenal complications, compared with patients without a tendency for gastroduodenal complications, an increase in H. pylori infection was shown in the first days of the burn toxemia period, due to the severity of the burn and a tendency for microcirculation disorders in the gastrointestinal mucosa and inflammation in the epithelium. The increased density of H. pylori in the gastric mucosa has an aggressive effect, further aggravating the inflammatory activity of the mucous membrane in the gastroduodenal region. It also persists for a long time against the background of persistent inflammation. As a result, acute inflammation occurs in the gastroduodenal region and causes bleeding.

The development of vascular endothelial dysfunction can lead to vascular damage. We observed high concentrations of VEGF during the acute period of burns. High concentrations of VEGF can lead to the release of compensatory vasoactive substances that cause pathogenic hypoxic changes in the vascular endothelium. We found that VEGF concentrations were consistently higher in patients with a tendency for gastroduodenal complications than in patients without a tendency for gastroduodenal complications. This may be associated with the proliferation of endothelial cells, contributing to the proliferation of blood vessels and the formation of new vessels against the background of chronic diseases of the gastroduodenal zone.

The observed relationship between the amount of VEGF in the blood serum and the severity of gastroduodenal bleeding can be explained by the activation of cytokine production under conditions of tissue hypoxia, interpreted as a mechanism aimed at restoring the microcirculatory bed, contributing to the improvement of reparative processes in the mucous membrane and accelerating the epithelialization of ulcerative defects in patients. When there is an excessive increase in VEGF, especially against the background of atrophy, inflammation of the mucous membrane and H. pylori infection can accelerate the processes of neoangiogenesis. Thus, in patients with severe burns and gastroduodenal bleeding, the observed increase in the amount of VEGF depends on the severity of bleeding and infection with H. pylori; thus VEGF is diagnostically valuable.

The criterion for the effectiveness of therapeutic effect produced with esophagogastroduodenofibroscopy is the achievement of final hemostasis, that is, the absence of recurrence of hemorrhage. Eight of 52 patients with severe burns had recurrence of acute gastroduodenal ulcerative bleeding (15.4% of the number of patients with endoscopic hemostasis).

With argon plasma coagulation, bleeding recurrence was shown in 13.7% of cases. However, recurrence occurred after injection hemostasis in 24% of cases and after radiowave hemostasis in 23% of cases (P > .05). We found that argon plasma coagulation is a more effective method of hemostasis for bleeding from acute stress gastroduodenal ulcers after severe burn injuries than injection and radiowave hemostasis.

The criterion for the effectiveness of the therapeutic effect of antisecretory drugs in acute gastroduodenal ulcerative bleeding was the absence of recurrence of hemorrhage and the presence of stable hemostasis. In the main group of patients, against the background of antisecretory therapy, bleeding recurrence occurred in 5 patients (11.9%); in the control group, 8 patients (22.2%) had bleeding recurrence.

Stable hemostasis after bleeding from acute gastroduodenal ulcers in patients with severe burns was achieved in 77.8% of the control group against the background of antisecretory therapy with H₂-receptor blockers and in 88.1% of the main group against the background of antisecretory treatments and with the addition of dopamine D₂-receptor blockers (domperidone) and L-arginine. The dopamine D2-receptor blocker (domperidone) has a protective effect on the mucous membranes of the gastroduodenal zones and a positive regulatory effect on the tone and contractile activity of the gastrointestinal tract. Thus, the motor-evacuation functions of the gastroduodenal zones are quickly restored.

Recent studies conducted in compliance with the principles of evidence-based medicine have indicated the possibility of effective and safe use of L-arginine in clinical practice for various pathologies, including diseases of the gastrointestinal tract.^11,12 The use of antisecretory drugs with the addition of dopamine D2-receptor blockers (domperidone) and L-arginine in terms of ensuring stable hemostasis showed that it is the most effective in treatment of bleeding from acute ulcers in patients with severe burns (P > .05).

CONCLUSIONS
In patients with severe burns and acute gastroduodenal ulcers complicated by bleeding, severe organ dysfunction and a high tendency for severity of the patient’s condition (27.5%) can occur. In our study, multiorgan failure resulted in death in 72.7% of patients with bleeding from gastroduodenal ulcers.

The clinical manifestations of gastroduodenal bleeding in patients with severe burns patients depend on the condition and duration of the pereulcerogenic area, the nature of erosions, the depth and size of the ulcer, and the course of epithelialization and functional changes (deformities and motility) in the gastroduodenal area.

The risk of developing gastroduodenal erosive and ulcerative complications, including gastroduodenal bleeding, increases with an increase in the total area of the burn lesion, especially with an area of more than 40%. For these patients, it is necessary to perform esophagogastroduodenofibroscopy on hospital admission, which can allow the timely detection of erosive and ulcerative lesions from the gastrointestinal tract and the ability to begin treatment and prevention of gastroduodenal bleeding.

In patients with gastroduodenal bleeding after severe burn injuries, the observed increase in VEGF depended on the severity of bleeding and infection with H. pylori and was diagnostically valuable.

If occurrence of acute gastroduodenal erosions and ulcers in patients with severe burns is prevented, the risk of their formation decreases, as well as the likelihood of bleeding.

Argon plasma coagulation is the most effective method of hemostasis for bleeding from gastroduodenal ulcers, compared with injection and radiowave methods.

Stable hemostasis during bleeding from acute gastroduodenal ulcers in patients with severe burns was achieved in 77.8% of the control group against the background of antisecretory therapy with H₂-receptor blockers and in 88.1% of the main group against the background of antisecretory treatments and with the addition of dopamine D₂-receptor blockers (domperidone) and L-arginine.

The operation of choice for bleeding from stressful gastroduodenal ulcers in patients with severe burns is organ-preserving interventions.

REFERENCES

1. Wagner DO, Verbitsky VG. Comprehensive prevention of gastrointestinal bleeding in patients with shockogenic thermal trauma. Materials of the All-Russian Scientific and Practical Conference With International Participation, Dedicated to 70th Anniversary of the First Burn Center of Russia. St. Petersburg, Russia; 2016:19-22.
   CrossRef - PubMed
   
2. Chichkov OV, Skrebin ON, Gorshenin TL, et al. [Features of the clinic, diagnosis and treatment of patients with bleeding gastroduodenal ulcer, victims of burns.] J Basic Res Penza. 2014;10-11:208-211.
   CrossRef - PubMed
   
3. Fayazov A, Akhmedov A, Mirzakulov A, Nabiev A. Prediction of development of acute gastroduodenal complications in patients with severe burns. Burn Care Prev. 2021;1(3):99-106
   CrossRef - PubMed
   
4. Fayazov AD, Akhmedov AI, Babajanov AS, et al. [Improvement of the treatment of severely burned against the background of chronic gastroduodenal ulcers.] Bull Sci Educ. 2021;4(107):18-26.
   CrossRef - PubMed
   
5. Romanenkov NS. [Possibilities of modern medical technologies in the prevention and elimination of complications of gastric ulcer and duodenal ulcer in victims of burns.] Abstract of Thesis of Candidate of Medical Sciences. Velikiy Novgorod; 2014.
   CrossRef - PubMed
   
6. Akhmedov AI, Fayazov AD, Babajanov AS, et al. The possibility of predicting the development of acute gastroduodenal complications in severely burned patients. XXII International Correspondence Scientific Specialized Conference. International Scientific Review of the Problems of Natural Sciences and Medicine, Boston, MA; 2021. https://scientific--conference-com.translate.goog/grafik/grafik-2019-pervoe-polugodie.html?_x_tr_sl=ru&_x_tr_tl=en&_x_tr_hl=en&_x_tr_pto=ajax,sc,elem
   CrossRef - PubMed
   
7. Shibuya M. Vascular endothelial growth factor and its receptor system: physiological functions in angiogenesis and pathological roles in various diseases. J Biochem. 2013;153(1):13-19. doi:10.1093/jb/mvs136
   CrossRef - PubMed
   
8. Yenikomshian H, Reiss M, Nabavian R, Garner WL. Gastric feedings effectively prophylax against upper gastrointestinal hemorrhage in burn patients. J Burn Care Res. 2011;32(2):263-268. doi:10.1097/BCR.0b013e31820aafe7
   CrossRef - PubMed
   
9. Cook DJ, Fuller HD, Guyatt GH, et al. Risk factors for gastrointestinal bleeding in critically ill patients. Canadian Critical Care Trials Group. N Engl J Med. 1994;330(6):377-381. doi:10.1056/NEJM199402103300601
   CrossRef - PubMed
   
10. Huang JQ, Sridhar S, Hunt RH. Role of Helicobacter pylori infection and non-steroidal anti-inflammatory drugs in peptic-ulcer disease: a meta-analysis. Lancet. 2002;359(9300):14-22. doi:10.1016/S0140-6736(02)07273-2
    CrossRef - PubMed
    
11. Lau JY, Sung JJ, Lee KK, et al. Effect of intravenous omeprazole on recurrent bleeding after endoscopic treatment of bleeding peptic ulcers. N Engl J Med. 2000;343(5):310-316. doi:10.1056/NEJM200008033430501
    CrossRef - PubMed
    
12. Serrano P, Lanas A, Arroyo MT, Ferreira IJ. Risk of upper gastrointestinal bleeding in patients taking low-dose aspirin for the prevention of cardiovascular diseases. Aliment Pharmacol Ther. 2002;16(11):1945-1953. doi:10.1046/j.1365-2036.2002.01355.x
    CrossRef - PubMed