Federal Scientific Clinical Center of Miners’ Health Protection, 

Leninsk-Kuznetsky, Russia

Gastroduodenal bleeding is one of the most common complications of gastric and duodenal ulcer disease. Choric ulcer in upper digestive tract is complicated by bleeding in 15 % of patients [16]. Despite of apparent success in treating such patients, the general mortality from ulcerous bleeding is 13-15 %, the postsurgical one ‒ 12-21 % [6, 9, 19, 20]. The modern curative techniques (endoscopic hemostasis, drug therapy, blood loss correction) allow achieving primary hemostasis using conservative techniques in 90-95 % of patients. However recurrent bleeding develops in 10-35 % of observed cases. Surgical interventions for recurrent bleeding are performed in extremely unfavorable situations at the background of disordered compensatory mechanisms of the body that result in high mortality (35-75 %) [6, 10, 15]. According to many authors, unsatisfactory results of surgical treatment are conditioned by significant amount of patients with recurrent bleeding.

Recurrent bleeding evolves in each seven patient during 24-36 hours, and recurrent bleeding of severe degree ‒ approximately in 50 %. At this time, surgical interventions are associated with higher mortality [1]. However widening the indications for emergent surgery is not a solution for such complex problem, because emergent operations result in two time higher mortality in case of hemorrhagic shock [4]. A possibility of reliable prediction of recurrent bleeding plays a crucial role in solving the problem. The literature includes the techniques for prediction of possible recurrent bleeding: with HLV antigens [14], with combination of clinical endoscopic signs of risk factors [18], with computer software data [12]. However there is an actual issue regarding prediction of recurrent bleeding, especially considering false positive results in 10-18 % [2, 5, 21, 22].

Research objective ‒ to assess endoscopic monitoring influence on clinical laboratory picture of recurrent gastroduodenal ulcer bleeding.

MATERIALS AND METHODS

The present study is based on the previous analysis including 205 patients with acute gastroduodenal ulcerous bleeding from upper digestive tract who received treatment in the surgery department, Clinical Center of Miners’ Health Protection. The excluded subjects were the patients with combination of ulcerous bleeding and perforated hollow organs.

The mean age of the male patient was 43.2 ± 16.7. The mean age of the female patients was 44.8 ± 15.4. Mainly, the working age patients (age of 21-50, 56 %) were presented.

Two groups of the patients were defined according to endoscopic monitoring influence on clinical diagnostic presentation of recurrent gastroduodenal ulcer bleeding (Table 1). The main group included the patients who received endoscopic monitoring using different techniques of endoscopic hemostasis. The entity of monitoring is preventing recurrent bleeding. The comparison group included the patients with primary endoscopic hemostasis and recurrent examination in 8-12 hours regardless of bleeding activity. The choice of recurrent (preventive) hemostasis was defined with defect location, vessel location in defect and hemostasis stability.

Table 1
Distribution of patients into groups

The estimation of bleeding severity on admission was performed with the recommendations by A.I. Gorbashko (1998) considering Hb, Ht, HR and systolic arterial pressure (table 2).

Table 2
Bleeding severity at the moment of admission according to A.I. Gorbashko (1998)

As for admission time, the main group included more patients of severe and middle degrees (not statistically significant).

During analysis of primary bleeding source one could see that gastric ulcers (acute and chronic) take the second place in regard to frequency, and they are equally presented in both groups (table 3). Chronic duodenal ulcers take the leading place in both groups. There were no statistically significant differences in localization of the first source of bleeding in the groups.

Table 3
Distribution of patients depending on location of primary source of bleeding

All patients had clinical manifestations of gastrointestinal diseases during several years (table 4).

Table 4
Clinical characteristics of patients with peptic ulcer disease

Note: the number of patients differs from the real one bacause of combination of different complaints in one patient.

Regardless of bleeding activity, all patients received systemic hemostatic, anti-secretory therapy and transfusion of red packed cells in relation to blood loss degree. There were no differences in characteristics and quality of drug therapy between the groups.

EXAMINATION METHODS

The general clinical methods were used for all patients: examination, collection of anamnesis data, laboratory study of the blood with estimating red blood cells (RBC), hemoglobin (Hb) and hematocrit (Ht) on admission and 24 hours after recurrent bleeding. Abdominal X-ray examination was conducted for identification of hollow organ perforations.

Fibrogastroduodenoscopy was the leading examination technique. Besides the fact of bleeding, the visual analysis included localization and characteristics of bleeding defect, its size, shape, characteristics of the bottom and edges. Bleeding activity was estimated according to J. Forrest.

The study was based on endoscopic monitoring (EM). EM is dynamic endoscopic control of bleeding source during definite time intervals depending on bleeding activity. Monitoring time intervals were 2-8 hours. Such time intervals are associated with retrospective analysis of recurrent bleeding in the patient who received their treatment for acute gastroduodenal ulcer bleeding in our clinic during 2002-2012 [13]. Agonal state of a patient is the single contraindication for emergent endoscopic intervention.

For the patients with suspected acute gastroduodenal ulcer bleeding the bleeding source endoscopic diagnostics and primary endoscopic hemostasis were conducted on admission in the room of emergent examinations. Both monotherapy and combined hemostasis were used. The statistical analysis was performed with Statistica 6.1 (Statsoft Inc., USA). Student’s test was used for estimating the statistical significance in mean values. Mann-Whitney and chi-square non-parametric tests were used for cases with absent normal distribution in the samples. The critical level of significance was 0.05.

RESULTS

Bleeding stopped in most patients (Forrest-II) on admission: 67 % in the main group, 67.6 % in the comparison group. Active bleeding rates were 24 % and 24.8 % correspondingly (Fig. 1).

Figure 1

Bleeding activity on admission (%)

The retrospective analysis for 105 patients (comparison group) with acute gastroduodenal ulcerous bleeding from upper digestive tract found 16 (15.2 %) cases of recurrent bleeding. F-IA bleedings were associated with 5 (45.5 %) cases of recurrent bleeding, F-IB ‒ with 3 (20 %), F-IIA ‒ with 4 (22.2 %), F-IIB ‒ with 2 (6.9 %), F-IIC ‒ with 2 (8.3 %).

All recurrent cases were accompanied by the typical clinical picture: collapse, hypotony, tachycardia, haematemesis, hemafecia or melena. During retrospective investigation of mean time of recurrent bleeding development we concluded that time of recurrence is directly related to primary degree of its activity on admission (according to J. Forrest).

On the basis of the above mentioned data we offered to develop endoscopic monitoring technique depending on its primary activity. Such monitoring is based on preventing recurrent bleeding or on identification at early (preclinical) stages.

EM was conducted for 100 patients (main group) with acute gastroduodenal ulcerous bleeding from upper digestive tract. There were 36 cases of recurrent bleeding. These figures are significantly higher compared to the literature data. Forrest IA recurrence was noted in 72.7 %, Forrest IIA ‒ in 73.7 %. For other groups the recurrence rate is not above 30 %. There were no recurrent bleeding cases in the group of Forrest III bleeding activity. For this group the time interval was 12-24 hours: this is the time of planned revision next day after admission (Fig. 2). The analysis of this diagram shows the clear statistical reliability for prevailing recurrent bleeding in the main group.

Figure 2

Recurrent bleeding rate in the examined groups

Considering the higher rate of recurrence in the main group compared to the comparison group and the literature data, we analyzed bleeding recurrence rate and development of typical clinic presentation of recurrent bleeding (Fig. 3).

Figure 3

The comparative frequency of clinical and endoscopic recurrent bleeding

 

In the comparison group all recurrent bleeding cases were diagnosed with clinic signs (collapse, hypotony, haematemesis, hemafecia and melena) and confirmed with endoscopy. In the main group the typical clinical presentation of recurrent bleeding was diagnosed in 3 (8.4 %) patients only (p < 0.05), including 1 (2.8 %) case with F-IA bleeding activity and 2 cases with F-IIA (5.6 %).

As for time of recurrent bleeding in the main group, all clinically significant cases were noted during 6 hours after admission (Fig. 4). In the control group 12 (75 %) recurrent cases were noted during the first hours after admission, and 3 (18.6 %) patients had recurrent bleeding during 6-12 hours. It may indicate the necessity for preventive hemostasis during repetitive examinations.

Figure 4

The number of clinically diagnosed recurrent bleedings in terms of origin time (%)

 

After receiving such data we performed the retrospective analysis of bleeding ulcers and their influence on clinical and laboratory presentation.

The significant worsening patient’s state was observed during investigation of the clinical presentation on hospital admission and for the moment of recurrent bleeding (table 5). Also changes were found regarding the characteristics of complaints, which clearly testify recurrent bleeding and worsening patient’s state. It is especially concerning the comparison group.

Table 5
Influence of recurrent bleeding on clinical presentation of disease course

Note: a possibility of several simultaneous leading complaints in one patient.

The diagrams of the levels of RBC, Hb and Ht in both groups show the changes in red blood values on admission, at the moment of recurrent bleeding and 24 hours after recurrent bleeding. There were no statistically significant differences between the groups according to the first two diagrams of red blood (Fig. 5, 6).

Figure 5

Laboratory values of “red” blood in patients with gastrointestinal bleeding on admission

Figure 6

Laboratory values of “red” blood in patients with gastrointestinal bleeding for the moment of recurrent bleeding

But after the analysis of red blood values after 24 hours, we received the following results: the clear statistically significant differences in RBC (p = 0.004), Hb (p = 0.041) and Ht (p = 0.046) with worsening values in the comparison group (Fig. 7).

 Figure 7

Laboratory values of “red” blood in patients with gastrointestinal bleeding 24 hours after recurrent bleeding

 

CONCLUSION

1. Ulcerous bleeding endoscopic monitoring allows identifying recurrent bleeding at early preclinical stages.

2. Laboratory values of red blood are not informative for a surgeon compared to endoscopic monitoring data; they become available only 24 hours later.

3. Endoscopic monitoring allows controlling patient’s state and providing adequate and timely treatment.