Advertisement

A Comparative Study of the Resorption and Immune Response for Two Starch-Based Hemostat Powders

Open AccessPublished:October 31, 2022DOI:https://doi.org/10.1016/j.jss.2022.09.022

      Abstract

      Introduction

      Powder hemostats are valuable adjuncts to minimize intraoperative and postoperative complications. In addition to promotion of rapid coagulation, resorption, and biocompatibility are desirable attributes. Plant starch-based polysaccharide hemostat powders are effective and widely used hemostatic agents, however their source and/or processing can affect characteristics such as in vivo degradability. For example, Arista is a purified/hydrolyzed starch powder that is rapidly resorbed in vivo; whereas PerClot shows slow resorption and preservation of a crystalline form.

      Materials and methods

      In the present study, we compared the cellular response to the hemostatic agents PerClot and Arista both in vitro and in vivo, and used potato starch and urinary bladder extracellular matrix (UBM-ECM) as high crystallinity/slowly resorbable and prohealing controls, respectively.

      Results

      All test articles and their degradation products were cytocompatible in vitro as measured by cell viability and metabolic activity of bone-marrow macrophages. PerClot induced a stronger proinflammatory, M1-like macrophage response in vitro (P < 0.001) than Arista, likely due to differences in source composition. Histologic examination of the in vivo surgical site showed the almost complete degradation of Arista after 12 h (day 0), whereas both PerClot and potato starch were still present at 28 d with crystals identifiable with polarized light microscopy and periodic acid Schiff (PAS) staining. Macrophage phenotype in vivo showed no differences between PerClot and Arista. Collagen deposition and mononuclear cell accumulation consistent with an early foreign body response were present around PerClot and potato starch crystals, whereas no such cell or connective tissue deposition was noted at the site of Arista or UBM-ECM placement.

      Keywords

      Introduction

      Blood loss contributes to the morbidity and mortality associated with trauma and surgery. Accordingly, effective hemostatic agents are of great clinical interest and are characterized often by biocompatibility, rapid formation of a fibrin clot, and ease of application to the site of interest.
      • Hickman D.A.
      • Pawlowski C.L.
      • Sekhon U.D.S.
      • Marks J.
      • Gupta A.S.
      Biomaterials and advanced technologies for hemostatic management of bleeding.
      ,
      • Day M.W.
      Control of traumatic extremity hemorrhage.
      Powder hemostat products are typically composed of plant carbohydrates, minerals,
      • Barkun A.N.
      • Moosavi S.
      • Martel M.
      Topical hemostatic agents: a systematic review with particular emphasis on endoscopic application in GI bleeding.
      ,
      • Mourad F.H.
      • Leong R.W.
      Role of hemostatic powders in the management of lower gastrointestinal bleeding: a review.
      or extracellular matrix (ECM)-based materials.
      • Singh R.K.
      • Baumgartner B.
      • Mantei J.R.
      • et al.
      Hemostatic comparison of a polysaccharide powder and a gelatin powder.
      ,
      • Bruckner B.A.
      • Ngo U.
      • Ramchandani M.
      • Suarez E.
      • Awad S.
      • Reardon M.
      Application techniques of a novel hemostat in cardiac operations: HEMOBLAST.
      Plant starch-based polysaccharide hemostat powders have proven effective in both preclinical
      • Murat F.J.
      • Ereth M.H.
      • Dong Y.
      • Piedra M.P.
      • Gettman M.T.
      Evaluation of microporous polysaccharide hemospheres as a novel hemostatic agent in open partial nephrectomy: favorable experimental results in the porcine model.
      ,
      • MacDonald M.H.
      • Wang A.Y.
      • Clymer J.W.
      • Hutchinson R.W.
      • Kocharian R.
      An in vivo comparison of the efficacy of hemostatic powders, using two porcine bleeding models.
      and clinical
      • Antisdel J.L.
      • West-Denning J.L.
      • Sindwani R.
      Effect of microporous polysaccharide hemospheres (MPH) on bleeding after endoscopic sinus surgery: randomized controlled study.
      • Facciorusso A.
      • Straus Takahashi M.
      • Eyileten Postula C.
      • Buccino V.R.
      • Muscatiello N.
      Efficacy of hemostatic powders in upper gastrointestinal bleeding: a systematic review and meta-analysis.
      • Huang R.
      • Pan Y.
      • Hui N.
      • et al.
      Polysaccharide hemostatic system for hemostasis management in colorectal endoscopic mucosal resection.
      settings. The composition and processing of these products affect clinically relevant properties such as biodegradability, rate of resorption, and biocompatibility. Arista Absorbable Hemostat (Arista AH), for instance, is a purified plant-based starch that induces rapid coagulation, reduces blood loss, and is completely degraded within 24-48 h in vivo.
      • Antisdel J.L.
      • Janney C.G.
      • Long J.P.
      • Sindwani R.
      Hemostatic agent microporous polysaccharide hemospheres (MPH) does not affect healing or intact sinus mucosa.
      ,
      • Tschan C.A.
      • Nie M.
      • Schwandt E.
      • Oertel J.
      Safety and efficacy of microporous polysaccharide hemospheres in neurosurgery.
      PerClot Polysaccharide Hemostatic System (PerClot PHS) is also a resorbable plant-starch hemostat, can maintain a crystalline form for up to 14 d in vivo and has been associated with a fibrotic healing response
      • Wang Y.
      • Xu M.
      • Dong H.
      • et al.
      Effects of PerClot(R) on the healing of full-thickness skin wounds in rats.
      ,
      • Nilsson G.J.
      Characterization of Crystallinity in Starch Hemostatic Products.
      similar to the fibrotic encapsulation reported to occur with nonresorbable polymeric carbohydrates.
      • Marques A.P.
      • Reis R.L.
      • Hunt J.A.
      An in vivo study of the host response to starch-based polymers and composites subcutaneously implanted in rats.
      • Santos T.C.
      • Marques A.P.
      • Höring B.
      • et al.
      In vivo short-term and long-term host reaction to starch-based scaffolds.
      • Klopfleisch R.
      • Jung F.
      The pathology of the foreign body reaction against biomaterials.
      In the surgical setting, such fibrotic healing response poses a risk of postoperative complications and an increase in medical costs.
      • Capella-Monsonís H.
      • Kearns S.
      • Kelly J.
      • Zeugolis D.I.
      Battling adhesions: from understanding to prevention.
      In fact, surgical glove starch powder has shown to promote granulomas, adhesion formation, and inflammation for decades.
      • Edlich R.F.
      • Long W.B.
      • Gubler D.K.
      • et al.
      Dangers of cornstarch powder on medical gloves: seeking a solution.
      The objective of the present study was to assess the in vitro and in vivo degradation profile and corresponding response, including the immune response, elicited by these two clinically available starch-based hemostat powders with different levels of crystallinity. We hypothesize that the semicrystalline nature and slow resorption of PerClot will elicit a greater proinflammatory response than that of Arista which is both noncrystalline and rapidly resorbed.

      Materials and Methods

      Experimental design overview

      Arista and PerClot crystallinity, degradability, and immune responses were compared in vitro and in vivo. In vitro, crystallinity was assessed before and after enzymatic degradation and the immune reaction was analyzed by exposing murine primary bone marrow naive and M1-activated macrophages to the powders, where their phenotype was analyzed by immunolabeling. In vivo, an abdominal pocket model in rats was performed for 28 d, where the presence of crystals, presence of immune cells, macrophage activation, and collagen deposition were assessed.

      Materials

      Arista Absorbable Hemostat (Arista) powder was provided by BD Bard (New Jersey, US), while PerClot hemostat powder was purchased from Cryolife Europa, Ltd (Hampshire, UK). Extra pure potato starch was purchased from Acros Organics (New Jersey, US) and sterilized by electron beam radiation to be consistent with the sterilized hemostat powders. Porcine urinary bladder matrix (UBM)-ECM was manufactured as previously described
      • Dearth C.L.
      • Slivka P.F.
      • Stewart S.A.
      • et al.
      Inhibition of COX1/2 alters the host response and reduces ECM scaffold mediated constructive tissue remodeling in a rodent model of skeletal muscle injury.
      and comminuted with a Wiley mill.

      In vitro degradation assay and crystallinity assessment

      The amylase-mediated in vitro degradation of Arista versus PerClot was assessed over time by two methods. The first method was based on the ability of both products to absorb fluid and form a gel. The swollen volume (SV) of the viscous gels was monitored at various time points during enzymatic degradation. Briefly, phosphate buffered saline (PBS) with α-amylase from porcine pancreas (Sigma–Aldrich, Missouri, US) was made at a concentration of 200 U/L. The powder (0.5 g) was added to a 10 mL graduated cylinder and hydrated with 13 mL of amylase degradation media. The cylinder was covered with Parafilm and placed in an incubator at 37°C. The SV of the hydrated starch gels in the graduated cylinders was visually determined and recorded at various time points out to 4 wk.
      For the second method, the mass fraction of the powder remaining after exposure to the amylase degradation media was measured in triplicate at 1, 2, and 4 wk. The amylase media containing the degraded materials was passed through cellulose filter papers having a controlled pore size of 8 μm (one filter for each sample) to isolate and collect the remaining product. Material in filter papers was rinsed with purified water then vacuum dried. Finally, dry remnant product was extracted from filter papers, and weighed.
      Polarized light microscopy images of the PerClot and Arista groups without amylase (0 U/L) were taken at time zero to observe baseline crystallinity and in 200 U/L amylase degradation media at 1, 2, 3, and 4 wk to evaluate selective degradation and residual crystallinity (Axio Imager Digital Microscope with polarized lens, Zeiss, New York, US). Potato starch served as a high crystallinity control for the mass loss and microscopy analysis. Potato starch did not absorb fluid and form a hydrated gel and therefore could not be used as a control for SV measurements.

      Isolation of bone marrow macrophages

      All experiments involving laboratory animals were approved by the Institutional Animal Care and Use Committee at the University of Pittsburgh (IACUC protocol #18083483). Murine bone marrow macrophages were isolated as previously described.
      • Huleihel L.
      • Dziki J.L.
      • Bartolacci J.G.
      • et al.
      Macrophage phenotype in response to ECM bioscaffolds.
      Briefly, female 6- to 8-wk-old C57bl/6 mice (Jackson Laboratories, Maine, US) were euthanized via carbon dioxide asphyxiation and cervical dislocation. Using aseptic technique, the skin from the proximal hind limb to the tarsus was removed, the tarsus and stifle were disarticulated, the muscle excised with a scalpel and the tibia isolated. Similarly, the coxafemoral joint was disarticulated and muscle excised for isolation of the femur. Bones were kept on ice and rinsed in a sterile dish containing macrophage complete medium, which consisted of DMEM (Gibco, Massachusetts, US), 10% fetal bovine serum , 10% L929 fibroblast supernatant, 0.1% beta-mercaptoethanol (Sigma–Aldrich, Missouri, US), 1% penicillin/streptomycin, 10 mM nonessential amino acids (Gibco, Massachusetts, US), and 10 mM Hepes buffer. In a sterile environment, the ends of each bone were transected with sterile scissors and the marrow cavity flushed with complete macrophage medium using a 30-G needle. Cells were then washed and plated at 106 cell/mL, and allowed to differentiate into macrophages for 7 d at 37°C and 5% CO2, changing medium every 48 h.

      Cytocompatibility of hemostat powders

      Naive macrophages were treated with basal media consisting of DMEM, 10% fetal bovine serum, 1% penicillin/streptomycin, and containing one of the hemostat powders (Arista, PerClot, potato starch as high crystallinity control, or degraded UBM-ECM as prohealing control), their degradation products (powders after enzymatic degradation), pepsin, or α-amylase at 2.5, 5.0, 7.5, 10.0 and 15.0 mg/mL or equivalent dilutions for the enzymes. Basal media alone was used as control. After 24 h of incubation, cells were washed with PBS and metabolic activity was measured using a 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide assay (Invitrogen, Massachusetts, US), whereas cell viability was assessed using a LIVE/DEAD assay (Invitrogen, Massachusetts, US).

      Macrophage activation and powder treatment

      Naive macrophages were treated with basal media and 20 ng/mL IFNγ and 100 ng/mL lipopolisaccharyde (LPS) to induce a M1 phenotype or 20 ng/mL IL-4 to induce M2 phenotype. To test the effect of each powder on naive macrophages phenotype, these were treated with basal media containing either (i) Arista, (ii) degradation products of Arista, (iii) PerClot, (iv) degradation products of PerClot, (v) potato starch, (vi) degradation products of potato starch, (vii) degraded UBM-ECM, (viii) pepsin, or (ix) α-amylase at 5 mg/mL or equivalent dilution. Cells were treated for 24 h at 37 ⁰C and 5% CO2, then washed with PBS, and fixed in 4% paraformaldehyde for 20 min at room temperature for immunolabeling.
      To assess the effect of hemostat powders on M1 activated macrophages, naive macrophages were treated with 20 ng/mL IFNγ and 100 ng/mL LPS for 6 h at 37°C and 5% CO2. Then, M1 activated macrophages were incubated with the same conditions tested in naive macrophages for 24 h at 37°C and 5% CO2.

      Immunolabelling of macrophages

      Fixed cells were washed in PBS and incubated in blocking solution (PBS, 0.1% Triton-X and 5% normal goat serum) for 1 h at room temperature. Blocking solution was removed and cells were incubated during 1 h at room temperature in blocking solution containing primary antibodies for F4/80 (1:200, NB600404, Fisher Scientific, Massachusetts, US) as pan-macrophage marker, iNOS (1:100, PA3030A, Thermo Fisher Scientific, Massachusetts, US) as M1 proinflammatory marker, and Fizz1 (1:200, 50-402-368, Peprotech, New Jersey, US) and Arginase 1 (1:200, ab91279, Abcam, Massachusetts, US) as M2 pro-healing markers. After washing three times in PBS, cells were incubated in blocking solution containing the correspondent secondary antibodies (Alexa Fluor donkey antirat 488 or donkey antirabbit 488, Invitrogen) for 1 h at room temperature and 1:200 dilution. After washing again with PBS, nuclei were counterstained with 4',6-diamidino-2-phenylindole and images of three 10× fields were taken for each well using a live-cell microscope (Observer.Z1, Zeiss, New York, US). Exposure times for the different treatments were standardized to M1 or M2 activated control macrophages. Images were quantified using CellProfiler (Broad Institute, Massachusetts, US) for the percentage of positive F4/80, iNOS, Arginase, and Fizz1 cells. Arginase and iNOS percentages were employed to calculate M2:M1 ratio for each condition.

      In vivo abdominal pocket model in rat

      An intraabdominal wall pocket model in 6-8 wk old Sprague–Dawley rats (Jackson Laboratories, US) was used to evaluate the immune response elicited by each hemostat powder in vivo (N = 5 per group, 5 groups, 5 time points). This procedure was conducted with the approval of the Institutional Animal Care and Use Committee at the University of Pittsburgh (IACUC protocol #20057255). Briefly, rats were anaesthetized with 5% isoflurane and the ventral abdomen was shaved and scrubbed with betadine solution and 70% ethanol. A 2 cm right paramedian full-thickness incision was made and the cut surface of the intraabdominal wall was exposed and isolated. A 1 cm long pocket defect was created in the muscle and filled with 15-20 mg of either Arista, PerClot, potato starch, or UBM-ECM powders. A defect without any powder was used as control. The pocket defect was closed with 4-0 nonresorbable polypropylene suture and the full thickness abdominal wall defect was then repaired with 3-0 absorbable polyglycolic acid suture. Subcutaneous injections of 0.05 mg/kg buprenorphine every 12 h and 5 mg of oral enrofloxacin every 24 h were administered for 72 h postoperatively. Rats were euthanized after day 0 (12 h) and at days 3, 7, 14, and 28 by CO2 asphyxiation and cervical dislocation, and the abdominal muscle containing the pocket defect was harvested and fixed in 10% buffered formalin for subsequent preparation of slides for histologic examination.

      Histology and immunolabeling

      Formalin-fixed tissue samples were embedded in paraffin and cut transversally in 5 μm thick sections. The sections were cleared in xylene, rehydrated in decreasing ethanol dilutions, stained with hematoxylin and eosin (H&E), periodic acid Schiff (PAS), or picrosirius red, and coverslipped. Bright field images and polarized light images were obtained with a live cell microscope (Observer.Z1, Zeiss, US). Polarized light and H&E images at low magnification (10×) were used to semiquantitatively assess the presence of crystals. H&E images under bright field at high magnification (40×) were used to assess the presence of immune cells. PAS images at low magnification (10×) were taken to assess the presence of starch-based (carbohydrates) polymers. Picrosirius red images were taken under bright field and polarized light to assess collagen deposition at day 28. Image analysis and semiquantitative and quantitative scoring was performed as described in Table, by 3 trained and blinded individuals.
      TableHistology scoring for in vivo abdominal pocket model.
      CrystalsNeutrophils and MNGCEosinophils/PAS
      ScoreDetectionScoreNumber of cellsScoreNumber of cells
      0No detected00-None
      1Minimal/barely detected11 to 5+1 to 5
      2Mild/slightly detectable26 to 15++More than 5
      3Moderate/easily detectable316 to 40+++More than 10
      4Markedly/very evident441 or more
      For immunolabelling of paraffin sections, the slides were cleared and rehydrated in xylene and decreasing ethanol dilutions, respectively. Antigen retrieval was achieved by boiling the slides in citric acid solution (10 mM, pH 6) for 1 min. After rinsing in Tris buffered-saline 0.01% Tween 20 (TBST), sections were blocked in Pierce protein-free blocking buffer (PI3751, Fisher Scientific, US) for 45 min at room temperature and incubated in primary antibody solution diluted in blocking buffer for 3 min under microwave heating. Sections were then washed three times in TBST, incubated with horseradish peroxidase (HRP)-conjugated secondary antibodies for 3 min under microwave heating, washed three more times in TBST, and incubated for 10 min at room temperature in the correspondent OPAL signal generation kit (Akoya Biosciences, Massachusetts, US) dilution. The sections were then exposed to anti CD206 (AF2535, R&D Systems, Minnesota, US), TNF-α (ab6671, Abcam, Massachusetts, US) and CD68 (MCA341R, Bio-Rad, California, US) primary antibodies and rabbit antigoat HRP, goat antirabbit HRP, and goat antimouse HRP secondary antibodies (Dako, California, US), respectively. Draq5 (Fisher Scientific, US) was used to counterstain nuclei. Sections were washed in PBS and mounted in Fluoromount (Fisher Scientific, Massachusetts, US) mounting medium. Finally, images were taken with a live cell microscope (Observer Z1, Zeiss, New York, US), and analyzed with CellProfiler (Broad Institute, Massachusetts, US) software to calculate the percentage of macrophages (CD68 positive cells), those expressing TNF-α (M1), CD206 (M2), and the M2/M1 ratio.

      Statistical analysis

      Statistical analysis was performed with Graph Pad Prism 9 (Graph Pad Holdings LLC, California, US). Normal distribution of data was assessed using Shapiro-Wilk test. One-way ANOVA and Fisher's least significant difference (LSD) post hoc analysis were used to assess statistical significance for normally distributed data, which was assumed when P < 0.05. Kruskal–Wallis and Dunn's post hoc nonparametric tests were used for nonnormally distributed data.

      Results

      In vitro degradation and crystallinity assessment

      Polarized microscopy imaging of the materials hydrated in PBS without amylase illustrated that Arista is uniform microspheres composed of noncrystalline starch while PerClot is comprised of both amorphous and crystalline starch with granular morphology similar to raw potato starch (Fig. 1).
      Figure thumbnail gr1
      Fig. 1Bright field (top row) and polarized light (bottom row) microphotographs of Arista, PerClot, and potato starch hydrated in PBS (10× magnification). Scale bar (valid for all images) 100 μm.
      After initial exposure to amylase degradation media, Arista underwent a controlled loss of SV within 2-4 d (Supplementary Fig. 1). PerClot appeared to follow a biphasic degradation trend in amylase degradation media. Within the first 6 h of amylase exposure, PerClot had rapid initial loss of SV followed by a period of very slow degradation of PerClot‘s residual crystalline portion for the remainder of the 4 wk. Polarized microscopy imaging confirmed that there was no crystallinity detected in Arista at any time point, whereas PerClot showed selective degradation of the hemostatic agent's amorphous starch components leaving a highly concentrated crystalline starch in the residual PerClot material out to 4 wk (Fig. 2B).
      Figure thumbnail gr2
      Fig. 2Enzymatic degradation in vitro of PerClot and Arista. The remaining mass after degradation in 200 U/L α-amylase approximately corresponded to a 75% and 12% in potato starch and PerClot, respectively after 28 d, whereas no material remained in Arista (A). Bright field and polarized light microphotographs of PerClot and Arista (B) confirmed such observations. Data presented as average ± standard deviation (n = 3), ∗∗∗ indicates a significant difference where P < 0.001. Scale bar (valid for all images) 100 μm.
      The mass loss measurements following amylase degradation indicated that Arista was absent by the first time point at 7 d while PerClot and potato starch had 12% and 75% remaining, respectively (Fig. 2A). A similar amount of mass for PerClot and raw potato starch persisted through the last time point at 28 d.

      Cytocompatibility in vitro

      Cytocompatibility in vitro of starch hemostat powders, their degradation products and hydrolytic enzymes was evaluated by the metabolic activity of murine naive BM macrophages and by cell viability analysis. No reduction in the metabolic activity was present when macrophages were incubated with the different concentrations of hemostat powders (Fig. 3A), their degradation products (Fig. 3B), or the equivalent dilutions of enzyme (Fig. 3C). Arista degradation products at high concentrations (7.5 to 15 mg/mL) elicited a significant increase in metabolic activity but did not affect the cell viability of macrophages (Fig. 3D). Similarly, the rest of the conditions did not affect cell viability. In view of these results, the rest of the experiments in vitro were conducted with a concentration of 5 mg/mL or equivalent for the enzymes.
      Figure thumbnail gr3
      Fig. 3Metabolic activity and viability of naive macrophages. Metabolic activity of naive macrophages showed no significant decrease when incubated with the hemostat powders (A), their degradation products (B), nor hydrolytic enzymes (C). A significant increase of metabolic activity was elicited by the degradation products of Arista at 7.5, 10, and 15 mg/mL (B), although this effect was not related to any measurable detrimental impact on macrophage cell viability (D), equally to that observed in the rest of concentrations and conditions tested. Data are expressed as average and standard deviation of values normalized to control (n = 3, N = 3). ∗ Indicates a significant difference where P < 0.05.

      Naive macrophage phenotype analysis

      Mouse naive bone marrow macrophage phenotype was determined after treatment with the hemostat powders, their degradation products, and catalytic enzymes at 5 mg/mL. To this end, iNOS as M1 marker, Arginase and Fizz1 as M2 markers, and F4/80 as pan macrophage marker were assessed by immunocytochemistry. LPS/IFNγ (M1) and IL-4 (M2) treated macrophages were used as positive controls. Immunolabelling images showed the presence of iNOS positive cells in LPS, amylase, UBM-ECM, PerClot and degraded PerClot conditions, whereas the number of cells with M2 marker Arginase was higher in IL-4, UBM-ECM and degraded Arista conditions. Fizz1 positive cells were only observed in IL-4 and UBM-ECM groups. F4/80 positive cells seemed lower in IL-4, amylase and the degradation products of PerClot and potato starch (Supplementary Fig. 2).
      Quantification of the percentage of cells expressing each marker confirmed the observations from immunofluorescence images. Naive macrophages treated with IL-4, amylase, and the degradation products of PerClot and potato starch showed a lower percentage of F4/80 positive cells (P < 0.001) (Fig. 4A), while higher percentage of cells expressing iNOS (Fig. 4B) were found in macrophages treated with LPS, amylase, UBM-ECM, and PerClot and its degradation products (P < 0.001). These results were in distinct contrast to the results observed for Arista and potato starch. M2 marker Fizz1 was only expressed in IL-4 treated macrophages (Fig. 4C), while higher percentage of cells positive for Arginase were present in UBM-ECM (P < 0.01) and pepsin (P < 0.05) (Fig. 4D). M2:M1 percentage of macrophages ratio was higher when treated with the degradation products or Arista and with potato starch, but without significant differences (Fig. 4E).
      Figure thumbnail gr4
      Fig. 4Naive macrophage phenotype response to hemostat powders measured by quantification of cell percentage expressing F4/80 (A), iNOS (B), Fizz1 (C), and Arginase (D). Results showed the higher expression of iNOS (M1 marker) in amylase, UBM-ECM, PerClot, and degraded PerClot, whereas M2 marker Arginase was observed in UBM-ECM and pepsin conditions. M2:M1 ratio (E) was higher in degraded Arista and Potato starch, but with no significant differences. Data are expressed as average and standard deviation (n = 3, N = 3). ∗, ∗∗, ∗∗∗ indicate a significant difference with the control where P < 0.05, P < 0.01 and P < 0.001, respectively.

      M1-activated macrophage phenotype analysis

      Macrophages were activated to a M1 phenotype with LPS/IFNγ, then treated with hemostat powders or their degradation products, and their phenotype was characterized after 24 h. Immunofluorescence images (Supplementary Fig. 3) showed macrophages in all groups expressed the M1 marker iNOS, although a lower number of positive cells were present in the IL-4, UBM-ECM, Arista and their degradation products groups, and the PerClot group before degradation conditions. M2 markers were not present except for Arginase in IL-4 treated macrophages. F4/80 positive cells were noticeably lower in amylase, Arista and its degradation products, degradation products of PerClot, and potato starch and its degradation products.
      Image analysis quantification confirmed the lower percentage of F4/80 cells treated with amylase, Arista, the degradation products of Arista, PerClot and potato starch (P < 0.001), and potato starch before degradation (P < 0.01) (Fig. 5A). Quantification of positive iNOS cells (Fig. 5B) showed no statistically significant differences with the control. The M2 marker Fizz1 was not detected in any of the conditions (Fig. 5C) and Arginase positive percentage of cells was only significantly higher (P < 0.001) in IL-4 (Fig. 5D).
      Figure thumbnail gr5
      Fig. 5M1 activated macrophages phenotype after treatment with hemostat powders assessed by quantification of cell percentage expressing F4/80 (A), iNOS (B), Fizz1 (C), and Arginase (D). Results revealed expression of iNOS (M1 marker) in all the conditions with lower trends in pepsin, UBM-ECM, Arista and its degradation products and PerClot. Arginase was only observed in the positive control and Fizz1 was not observed in any condition. F4/80 expressing cells were lower in amylase, Arista and its degradation products, the degradation products of PerClot and potato starch and its degradation products. Data are expressed as average and standard deviation (n = 3, N = 3). ∗, ∗∗, ∗∗∗ indicate a significant difference with the control where P < 0.05, P < 0.01 and P < 0.001, respectively.

      In vivo abdominal pocket model

      Morphologic analysis

      Macroscopically, the surgical site showed no signs of inflammation at any time. The sham and Arista groups showed minimal evidence of the surgical pocket at 1 wk and no remaining Arista material could be found at any time point. UBM-ECM powder was still visible at 7 d but was no longer visible by 14 d (Supplementary Fig. 4). The PerClot and potato starch groups showed mild adhesions on the peritoneal surface of the surgical site, being more prominent in the potato starch group.
      Polarized microscopy on H&E sections showed the clear presence of crystals in the PerClot and potato starch groups throughout the 28 d of study (Fig. 6), although they were more prominent in the potato starch group. Semiquantitative analysis confirmed such observations, showing a higher score of crystals (P < 0.05 at days 0,3,7, and 28 and P < 0.001 at day 14) in PerClot and potato starch (P < 0.001) than in sham, UBM-ECM and Arista groups which only showed background signal derived from the suture material. When comparing values for the potato starch and PerClot groups, a lower score was present at all time points (P < 0.05 at day 0, P < 0.001 in the other time points) in the PerClot group which showed a decreasing score over time.
      Figure thumbnail gr6
      Fig. 6Hematoxylin and eosin-stained tissue sections under bright field and polarized microscopy and quantification of crystals with the methodology specified in . Crystals were detected throughout the 28 d of the study in PerClot and potato starch groups. Data are expressed as median and interquartile range (N = 5). ∗, ∗∗, ∗∗∗, ∗∗∗∗ indicate a significant difference where P < 0.05, P < 0.01, P < 0.001 and P < 0.0001, respectively. Scale bar 100 μm (valid for all pictures).
      Similar results were observed in PAS-stained sections, which assessed the presence of carbohydrates (starch) within the tissue at the surgery site. The sham and UBM-ECM conditions did not show any presence of carbohydrates as expected, whereas Arista showed almost complete degradation after day 0, where only scattered positive staining particles could be seen, and these completely cleared after 3 d (Fig. 7). In contrast, the PerClot and potato starch groups showed particles that were present until day 28, although with lower incidence in the PerClot group as confirmed by the qualitative scoring.
      Figure thumbnail gr7
      Fig. 7PAS staining of tissue sections to assess the presence of starch particles (bright purple). Arista showed only scattered particles at day 0, showing none from day 3. The PerClot and potato starch groups showed carbohydrate particles throughout the 28 d of the study, with greater prominence in the potato starch group. Qualitative scoring confirmed such observations. Scale bar 100 μm (valid for all pictures).
      Quantification and scoring of cell populations in high magnification images showed a higher presence of eosinophils in the UBM-ECM group at day 3, while at day 7 UBM-ECM, PerClot and potato starch groups showed a higher presence of eosinophils. A greater number of neutrophils were present in the UBM-ECM groups day 3 (Fig. 8C) (P < 0.01), whereas no differences were present between groups at any subsequent time point.
      Figure thumbnail gr8
      Fig. 8Qualitative scoring of eosinophils (A) and semiquantitative scoring of neutrophils and multinucleated giant cells (MNGC) at days 0 (B), 3 (C), 7 (D), 14 (E), and 28 (F). Eosinophils were more present in UBM-ECM, PerClot and potato starch groups until day 7, while the UBM-ECM group attracted more neutrophils at day 0 and day 3. Data expressed as median and interquartile range (N = 5). ∗∗ indicates a significant difference where P < 0.01.
      Multinucleated giant cells (MNGC) only showed a higher presence in the potato starch group than in sham and UBM-ECM groups at day 14 (Fig. 8E), with no significant differences in the rest of time points.

      Macrophage immunofluorescence analysis

      Immunolabelling of tissue sections for macrophages for CD68 (pan macrophage marker), TNF-α (M1 marker), and CD206 (M2 marker) was used to assess the innate immune response in vivo. Immunofluorescence images (Fig. 9) showed a high cellularity and mixed population of TNF-α and CD206 positive cells at day 3, decreasing gradually in time. Also, the proportion of CD206 cells was visually greater at days 7 and 14 in the UBM-ECM group, where the gradual degradation of the material (red autofluorescence) could also be observed.
      Figure thumbnail gr9
      Fig. 9Immunofluorescence images in vivo of tissue sections detecting CD68 (red, pan macrophage marker), TNF-α (orange, M1 marker), CD206 (green, M2 marker), and nuclei (blue). Scale bar 50 μm (valid for all images).
      Immunofluorescence microscopy images were analyzed to quantify the percentage of macrophages, M1 positive macrophages, M2 positive macrophages, and the M2:M1 ratio (Fig. 10). All groups showed a high percentage of macrophages at day 3, decreasing from day 7. The Arista group showed the lowest percentage of macrophages at days 14 and 28, being significantly less than the potato starch group (P < 0.05) and with PerClot at day 14 (P < 0.01). The percentage of M1 positive macrophages was greater in the Arista than in UBM-ECM and PerClot groups (P < 0.05 and P < 0.01, respectively) but showed no differences from the sham group (P > 0.05) at day 7; the rest of time points did not show any significant differences between groups. The UBM-ECM group elicited the highest percentage of M2 macrophages at days 7 (P < 0.05 with Arista) and 14 (P < 0.001 with sham, P < 0.01 with PerClot), while at day 28, M2 macrophage percentage was higher in the Arista group (P < 0.0001 with sham, P < 0.01 with UBM-ECM and potato starch groups). The UBM-ECM group showed the higher M2:M1 ratio at days 7 (P < 0.01) and 14 (P < 0.05 with sham), while no significant differences were found between the hemostat powder groups.
      Figure thumbnail gr10
      Fig. 10Quantification of total macrophage percentage, M1 positive macrophages percentage, M2 positive macrophages percentage, and M2:M1 ratio from immunofluorescence images. Data are expressed as average and standard deviation (N = 5). ∗, ∗∗, ∗∗∗, ∗∗∗∗ indicate a significant difference where P < 0.05, P < 0.01, P < 0.001 and P < 0.0001, respectively.

      Collagen deposition analysis

      Picrosirius red images taken under polarized microscopy were used to assess the presence of collagen deposition in the tissue sections at day 28. The sham, UBM-ECM and Arista groups did not show the presence of collagen deposition (Fig. 11). In contrast, the PerClot and potato starch groups showed deposition of red (mature) and green/orange (immature, loosely packed) collagen around the crystals present in the abdominal pocket area, which were also detectable under polarized microscopy; although both crystals and collagen were more prominent in the potato starch group. H&E images at day 28 also showed signs of fibrotic encapsulation around PerClot and potato starch crystals (Fig. 11).
      Figure thumbnail gr11
      Fig. 11Bright field and polarized microscopy images of picrosirius red and H&E-stained tissue sections at day 28. Higher presence of deposited collagen can be observed around PerClot and potato starch crystals. High magnification H&E showed initiation of fibrotic encapsulation around the crystals (C), characterized by the accumulation of cells (black arrows) and ECM deposition in PerClot and potato starch. Yellow line indicates the area of the polarized images at high magnification (Polarized 10×). S = Suture; P = Pocket. Scale bar 500 μm for low magnification, 100 μm for 10× magnification and 50 μm for 40× magnification.

      Discussion

      Plant starch-based polysaccharide hemostat powders are used to promote clotting and reduce coagulation time in a variety of surgical procedures.
      • Antisdel J.L.
      • West-Denning J.L.
      • Sindwani R.
      Effect of microporous polysaccharide hemospheres (MPH) on bleeding after endoscopic sinus surgery: randomized controlled study.
      ,
      • Facciorusso A.
      • Straus Takahashi M.
      • Eyileten Postula C.
      • Buccino V.R.
      • Muscatiello N.
      Efficacy of hemostatic powders in upper gastrointestinal bleeding: a systematic review and meta-analysis.
      These powders absorb blood, concentrate cells and coagulation factors, and arrest hemorrhage within seconds to minutes of application.
      • Zheng C.
      • Bai Q.
      • Wu W.
      • et al.
      Study on hemostatic effect and mechanism of starch-based nano-microporous particles.
      Ideally, once these materials have completed their task, they are reabsorbed and do not interfere with the healing process. However, starch-based materials, if not processed with a hydrolysis step that promotes their degradability, do not resorb quickly and can remain at the anatomic site of application for weeks to months.
      • Santos T.C.
      • Marques A.P.
      • Höring B.
      • et al.
      In vivo short-term and long-term host reaction to starch-based scaffolds.
      Although starch-based polymers generally do not trigger a notable acute inflammatory reaction,
      • Torres F.G.
      • Troncoso O.P.
      • Pisani A.
      • Gatto F.
      • Bardi G.
      Natural polysaccharide nanomaterials: an overview of their immunological properties.
      their prolonged presence after implantation can lead to a foreign body response and chronic inflammation. The present study compared two clinically available hemostat starch powders: Arista, a rapidly resorbable starch powder with no crystallinity, and PerClot a slowly resorbable, semicrystalline starch powder.
      Crystallinity of starch is caused by the compact packing of amylose and amylopectin, resulting in crystallites that orient radially to the principal axis causing a specific birefringence pattern called Maltese crosses
      • Torres F.G.
      • Troncoso O.P.
      • Pisani A.
      • Gatto F.
      • Bardi G.
      Natural polysaccharide nanomaterials: an overview of their immunological properties.
      under polarized light, as observed in PerClot and potato starch in this study. This crystallinity was also present after α-amylase in vitro digestion of PerClot and potato starch in contrast to Arista that showed no crystallinity and complete degradation. Residual PerClot crystals showed a noticeably lower size than those present in potato starch after degradation. These findings suggest partial degradability of PerClot. These differences between PerClot and potato starch could account for different crystallinity ratios, particle size, and/or particle topography.
      • Torres F.G.
      • Troncoso O.P.
      • Pisani A.
      • Gatto F.
      • Bardi G.
      Natural polysaccharide nanomaterials: an overview of their immunological properties.
      ,
      • Dhital S.
      • Warren F.J.
      • Butterworth P.J.
      • Ellis P.R.
      • Gidley M.J.
      Mechanisms of starch digestion by alpha-amylase-Structural basis for kinetic properties.
      The processing of Arista eliminates crystallinity and increases the accessibility of α-amylase to α-glucan chains for their degradation. This enzymatic interaction process is sterically hindered in highly packed/crystalline starch structures,
      • Dhital S.
      • Warren F.J.
      • Butterworth P.J.
      • Ellis P.R.
      • Gidley M.J.
      Mechanisms of starch digestion by alpha-amylase-Structural basis for kinetic properties.
      thus explaining the slow degradation of PerClot and potato starch.
      BM naive macrophage metabolic activity and cell viability analysis in vitro showed the cytocompatibility of all the starch powders, as has been shown in previous work with similar starch-based materials in 3T3 fibroblasts
      • Torres F.G.
      • Troncoso O.P.
      • Grande C.G.
      • Díaz D.A.
      Biocompatibilty of starch-based films from starch of Andean crops for biomedical applications.
      and monocyte-like THP-1 cells.
      • Gatto F.
      • Troncoso O.P.
      • Brunetti V.
      • et al.
      Human monocyte response to Andean-native starch nanoparticles.
      The in vitro immune response results showed that PerClot and its degradation products trigger a higher percentage of M1 or proinflammatory cells than Arista and potato starch groups. These findings suggest that starch powder’s physical properties would not influence the different immune reaction observed in vitro, given the structural similarities between potato starch and PerClot. Instead, these differences are more likely the result of different starch source composition and the interaction of these components with the membrane of immune cells.
      • Gatto F.
      • Troncoso O.P.
      • Brunetti V.
      • et al.
      Human monocyte response to Andean-native starch nanoparticles.
      M2 markers were not influenced by any of the starch powders in naive macrophages, while M2:M1 ratio showed no statistically significant differences. Amylase also elicited an increase in iNOS positive cells likely due to residual enzyme activity. Interestingly, a lower number of cells expressing F4/80 and a pan macrophage marker were found in amylase degraded PerClot, and degraded potato starch treated cells, which could be related to the different activation status of the macrophages, were found under these conditions and/or the interaction of the carbohydrates and enzyme with the glycoproteic marker.
      • Dos Anjos Cassado A.
      F4/80 as a major macrophage marker: the case of the peritoneum and spleen.
      In addition, M1-activated macrophages phenotype was not modulated by any of the starch powders. Overall, these results suggest that PerClot is associated with a greater potential to induce a proinflammatory reaction than Arista.
      An intraabdominal pocket implantation model in rats for 28 d was conducted to characterize the immune response elicited in vivo by Arista and PerClot. Consistent with the in vitro results, Arista was mostly degraded after 12 h of implantation and completely degraded by 3 d. In contrast, PerClot and potato starch had residual crystalline material for the entire duration of the study, although PerClot showed partial degradation. Previous work on starch-blended scaffolds has reported the complete resorption of the starch fraction of these materials,
      • Marques L.
      • Holgado L.A.
      • Simões R.D.
      • et al.
      Subcutaneous tissue reaction and cytotoxicity of polyvinylidene fluoride and polyvinylidene fluoride-trifluoroethylene blends associated with natural polymers.
      whereas crystallinity of potato starch granules inversely relates to their resorption,
      • Velasquez D.
      • Pavon-Djavid G.
      • Chaunier L.
      • Meddahi-Pellé A.
      • Lourdin D.
      Effect of crystallinity and plasticizer on mechanical properties and tissue integration of starch-based materials from two botanical origins.
      matching the findings of the present study.
      None of the hemostat powders elicited an inflammatory reaction after implantation, and no differences in macrophage activation were observed. The control for a positive M2:M1 ratio was UBM-ECM powder, which showed a higher M2:M1 ratio after 7 and 14 d. These results are consistent with the reported antiinflammatory phenotype elicited by ECM degradation products.
      • Sadtler K.
      • Sommerfeld S.D.
      • Wolf M.T.
      • et al.
      Proteomic composition and immunomodulatory properties of urinary bladder matrix scaffolds in homeostasis and injury.
      ,
      • Badylak S.F.
      • Gilbert T.W.
      Immune response to biologic scaffold materials.
      Interestingly, the lack of any inflammatory response by macrophages in vivo contrasted with the macrophage phenotype assessment in vitro. These differences may account for the limitations of the in vitro setting compared to the complexity of the healing process in native tissue, where many role players (i.e., cell types, cytokines, chemokines, and physical stimuli) can influence macrophage phenotype. In addition, the inflammation elicited by the surgical procedure itself could also have masked any initial low-mild inflammatory reaction elicited by any of the materials. A clearance of neutrophils and macrophages in the surgical site was observed in all groups after day 3, consistent with the lack of any notable inflammatory tissue response. Arista was found to elicit a lesser presence of eosinophils at day 7 compared to PerClot and potato starch, and a less prominence of macrophages than potato starch at days 14 and 28. Also, potato starch showed a higher presence of MNGC at day 14. These collective results indicate that the high degree of crystallinity in potato starch, and therefore low degradability, results in the initiation of a foreign body reaction.
      • Santos T.C.
      • Marques A.P.
      • Höring B.
      • et al.
      In vivo short-term and long-term host reaction to starch-based scaffolds.
      ,
      • Velasquez D.
      • Pavon-Djavid G.
      • Chaunier L.
      • Meddahi-Pellé A.
      • Lourdin D.
      Effect of crystallinity and plasticizer on mechanical properties and tissue integration of starch-based materials from two botanical origins.
      PerClot showed similar trends to potato starch although without significant differences from Arista, which could be due to the lower extent of crystallinity of PerClot versus that of potato starch.
      • Velasquez D.
      • Pavon-Djavid G.
      • Chaunier L.
      • Meddahi-Pellé A.
      • Lourdin D.
      Effect of crystallinity and plasticizer on mechanical properties and tissue integration of starch-based materials from two botanical origins.
      However, both PerClot and potato starch showed the deposition of collagenous connective tissue around the remaining crystals at day 28, which could indicate a foreign body response and fibrotic encapsulation of the undegraded crystal particles.
      The response elicited by PerClot in the present study indicates potential fibrosis and adhesion formation in the surgery area, which entails a risk of mid and long-term postoperative complications and increase of medical costs.
      • Capella-Monsonís H.
      • Kearns S.
      • Kelly J.
      • Zeugolis D.I.
      Battling adhesions: from understanding to prevention.
      ,
      • Trochsler M.
      • Maddern G.J.
      Adhesion barriers for abdominal surgery: a sticky problem.
      In recent clinical studies, PerClot has been reported to increase the rate of complications such as a higher risk of infection in neck and head surgery,
      • Rao K.
      • Gomati A.
      • Yuen Hao Tong E.
      • W Ah-See K.
      • Shakeel M.
      Use of PerClot® in head and neck surgery: a Scottish centre experience.
      higher incidence of inflammation markers, and fever in cardiac surgery.
      • Tscholl V.
      • Spann F.
      • Moses J.
      • et al.
      Prospective randomized study evaluating the effects of PerClot® (Polysaccharide Hemostatic System) application in patients with high bleeding risk undergoing cardiac rhythm device implantation.
      Although these complications were not related to a long-term presence of the material, they indicate the potential inflammatory effect of PerClot. Rapidly resorbable starch hemostat powders, such as Arista,
      • Tschan C.A.
      • Nie M.
      • Schwandt E.
      • Oertel J.
      Safety and efficacy of microporous polysaccharide hemospheres in neurosurgery.
      ,
      • Antisdel J.L.
      • Matijasec J.L.
      • Ting J.Y.
      • Sindwani R.
      Microporous polysaccharide hemospheres do not increase synechiae after sinus surgery: randomized controlled study.
      may decrease the risk of postoperative complications and prevent a long-term foreign-body response.
      The present study comparatively assessed the immune response to two starch-based hemostat powders using a macrophage in vitro model and a rat abdominal pocket in vivo model. Although these models fulfilled the objective of the study, there are limitations. First, in vitro models lack the heterogeneity in cell populations, stimuli and signals present during the in vivo immune response, and the results obtained in vitro must be interpreted with these considerations. Second, the present in vivo model may not be fully representative of every clinical scenario in which these hemostats are employed. For example, more challenging conditions such as more severe bleeding and/or larger surface area may be encountered in the clinical scenario. Lastly, the animals employed in this study for both macrophage isolation and surgeries were all female for husbandry and animal well-being reasons. Although polysaccharide-based biomaterials are known to present certain sex-bias upon immune response,
      • Xu J.
      • Nie N.
      • Wu B.
      • et al.
      The personalized application of biomaterials based on age and sexuality specific immune responses.
      such bias affects mostly the CD4+ population and are not related to the foreign body reaction to biomaterials and encapsulation observed in the present study. The authors do not expect that there is an influence of sex in the results observed in this study. However, further in vivo and clinical studies are needed to better understand the extents and mechanisms by which highly crystalline starch-based hemostats trigger an inflammatory foreign body response and encapsulation across sexes.

      Conclusions

      The present study highlights differences in the crystallinity, degradation rate, inflammatory cell response, and fibrotic reaction to two starch-based hemostats. These differences may explain clinical outcome differences between the two products.

      Author Contributions

      HCM carried out the experimental in vitro and in vivo work. AS and BC assisted with in vivo work, whereas MF, GL, and KG assisted in the in vitro portion. HCM wrote the manuscript, while HCM, GL, KG, and SFB edited the manuscript. SFB was responsible for obtaining funding for the project, participated in the study design and interpretation of data, and supervised the study.

      Acknowledgments

      Funding for this work was provided in part by Becton Dickinson. The assistance of the staff of the Division of Laboratory Animal Research at the University of Pittsburgh is gratefully acknowledged.

      Disclosure

      S. Badylak conducts sponsored research for Becton Dickinson and MF, GL, and KG are employees of BD.

      Funding

      None.

      Supplementary Materials

      References

        • Hickman D.A.
        • Pawlowski C.L.
        • Sekhon U.D.S.
        • Marks J.
        • Gupta A.S.
        Biomaterials and advanced technologies for hemostatic management of bleeding.
        Adv Mater. 2018; 301700859
        • Day M.W.
        Control of traumatic extremity hemorrhage.
        Crit Care Nurse. 2016; 36: 40-51
        • Barkun A.N.
        • Moosavi S.
        • Martel M.
        Topical hemostatic agents: a systematic review with particular emphasis on endoscopic application in GI bleeding.
        Gastrointest Endosc. 2013; 77: 692-700
        • Mourad F.H.
        • Leong R.W.
        Role of hemostatic powders in the management of lower gastrointestinal bleeding: a review.
        J Gastroenterol Hepatol. 2018; 33: 1445-1453
        • Singh R.K.
        • Baumgartner B.
        • Mantei J.R.
        • et al.
        Hemostatic comparison of a polysaccharide powder and a gelatin powder.
        J Invest Surg. 2019; 32: 393-401
        • Bruckner B.A.
        • Ngo U.
        • Ramchandani M.
        • Suarez E.
        • Awad S.
        • Reardon M.
        Application techniques of a novel hemostat in cardiac operations: HEMOBLAST.
        J Card Surg. 2019; 34: 849-853
        • Murat F.J.
        • Ereth M.H.
        • Dong Y.
        • Piedra M.P.
        • Gettman M.T.
        Evaluation of microporous polysaccharide hemospheres as a novel hemostatic agent in open partial nephrectomy: favorable experimental results in the porcine model.
        J Urol. 2004; 172: 1119-1122
        • MacDonald M.H.
        • Wang A.Y.
        • Clymer J.W.
        • Hutchinson R.W.
        • Kocharian R.
        An in vivo comparison of the efficacy of hemostatic powders, using two porcine bleeding models.
        Med Devices (Auckl). 2017; 10: 273-279
        • Antisdel J.L.
        • West-Denning J.L.
        • Sindwani R.
        Effect of microporous polysaccharide hemospheres (MPH) on bleeding after endoscopic sinus surgery: randomized controlled study.
        Otolaryngol Head Neck Surg. 2009; 141: 353-357
        • Facciorusso A.
        • Straus Takahashi M.
        • Eyileten Postula C.
        • Buccino V.R.
        • Muscatiello N.
        Efficacy of hemostatic powders in upper gastrointestinal bleeding: a systematic review and meta-analysis.
        Dig Liver Dis. 2019; 51: 1633-1640
        • Huang R.
        • Pan Y.
        • Hui N.
        • et al.
        Polysaccharide hemostatic system for hemostasis management in colorectal endoscopic mucosal resection.
        Dig Endosc. 2014; 26: 63-68
        • Antisdel J.L.
        • Janney C.G.
        • Long J.P.
        • Sindwani R.
        Hemostatic agent microporous polysaccharide hemospheres (MPH) does not affect healing or intact sinus mucosa.
        Laryngoscope. 2008; 118: 1265-1269
        • Tschan C.A.
        • Nie M.
        • Schwandt E.
        • Oertel J.
        Safety and efficacy of microporous polysaccharide hemospheres in neurosurgery.
        Neurosurgery. 2011; 69: ons49-ons63
        • Wang Y.
        • Xu M.
        • Dong H.
        • et al.
        Effects of PerClot(R) on the healing of full-thickness skin wounds in rats.
        Acta Histochem. 2012; 114: 311-317
        • Nilsson G.J.
        Characterization of Crystallinity in Starch Hemostatic Products.
        Chalmers Annual Materials Science Initiative Seminar, Gothenburg, Sweden2013
        • Marques A.P.
        • Reis R.L.
        • Hunt J.A.
        An in vivo study of the host response to starch-based polymers and composites subcutaneously implanted in rats.
        Macromolecules Biosci. 2005; 5: 775-785
        • Santos T.C.
        • Marques A.P.
        • Höring B.
        • et al.
        In vivo short-term and long-term host reaction to starch-based scaffolds.
        Acta Biomater. 2010; 6: 4314-4326
        • Klopfleisch R.
        • Jung F.
        The pathology of the foreign body reaction against biomaterials.
        J Biomed Mater Res A. 2017; 105: 927-940
        • Capella-Monsonís H.
        • Kearns S.
        • Kelly J.
        • Zeugolis D.I.
        Battling adhesions: from understanding to prevention.
        BMC Biomed Eng. 2019; 1: 5
        • Edlich R.F.
        • Long W.B.
        • Gubler D.K.
        • et al.
        Dangers of cornstarch powder on medical gloves: seeking a solution.
        Ann Plast Surg. 2009; 63: 111-115
        • Dearth C.L.
        • Slivka P.F.
        • Stewart S.A.
        • et al.
        Inhibition of COX1/2 alters the host response and reduces ECM scaffold mediated constructive tissue remodeling in a rodent model of skeletal muscle injury.
        Acta Biomater. 2016; 31: 50-60
        • Huleihel L.
        • Dziki J.L.
        • Bartolacci J.G.
        • et al.
        Macrophage phenotype in response to ECM bioscaffolds.
        Semin Immunol. 2017; 29: 2-13
        • Zheng C.
        • Bai Q.
        • Wu W.
        • et al.
        Study on hemostatic effect and mechanism of starch-based nano-microporous particles.
        Int J Biol Macromol. 2021; 179: 507-518
        • Torres F.G.
        • Troncoso O.P.
        • Pisani A.
        • Gatto F.
        • Bardi G.
        Natural polysaccharide nanomaterials: an overview of their immunological properties.
        Int J Mol Sci. 2019; 20: 5092
        • Dhital S.
        • Warren F.J.
        • Butterworth P.J.
        • Ellis P.R.
        • Gidley M.J.
        Mechanisms of starch digestion by alpha-amylase-Structural basis for kinetic properties.
        Crit Rev Food Sci Nutr. 2017; 57: 875-892
        • Torres F.G.
        • Troncoso O.P.
        • Grande C.G.
        • Díaz D.A.
        Biocompatibilty of starch-based films from starch of Andean crops for biomedical applications.
        Mater Sci Eng C. 2011; 31: 1737-1740
        • Gatto F.
        • Troncoso O.P.
        • Brunetti V.
        • et al.
        Human monocyte response to Andean-native starch nanoparticles.
        Starch Stärke. 2016; 68: 1016-1023
        • Dos Anjos Cassado A.
        F4/80 as a major macrophage marker: the case of the peritoneum and spleen.
        in: Kloc M. Macrophages: Origin, Functions and Biointervention. Springer International Publishing, Cham2017: 161-179
        • Marques L.
        • Holgado L.A.
        • Simões R.D.
        • et al.
        Subcutaneous tissue reaction and cytotoxicity of polyvinylidene fluoride and polyvinylidene fluoride-trifluoroethylene blends associated with natural polymers.
        J Biomed Mater Res B: Appl Biomater. 2013; 101: 1284-1293
        • Velasquez D.
        • Pavon-Djavid G.
        • Chaunier L.
        • Meddahi-Pellé A.
        • Lourdin D.
        Effect of crystallinity and plasticizer on mechanical properties and tissue integration of starch-based materials from two botanical origins.
        Carbohydr Polym. 2015; 124: 180-187
        • Sadtler K.
        • Sommerfeld S.D.
        • Wolf M.T.
        • et al.
        Proteomic composition and immunomodulatory properties of urinary bladder matrix scaffolds in homeostasis and injury.
        Semin Immunol. 2017; 29: 14-23
        • Badylak S.F.
        • Gilbert T.W.
        Immune response to biologic scaffold materials.
        Semin Immunol. 2008; 20: 109-116
        • Trochsler M.
        • Maddern G.J.
        Adhesion barriers for abdominal surgery: a sticky problem.
        Lancet. 2014; 383: 8-10
        • Rao K.
        • Gomati A.
        • Yuen Hao Tong E.
        • W Ah-See K.
        • Shakeel M.
        Use of PerClot® in head and neck surgery: a Scottish centre experience.
        Eur Arch Otorhinolaryngol. 2021; 278: 1965-1969
        • Tscholl V.
        • Spann F.
        • Moses J.
        • et al.
        Prospective randomized study evaluating the effects of PerClot® (Polysaccharide Hemostatic System) application in patients with high bleeding risk undergoing cardiac rhythm device implantation.
        Int J Cardiol. 2017; 248: 84-91
        • Antisdel J.L.
        • Matijasec J.L.
        • Ting J.Y.
        • Sindwani R.
        Microporous polysaccharide hemospheres do not increase synechiae after sinus surgery: randomized controlled study.
        Am J Rhinol Allergy. 2011; 25: 268-271
        • Xu J.
        • Nie N.
        • Wu B.
        • et al.
        The personalized application of biomaterials based on age and sexuality specific immune responses.
        Biomaterials. 2021; 278: 121177