Gastrointestinal (GI) bleeding is a common indication for endoscopy. For refractory cases, hemostatic powders (HP) represent “touch-free” agents.

To analyze short term (ST-within 72 h-) and long-term (LT-within 30 d-) success for achieving hemostasis with HP and to directly compare the two agents Hemospray (HS) and Endoclot (EC).

HP was applied in 154 consecutive patients (mean age 67 years) with GI bleeding. Patients were followed up for 1 mo (mean follow-up: 3.2 mo).

Majority of applications were in upper GI tract (89%) with following bleeding sources: peptic ulcer disease (35%), esophageal varices (7%), tumor bleeding (11.7%), reflux esophagitis (8.7%), diffuse bleeding and erosions (15.3%). Overall ST success was achieved in 125 patients (81%) and LT success in 81 patients (67%). Re-bleeding occurred in 27% of all patients. In 72 patients (47%), HP was applied as a salvage hemostatic therapy, here ST and LT success were 81% and 64%, with re-bleeding in 32%. As a primary hemostatic therapy, ST and LT success were 82% and 69%, with re-bleeding occurring in 22%. HS was more frequently applied for upper GI bleeding (P = 0.04)

Both HP allow for effective hemostasis with no differences in ST, LT success and re-bleeding.

Gastrointestinal (GI) bleeding represents a major challenge for the GI endoscopist both in term of frequency, which is estimated to be 150/100000 for upper and 33/100000 for lower GI bleeding with a mortality ranging between 2 and 10%[1-3], and in terms of technical efforts to reach a stable hemostasis. As the administration of direct oral anticoagulants[4] and the use of assistant devices in terminal cardiomyopathy[5] is increasing, sufficient and effective treatment of GI bleeding is mandatory while at the same time can be clinically challenging. In the last years, endoscopists increasingly face emergency bleeding in a clinical scenario in which coagulation parameters cannot always be corrected to normal range. Further, with increasing development of advanced endoscopic therapeutic procedures, iatrogenic bleeding after endoscopic resections represents another emerging problem[6]. Conventional treatment approaches achieve hemostasis in more than 90% of cases[7], however, depending on the bleeding site and source can be technically challenging, and might not be optimal for diffuse oozing bleeding as frequently observed in patients with impaired coagulation or cancer bleeding.

Hemostatic powders (HP) act as “touch-free” agents that can be easily administrated for the treatment of GI bleeding, which are generally safe and well tolerated[8-11]. Hemospray (HS, TC-325, Cook Medical, Bloomington, Indiana, United States) is an inert mineral based compound, which, in contact with blood, absorbs water and acts cohesively and adhesively, thereby forming a covering mechanical tamponade. By fluid absorption, HS enhances clot formation by deforming and packing erythrocytes, concentrates activated platelets with clotting factors and interacts with the fibrin matrix[12] and within 24 to 72 h, the adherent coat sloughs off into the GI lumen[10]. With his local hemostatic proprieties, first studies suggest that HS is equally effective in both patients with and without systemic antithrombotic therapy[9].

Endoclot (EC, Micro-Tech Europe, Düsseldorf, Germany) is a starch-derived agent composed of absorbable hemostatic polysaccharides. Similar to HS, in contact with blood, EC initiates a dehydration process leading to a concentration of clotting factors, platelet and erythrocytes thereby accelerating the physiological clotting cascade and the formation of a mechanical shell of gelled matrix which adheres to the bleeding tissue[13]. Although data on the efficacy of EC are still limited, first clinical evidences suggest that both HS and EC allow for effective bleeding control[8,11,14-20]. Further, no direct comparison of the efficacy of these two HP is available to date. Against this background we set off: (1) To analyze short and long term hemostatic effectiveness of HP; and (2) to compare the efficacy between the agents HS and EC in achieving hemostasis in a large cohort of patients treated for emergency GI bleeding in our center.

Herein we report on our experience in the treatment of GI bleeding both with HS and EC in a single tertiary university care center. To the best of our knowledge our work is the first to directly compare two different HP for the treatment of GI bleeding in the upper and lower GI tract. Our study confirms the findings of other investigators where an excellent immediate control of the bleeding source was achieved with HP[8,11,18,19,21]. HP exhibited an overall short-term success of 82% in our study. With this, ST success was higher in our cohort compared to a previous report on a smaller cohort by Chen and colleagues[8], although this study analyzed of success rates of HS only.

According to the literature, hemostatic success of HP within 7 to 8 d range between 51% and 87.5%[14,15,19]. Within this study, we performed FU for at least one month in the vast majority of patients (87%) and long-term success dropped to 67% in our study. Hence, these data are consistent with results from GRAPHE registry in which LT success rates of 66% were reported[19]. A graphic illustration (Figure 3) of the mean incidence of re-bleeding after application of HP according our and past studies[8-11,14-16,18,19,22] shows that RBR increase over time after HP application across studies and with this, although allowing for excellent immediate bleeding control, HP appears to be not suitable as a definitive long-term hemostasis tool in patients with a high-risk profile of bleeding recurrence. On the other hand, the benefit of HP is the high immediate hemostasis rate and that can be administered more than once without risk of “overdosing” or induction of bleeding due to mechanical irritation.

Open in New Tab   Full Size Figure   Download Figure

Figure 3 Incidence (%) of re-bleeding after application of hemostatic powder according our data and past studies[8-11,14-16,18,19,21,26].

When performing subgroup analyses according to bleeding etiology, overall ST and LT success in peptic ulcer disease was 81% and 68% with a RBR of 19%. With this, our results are consistent to those reported in the literature[11,13,14,17,18,23], with immediate hemostasis ranging between 78 and 96% and RBR between 10.5 and 38%. However, when analyzing peptic ulcer disease with Forrest Ia bleeding in our study, ST and LT success were only 67% and 33% respectively. Together with results from other studies that have reported a re-bleeding risk of Forrest Ia lesions under HP between 67% and 73%[10,11,13,14,17,18,24], our data show that HP are not effective as a first-line therapy in Forrest Ia peptic ulcer bleeding. Nevertheless, HP but might still be useful in this scenario as a bridging or rescue strategy until an alternative therapy as another endoscopic procedure, a radiological embolization or surgical therapy can be performed.

HP have also been reported to be effective as rescue therapy for variceal bleeding when band ligation fails[25] and also in gastric varices and gastric bleeding derived from portal hypertension[23]. Within our study, we observed an overall ST success of 85% and LT success of 56%. It is important to note that in the majority of applications for variceal bleeding, the bleeding was serious with 70% of patients presenting with hemorrhagic shock. Against this background, the overall ST success can be regarded as high, and thus HP might represent a promising addition to arsenal of the endoscopist for severe and refractory variceal bleeding.

Due to their touch-free application and large coverage, HP are also well suited for the treatment of tumor bleeding. As shown in our study, HP provide immediate hemostatic efficacy of 95%, a short-term success of 83% and a long-term success 87% in patients with diffuse tumor bleeding. With this, our results are comparable to previous studies, in which immediate efficacy of HP and RBR ranged between 93%-100% and 20%-32%, respectively[10,14,18,21,26]. Since tumor bleeding is frequently diffuse and exhibits a large bleeding area, high RBR ranging up to 49% have been reported with conventional hemostatic approaches[24,27]. Together, with data from the largest multicenter retrospective study, in which an immediate hemostasis of HP for tumor bleeding was achieved in almost 98% of patients, our data show that HP are allow for effective control of tumor bleeding.

To date, no direct comparison between HS and EC is available. When looking across studies rates for achieving primary hemostasis in the upper GI tract with EC and HS have been reported to range between 82%-100%[13,16] and 85%-98%[8,9,11,14,15,18,19,21], respectively. Our study is the first to directly compare the efficacy of HS and EC and no significant different in their hemostatic efficacy and RBR were observed between these two agents. Nevertheless some technical differences between the two HP should be noted: first HS is sprayed at high pressure with a propellant CO₂ cartridge. Such feature might be an advantage in cases of high pressure bleeding or scenarios where a large surface needs to be covered. On the other hand, high-pressure application can potentially cause further tissue injury to the point of perforation especially in friable or inflamed mucosa. Indeed, in two of the patients treated with HS (1.3%), perforation occurred as major adverse events after application of HS in the current study. Occurrence of intestinal perforation after HS application have been reported in other series as well[15,18], therefore some caution of using HS might be necessary. In contrast, with EC the pressure of spraying is much lower, allowing a more sectorial area of targeting, making EC more suitable for localized bleeding lesions like a peptic ulcers or a surface after resection. On the other hand, the area that can be covered with EC might be lower with EC as compared to HS and also high pressure bleeding might be less controlled. However, more systematic studies are clearly needed to investigate on these aspects.

For lower GI bleeding ST and LT success of HP were 75% and 56.3% with a RBR of 37.5%. Data on the role of HP for lower GI bleeding are relatively scarce to date and long-term FU data are completely lacking. In the largest series of low GI bleeding treated with EC, hemostasis was achieved in 83% of the cases with a RBR of 11%[16]. Although limited by the number of patients included in the study, our results do support the concept that HP represent valuable therapeutic options for lower GI bleeding when conventional hemostatic approaches fail.

Limitations of the current study also need to be addressed. Although our study included a large number of patients, its setting in a single high volume university centre might have led to a certain bias in terms of patients characteristics. As shown by the clinical data, a large percentage exhibited a variety of severe co-morbidities and therefore most likely do not represent an average cohort. Further, we did not utilize a randomized study protocol and the decision to apply HS or EC was at the discretion of the endoscopist and therefore subjective.

In conclusion, our study demonstrates that both HPs HS and EC allow for bleeding control with high short-term efficacy when used as primary or salvage therapy. Further, both EC and HS exhibit high efficacy for achieving hemostasis in impaired coagulation status or friable tissues. With these properties, HPs represent powerful and effective additions to the armentarium of the endoscopist for treatment of GI bleeding.