- Email: [email protected]
Hemostatic Effectiveness of Fibrin Pad After Partial Nephrectomy in Swine
Journal of Surgical Research 167, e291–e298 (2011) doi:10.1016/j.jss.2010.01.022
Hemostatic Effectiveness of Fibrin Pad After Partial Nephrectomy in Swine Richard W. Hutchinson, D.V.M., Ph.D.,*,1 Duan Broughton, D.V.M.,* Thomas A. Barbolt, Ph.D.,*,** Thomas Poandl, M.S.,* Tim Muench, D.V.M., Ph.D.,* Rick Rockar, V.M.D.,* Mark Johnson, M.S.,† and James Hart, M.D.* *Ethicon, Inc., a Johnson and Johnson Company, Somerville, New Jersey; and †MPI Research, Mattawan, Michigan
Submitted for publication September 10, 2009
Background. Current management of severe surgical or traumatic bleeding is often achieved by manual tamponade or occlusion using devices such as tourniquets or ligatures. There are some clinical scenarios where these options are either marginally effective or impractical. The present study evaluates a new combination device (Fibrin pad) consisting of biologically active components (human thrombin and fibrinogen) delivered to the targeted site by an absorbable synthetic matrix (oxidized regenerated cellulose and polyglactin 910) in a swine severe bleeding model. In this model, severe bleeding can be managed by concurrent use of several currently available treatments, or a more convenient option that offers performance and safety advantages. Materials and Methods. Partial nephrectomies were performed on swine and treated with either Fibrin pad (FP) or conventional therapy (CTR)-temporary occlusion of renal artery, electrocautery, SURGIFLO, EVITHROM, SURGICEL NU-KNIT, and PDS II suture). After intraoperative hemostasis was confirmed, the animals were closed and recovered, then survived for 2, 14, or 56 d. Results. Hemostasis was achieved at surgery and maintained in all FP and CTR treated animals. FP was as effective as CTR at establishing durable hemostasis. Treatment with FP did not require temporary occlusion of the renal artery and decreased the total treatment time by half. No animals in either group had complications related to postoperative bleeding at any time during the study. There was no evidence of pulmonary thrombi or evidence of thrombotic complications. No biologically significant adverse local **
Current Address: TAB Consulting, Cambridge, New York To whom correspondence and reprint requests should be addressed at Ethicon, Inc., a Johnson and Johnson Company, P.O. Box 151,Somerville, NJ 08876-0151. E-mail: [email protected]. 1
tissue response was present in association with the Fibrin pad at any study interval, and no biologically relevant or consistent changes in blood parameters were identified. Conclusions. Fibrin pad was as effective as CTR for the primary management of severe bleeding without occlusion of the renal artery and a shorter surgical time. No evidence of a systemic or local adverse response was identified due to exposure to the Fibrin pad. Ó 2011 Elsevier Inc. All rights reserved. Key Words: severe bleeding; animal model; partial nephrectomy; EVITHROM; SURGIFLO; SURGICEL.
INTRODUCTION
The delivery of fibrin glue components using a backing pad has been described using either a horse-derived backing [1], VICRYL* Mesh backing [2], and with Oxidized Regenerated Cellulose [3]. These products have been shown to be capable of maintaining shortterm hemostasis in severe aortic injuries [4] and renal models of severe bleeding [5]. These products have also been shown to have no excess risk of thrombogenicity [6, 7]. Fibrin Pad (FP) described in this report has been previously shown to be effective in achieving hemostasis in severe parenchymal trauma models in severe coagulopathic states [8]. The objective of this study was to compare the effectiveness of a new combination product (FP) to conventional therapy of sequential use of temporary renal artery occlusion, electrocautery, SURGIFLO þ EVITHROM, SURGICEL NU-KNIT, and PDS II suture for the primary management of severe bleeding (defined as more intense pulsatile and/or rapidly
e291
0022-4804/$36.00 Ó 2011 Elsevier Inc. All rights reserved.
e292
JOURNAL OF SURGICAL RESEARCH: VOL. 167, NO. 2, MAY 15, 2011
flowing bleeding that could come from one or more artery or vein). The animal model has been used for similar evaluations of hemostats [5], and because the kidney is one of the most perfused organs, it is considered a worst-case model of parenchymal bleeding. Successfully demonstrating durable hemostasis in a model of this severity is intended to support a variety of other clinical applications. The conventional therapy described above includes both active and inactive hemostatic modalities that induce clotting, and physically stop bleeding. Similarly, the test article also contains pharmaceutical and device components that contribute to hemostais by both active and passive mechanisms, however it can be applied without clamping the renal artery. Because the two options rely on similar dual mechanisms, the study objectives are logical. MATERIALS AND METHODS Test Article FP is an absorbable combination product consisting of a flexible, easy to conform, matrix consisting of layer of oxidized regenerated cellulose (ORC: similar to SURGICEL topical absorbable hemostat and INTERCEED adhesion prevention barrier) and polyglactin 910 fibers (similar to VICRYL suture and mesh products) with a layer of lyophilized human plasma-derived fibrinogen and thrombin (similar to EVICEL Fibrin sealant). Fibrin pad is intended to be used when standard surgical methods of hemostasis are impractical or ineffective.
Test System Thirty-six healthy female pigs, approximately 12–13 wk of age (weighing between 34 and 56 kg), were obtained from USDA approved vendors and underwent a partial nephrectomy of the caudal pole of the right kidney. The swine model was used based on its similarity to human coagulation biochemistry and gross renal anatomy [9]. This established swine model has been used for similar evaluations of hemostatic agents [5].
pre-weighed laparotomy sponges were placed in the abdominal cavity surrounding the right kidney. For treatment groups 1 and 2, the caudal pole of the right kidney was resected at a 1.0 6 0.5 cm depth, creating an approximate 2.5 6 1 cm diameter resection surface using a scalpel. The nephrectomy was intended to include the full depth of the renal cortex without intrusion into the renal pelvis. As such, several arcuate arteries are typically involved. An approximately 2 3 3 in. section of Fibrin pad was applied directly to the resection surface with the ORC backing up. Manual tamponade was applied for 3 min using pre-weighed laparotomy sponges. The time from amputating the kidney to hemostasis was recorded (including the 3 min for the manual tamponade after article application). Improperly applied articles were removed and another application of Fibrin pad was applied. All visible blood was collected, and the pre-weighed laparotomy sponges were removed and re-weighed. Five of 12 animals required reapplication in group 1; also, in group 2, 5 of 12 animals required reapplication. In treatment group 3, the renal artery was temporarily occluded by the use of atraumatic vascular clamps. The caudal pole was resected as previously described. Electrocautery was utilized to stop bleeding from larger vessels. The resection surface was dried as much as possible using pre-weighed laparotomy sponges. SURGIFLO and EVITHROM were then applied to the cut surface with moist gauze sponges. Manual tamponade was applied for approximately 3 min. A double layer of 2 3 3 in. SURGICEL NU-KNIT was then applied and sutured in place with overlying PDS bolstering sutures fixed to the renal capsule. The vascular clamp was released and then the resection surface was evaluated for bleeding. If necessary, an additional 3 min compression was applied to achieve hemostasis. The time from clamping the renal artery to release of the clamp and the observation of no active bleeding was considered the time to hemostasis. All visible blood was collected and the pre-weighed laparotomy sponges were removed and re-weighed as a measure of blood loss. For all animals, the mean arterial pressure (MAP) was measured via indirect blood pressure cuff and maintained within 620 mm Hg of the pre-nephrectomy MAP value by intravenous infusion of lactated Ringer’s solution. The capsule surrounding the right kidney, primary abdominal wall, fascia, subcutaneous layers, and skin were closed with appropriately sized suture and suture patterns. Following surgery, the animals were closely monitored during anesthetic recovery for physiologic disturbances, including cardiovascular/respiratory depression, hypothermia, and hypovolemia.
Clinical Pathology Study Design Treatment group 1 consisted of 12 pigs with the Fibrin pad applied to the resection surface and a 48 (62) h study interval. Treatment group 2 consisted of 12 pigs treated with Fibrin pad, and the animals were evaluated at 10 and 28 d with MRI, and necropsy and histopathology of selected tissues at 14 (62) or 56 (62) d post-surgery (6 animals at each time). The resection surface of treatment group 3 was treated with temporary renal artery occlusion, electrocautery, SURGIFLO, EVITHROM, SURGICEL NU-KNIT, and PDS II, and was evaluated similar to treatment group 2.
Surgical Model Following induction of anesthesia, a ventral midline incision was made from the xyphoid to just cranial of the pubis. Routine clamping and electrocautery was used to control cutaneous hemorrhage. The abdominal organs were inspected for evidence of past or present pathology. Because of the potential for the porcine spleen to sequester a variable red cell mass during hemorrhage, a splenectomy was performed. The spleen was weighed and an intravenous bolus injection of warm lactated Ringer’s solution at three times the weight of the explanted spleen was administered. Following the splenectomy,
For treatment group 1, standard hematology evaluations and a coagulation profile (prothrombin time (PT), activated partial thromboplastin time (aPTT), and fibrinogen) were conducted on all animals prior to surgery and at 1 and 2 d postoperatively. For treatment groups 2 and 3, a coagulation profile (PT and aPTT), and standard hematology, blood serum chemistry and urinalysis evaluations were conducted on all animals prior to surgery and at 10 6 2 and 28 6 2 d postoperatively, and prior to necropsy (14 or 56 d post-surgery). All blood samples were collected from the jugular vein or other suitable vein. The samples were collected into tubes containing EDTA for evaluation of hematology parameters, citrate for evaluation of coagulation parameters, and clot tubes for serum biochemistry samples. Urine samples were collected via cystocentesis.
MRI Evaluations The animals from treatment groups 2 and 3 were anesthetized and MRI evaluations of the resection sites (Northern Biomedical Research, Muskegon, MI) were conducted pre-surgery, and at 10 and 28 d. An independent radiologist evaluated the resection surface images and measured fluid accumulations when present.
HUTCHINSON ET AL.: FIBRIN PAD EFFECTIVENESS AFTER PARTIAL NEPHRECTOMY
Scheduled Euthanasia Treatment group 1 animals were euthanized at 48 6 2 h. Necropsies were conducted and the thoracic and abdominal vasculature was examined for thrombosis. A section from each lung lobe and the resection site were collected for microscopic evaluation. Animals from treatment groups 2 and 3 were divided into two. Half of the animals from each treatment group were necropsied at 14 (62) or 56 (62) d post-surgery. The lung (three sections from each lobe), liver, right and left kidneys, heart, ureters, bladder, pancreas, and adrenals were collected for microscopic evaluation. In addition, the arterial and venous vessels were traced to the abdominal aorta and vena cava, respectively, and examined for thrombosis. All harvested tissues were fixed in 10% neutral buffered formalin (NBF) for at least 48 h. Tissues were embedded, sectioned, mounted, and stained with hematoxylin and eosin according to standard procedures.
Statistical Comparisons Total treatment times and blood loss values for groups 2 and 3 were compared because these two groups were conducted together with the same surgical team in a randomized order. Group 1 was conducted at a later time, and although the same procedure was used, the surgical staff available at the time was different. Because the hemostasis time and blood loss evaluation is somewhat dependant on the surgeon, the inclusion of group 1 in the statistical comparison would add an additional variable that is not relevant when only groups 2 and 3 are compared. Groups 2 and 3 were compared using a t-test (Graphpad Prism ver. 5.02; GraphPad Software, San Diego, CA). When comparing peri-renal fluid, body weight, hematology (except leukocyte count), coagulation parameters, clinical chemistry, and urinalysis values, Levene’s was used to assess homogeneity of group variances for each specified endpoint and for all collection intervals. If Levene’s test was not significant (P > 0.01), a pooled estimate of the variance (mean square error or MSE) was computed from a one-way analysis of variance (ANOVA) and utilized by a Dunnett’s comparison of each treatment group with the CTR group. If Levene’s test was significant (P < 0.01), comparisons with the CTR group were made using Welch’s t-test with a Bonferroni correction. Historic data indicate that total and differential leukocyte counts are not normally distributed; therefore a log transformation was performed and then group pair-wise comparisons were conducted using Welch’s t-test with a Bonferroni correction (SAS ver. 8.2; SAS Institute, Cary, NC).
e293
animals gained weight over the study period for treatment groups 2 and 3. Intraoperative Hemostasis
Assessment of intraoperative bleeding demonstrated that FP was as effective as the CTR that included renal artery occlusion at establishing hemostasis. Twelve animals in the FP group (group 2) did not require occlusion of the renal artery, and hemostasis was obtained with between one and three applications of FP with an average manual tamponade time of 5.9 6 2.6 min, and was statistically significant between the Fibrin pad and CTR treatment groups (P ¼ 0.0013; Fig. 1). Similar results were repeated with another group (group 1) of 12 animals with one to two applications of FP and an average manual tamponade time of 7.1 6 4.2 min. The CTR group (group 3) required occlusion of the renal artery for an average of 10.8 6 2.6 min. Groups 2 and 3 were conducted in random order by the same surgical team, and group 1 was conducted by a different surgical team, which may account for slightly different hemostasis times for groups 1 and 2. A similar amount of blood loss was found in each treatment group with an average of 176.7 6 80.9 mL blood loss in animals receiving Fibrin pad (no renal artery occlusion) and an average of 139.9 6 63.0 mL of blood loss in animals receiving the conventional therapy (with renal artery occlusion). The difference in blood loss was not statistically significant (P ¼ 0.2272), and
Animal Welfare Compliance The in vivo portions of this study were conducted in facilities accredited by the Association for the Accreditation and Assessment of Laboratory Animal Care International (AAALAC). The guide for the care and use of laboratory animals was followed [10].
RESULTS Size of Partial Nephrectomy
The portion of kidney removed was on average 3.0 (60.19) cm long, 1.6 (60.05) cm wide, 1.1 (60.08) cm deep, and weighed 1.92 (60.51) g. Clinical Finidngs and Body Weights
Clinical findings were consistent for the age of the animals as well as for animals that had undergone a recent significant abdominal surgical procedure. All
FIG. 1. Box and whiskers plot comparing conventional treatment to FP treatment. The total treatment times for both groups are compared, the two groups were statistically different (P ¼ 0.0013). For the FP group, treatment time is measured from the time that the amputation occurs until the time that manual tamponade is completed and hemostasis is confirmed. For the CTR group, the time is begun at renal hilus clamping (immediately prior to renal tissue amputation) until hemostasis is confirmed after removal of the renal clamp. This time therefore reflects the renal ischemia time avoided by the use of FP (n ¼ 12/group). In this plot, the box indicates the 25th and 75th percentiles (with the median), and the whiskers indicate the minimum and maximum values.
e294
JOURNAL OF SURGICAL RESEARCH: VOL. 167, NO. 2, MAY 15, 2011
biologically trivial when one considers the potential hemorrhage from a partial nephrectomy wound. MRI Evaluations
MRI evaluations on d 10 and 28 demonstrated increased edema and significantly increased perirenal fluid in the FP group compared with the CTR group at d 10 (P < 0.001), but no significant difference at d 28 (P > 0.05). Edema and perirenal fluid accumulations were not present at necropsy for the animals necropsied at 14 or 56 d, so the significance of the MRI findings are unclear. The average area of fluid in the perirenal space was found to decrease over time in both treatment groups. At 10 d, the average area in the perirenal space for the FP group was 5.45 6 1.40 cm2 and for the CTR group it was 1.63 6 1.07 cm2. At 28 d, the average area in the perirenal space for the FP group was decreased to 1.98 6 1.05 cm2, and for the CTR group it was 0.22 6 0.34 cm2 (Fig. 2). Fluid accumulations were interpreted as transudates with blood products by MRI; however, necropsy findings and microscopic evaluation in the animals necropsied at 14 d do not correlate to the MRI interpretation (i.e.; no evidence of fluid accumulation with blood components). The necropsy findings and microscopic evaluation from treatment group 1 animals (48 6 2 h post-implantation) suggest that the edema fluid accumulation was within the adjacent surgical traumatized tissues, and when perirenal fluid was present, it was consistent with an ultrafiltrate of blood rather than a transudate with blood components.
FIG. 2. Accumulation of perirenal fluid. Calculated volumes of perirenal fluid accumulation as observed during MRI evaluations at 10 and 28 d. At 10 d the two treatments were statistically different (P < 0.001); at 28 d there was no significant difference (P > 0.05). The volume of fluid decreased in both groups, but was delayed in the FP group (compared with CTR). The mean and standard error (SEM) are plotted for each group.
Intraparenchymal edema fluid and hemorrhage was identified in the MRI evaluation. This change was not measured due to the inability to reliably define the edges of the alteration. The intraparenchymal changes were present in 10 of 11 Fibrin pad animals at 10 d postsurgery, and in 3 of 5 animals at 28 d post-implantation. Similar alterations were present in the CTR group for 11 of 11 animals at 10 d post-implantation, and 6 of 6 animals at 28 d post-implantation. Intraparenchymal hemorrhage, occasional clots in small vessels damaged by the resection procedure and necrosis was present in group 1 animals at 48 h and corresponded to the MRI reported alterations for treatment group 2 and 3 animals at 10 d. Macroscopically and microscopically the intraparenchymal areas were centrally located in the resection area and were cone-shaped but did not have a vascular orientation. The areas most likely represent damage due to blood from the resection surface forced along a path of least resistance (center of resection surface close to renal pelvis). Clinical Pathology
Hematology parameters were within the normal range for swine for treatment groups 2 and 3 (Table 1). A few parameters were noted to be statistically different between treatment groups 2 and 3; however, since the values were within the normal range for swine these differences were not considered biologically relevant and were believed to represent normal biological variation. No consistent hematology changes were observed in treatment group 1 (taken pre-surgery, 1 and 2 d post-surgery) except increased white blood cell counts 1 day post-implantation (data not shown). This change reflects a response to the surgical procedure and is not considered treatment related. Mean values for the clinical chemistries were within the normal range except for alkaline phosphatase, which was increased for treatment groups 2 and 3 at all time points (Table 2). The total protein was low at all time points for both treatment groups 2 and 3. Statistically different mean values were detected occasionally between treatment groups 2 and 3; however, the values were still within the normal range and the differences were not considered biologically significant. Prothrombin time (PT) and activated partial thromboplastin time (APTT) for all animals were within the expected range and were not statistically different between the treatment groups (Table 3). The urinalysis parameters were considered normal for treatment groups 2 and 3 (Table 4). Necropsy and Histopathology
At 2, 14, and 56 d post-implantation there was no evidence of biologically significant post-surgical
e295
HUTCHINSON ET AL.: FIBRIN PAD EFFECTIVENESS AFTER PARTIAL NEPHRECTOMY
TABLE 1 Selected Hematology Results (Mean ± SD). Results Significantly Different Than Control (P < 0.05) Are Marked with *
Leukocytes (103/uL) Erythrocytes (106/uL) Hematocrit (%) Platelets (103/uL) Absolute Reticulocytes (103/uL)
Control FP Control FP Control FP Control FP Control FP
Pretest
Day 10
Day 14
Day 28
Day 56
18.88 (4.689) 19.58 (4.661) 6.623 (0.597) 6.436 (0.7529) 34.17 (3.775) 33.71 (5.402) 467.2 (92.85) 425.6 (127.01) 57.19 (28.918) 50.28 (38.725)
19.48 (5.347) 22.33 (4.62) 6.489 (0.455) 6.358 (0.7151) 32.25 (1.659) 32.91 (3.767) 677.3 (242.45) 654.8 (101.34) 68.62 (17.887) 102.24 (46.107)*
18.95 (5.256) 22.33 (4.036) 6.617 (0.3625) 6.313 (0.4439) 32.97 (1.703) 31.43 (1.247) 504.3 (54.35) 604.2 (149.82) 88.82 (42.335) 61.53 (33.548)
19.73 (6.066) 22.7 (4.481) 6.483 (0.702) 6.512 (0.4527) 31.5 (2.505) 33.33 (2.343) 574.2 (208.43) 475.7 (43.81) 65.82 (17.633) 76.43 (32.829)
20.02 (4.386) 18.7 (2.433) 6.715 (0.3769) 6.44 (0.3301) 33.53 (1.877) 32.8 (1.219) 441.5 (81.97) 473.4 (52.3) 73.2 (38.516) 63.14 (22.082)
hemorrhage in any of the animals examined. No thrombi were identified in the major vasculature of the thoracic or abdominal cavities or in examined sections of lung for any of the animals evaluated. At necropsy, the expected findings indicative of abdominal surgery in the pig were present, including adhesion formation and variable fluid accumulations. There were no biologically relevant differences in adhesion formation between the Fibrin pad and the CTR treatment groups, or any evidence of a macroscopic adverse response to the treatment received. In treatment group 1, thin hematomas were occasionally located between the Fibrin pad and the resection surface. The hematomas merged with the resection surface and the Fibrin pad and were not considered clinically relevant.
At 48 h post-implantation (treatment group 1), occasional clots involving small blood vessels and glomeruli damaged by the resection were observed along the resection surface. These clots were considered normal and necessary for the cessation of hemorrhage from a resection surface. No other thrombi were present within the sections of kidney examined. Intrarenal hemorrhage, necrosis, and degeneration were present, and were the result of mechanical trauma associated with the surgical and implant procedure. Inflammation associated with the Fibrin pad was limited at 48 h. At 2 weeks, FP group (group 2) had decreased severity of inflammation (3 of 6 moderate, 3 of 6 severe), fibrosis/fibroplasia (2 of 6 moderate, 4 of 6 severe), and tubular degeneration/regeneration (1 minimal, 1 mild,
TABLE 2 Selected Clinical Chemistry Results (Mean ± SD). Results Significantly Different Than Control (P < 0.05) are marked with *
Chloride (mEq/L) Alkaline Phosphatase (U/L) GGT (U/L) AST (U/L) ALT (U/L) Sorbitol Dehydrogenase (U/L) Urea Nitrogen (mg/dL) Creatinine (mg/dL) Total Protein (g/dL) Albumin (g/dL) Globulin (g/dL)
Control FP Control FP Control FP Control FP Control FP Control FP Control FP Control FP Control FP Control FP Control FP
Pretest
Day 10
Day 14
Day 28
Day 56
102.7 (1.3) 103.4 (2.78) 114.6 (18.61) 122 (54.94) 33.4 (13.23) 36.4 (12.49) 33.3 (11.69) 41.8 (17.4) 36.2 (9.93) 39.2 (9.56) 5.43 (6.278) 3.14 (2.015) 9.1 (2.78) 8.3 (2.19) 1.11 (0.116) 1.13 (0.13) 6.61 (0.425) 6.56 (0.54) 3.52 (0.266) 3.39 (0.348) 3.09 (0.355) 3.17 (0.623)
100.9 (2.35) 102.4 (2.11) 94.5 (29.64) 100.5 (36.2) 32.7 (7.1) 36.5 (12.46) 29.5 (10.94) 28.2 (7.9) 37.9 (11.15) 38.9 (8.28) 2.03 (1.709) 3.48 (2.15) 8.8 (2.09) 8.5 (3.29) 1.03 (0.15) 1.09 (0.261) 6.28 (0.409) 6.37 (0.464) 3.29 (0.287) 3.27 (0.227) 2.98 (0.484) 3.1 (0.447)
100.7 (3.01) 102.7 (2.8) 101.8 (36.92) 84.2 (17.93) 32.8 (9.2) 31.8 (6.74) 37.3 (21.2) 23.2 (3.87) 39.5 (5.21) 38.8 (7.17) 2.27 (2.185) 2.83 (2.408) 7 (2.37) 9.2 (5.49) 1.07 (0.103) 1.2 (0.126) 6.45 (0.409) 6.32 (0.417) 3.28 (0.343) 3.35 (0.281) 3.17 (0.717) 2.97 (0.501)
100.7 (3.33) 101.2 (1.72) 96 (17.46) 112.7 (22.84) 30.3 (7.74) 36.3 (10.76) 28.3 (8.26) 26.2 (5.78) 39.2 (10.07) 41.2 (5.98) 1.03 (0.361) 1.42 (1.137) 9.7 (3.78) 8.2 (1.94) 1.28 (0.133) 1.2 (0.167) 6.55 (0.476) 6.73 (0.505) 3.5 (0.283) 3.47 (0.333) 3.05 (0.468) 3.27 (0.565)
101.5 (1.97) 103.7 (1.03)* 89 (16.27) 100.3 (16.99) 33.5 (10.25) 37.3 (8.57) 23.5 (4.51) 23.8 (7.05) 40.7 (9.48) 40.8 (4.36) 1.9 (1.066) 2.4 (0.663) 11.7 (4.68) 13.7 (3.08) 1.47 (0.175) 1.4 (0.21) 6.57 (0.393) 6.5 (0.329) 3.62 (0.286) 3.72 (0.172) 2.95 (0.575) 2.78 (0.397)
e296
JOURNAL OF SURGICAL RESEARCH: VOL. 167, NO. 2, MAY 15, 2011
TABLE 3 Coagulation Profile (Mean ± SD)
APTT (sec) Prothrombin Time (sec) Fibrinogenx
Control FP Control FP FP
Pretest
Day 1
Day 2
Day 10
Day 14
Day 28
Day 56
13.83 6 3.7* 13.63 6 2.0y 11.62 6 0.5* 14.12 6 0.7z 166.4 6 46.64
NC 16.9 6 3.1 NC 14.98 6 0.9 375.3 6 63.6
NC 14.43 6 2.8 NC 14.97 6 1.5 311.3 6 82.96
15.30 6 4.7 11.52 6 6.3 11.49 6 0.5 11.30 6 0.8 NC
15.44 6 1.4 15.78 6 1.0 11.26 6 0.5 11.82 6 0.5 NC
12.78 6 4.9 15.22 6 1.54 11.28 6 0.5 11.44 6 0.8 NC
13.08 6 5.8 15.58 6 1.18 11.44 6 0.6 11.56 6 0.3 NC
NC¼ Not Collected. Pretest value for groups 2 and 3, data not collected for group 1. y Pretest results for group 1 are presented, pretest APTT for groups 2 and 3 was 13.0865.1 sec. z Pretest results for group 1 are presented, pretest PT for groups 2 and 3 was 11.3560.6 sec. x Test only conducted for group 1. *
1 moderate, and 2 severe), and increased neovascularization (4 mild and 2 moderate) at d 14 compared with the CTR group (group 3). This may reflect the differences in the surgical manipulation and article application procedures between the CTR group versus the Fibrin pad group. Over the course of the study, the character of the inflammation seen in FP treated animals was observed to change from a primarily involving macrophages (6 of 6) response at d 14 to a more mature response involving lymphocytes, plasma cells, and macrophages at d 56. These changes reflect normal healing response and are not considered adverse effects. Some of the noted inflammation (i.e., the lymphoplasmacytic aggregates) associated with the Fibrin pad is suggestive of an immune response to the foreign (human) proteins in the Fibrin pad. The results of this study support the safety and biocompatibility of FP (Table 5). FP was considered essentially absorbed because less than 10% of the FP material remained at 56 d (data not shown). The material remaining represented the matrix (VICRYL) component of the Fibrin pad. DISCUSSION
Considering the morbidity, mortality, and extended hospitalization due to uncontrolled severe traumatic or surgical bleeding, the timely control of bleeding has obvious benefits for patients and for surgeons, and is an active area of research [11-14]. Surgeons encounter
bleeding from a variety of tissue types, intensity, and etiologies. The selection of appropriate methods or products to control bleeding is dependent upon many factors, which include but are not limited to: 1. Bleeding severity 2. Anatomic location of the source and the proximity of adjacent critical structures 3. Whether the bleeding is from a discrete source or from a broader surface area 4. Visibility and precise identification of the source 5. Access to the source 6. Tissue type and fragility/friability 7. Coagulation system status 8. Patient stability Bleeding of mild, moderate, or even severe intensity can frequently be quickly and successfully treated with traditional mechanical methods (i.e., suture, ligature, clips, and staples), or with various energy-based coagulation systems, (i.e., electrocautery, argon beam lasers, and ultrasound) [15, 16]. When these conventional measures are ineffective or impractical, adjunctive hemostasis techniques and products are typically utilized. These include topical absorbable hemostats (TAH), such as oxidized regenerated cellulose, gelatin, or collagen [17] and active hemostats, such as topical thrombin or fibrin sealants [18, 19]. These products have proven most efficacious in controlling slowly bleeding foci, diffuse oozing, bleeding from needle puncture sites, and diffuse
TABLE 4 Urinalysis (Mean ± SD)
Volume (ml) Specific Gravity pH
Control FP Control FP Control FP
Pretest
Day 10
Day 14
Day 28
Day 56
9.50 (1.7) 10 (0) 1.0100 (0.004) 1.0103 (0.004) 6.33 (0.86) 6.54 (1.1)
9.17 (1.9) 9.58 (1.4) 1.0104 (0.004) 1.0116 (0.004) 7.17 (0.72) 6.88 (0.9)
10 (0) 10 (0) 1.0125 (0.007) 1.0198 (0.009) 6.42 (0.74) 6.25 (1.3)
10 (0) 10 (0) 1.0053 (0.004) 1.0085 (0.004) 7.33 (0.75) 6.83 (0.7)
10 (0) 10 (0) 1.0082 (0.009) 1.0120 (0.012) 6.75 (0.69) 6.83 (0.4)
y
*
Not treatment related (attributed to major abdominal surgery). Within acceptable level as represents part of normal healing. z Defined as presence of macrophages. x Mechanically induced due to blood flow from resection surface along the path of least resistance (central resection site area closest to the renal pelvis).
N/A N/A N/A 7/12 minimal to mild
N/A
N/A 6/6 minimal to mild 0/6 6/6 minimal to mild 4/6y minimal to mild 1/6x Mild 0/6 6/6x minimal to mild 6/6z moderate to severe N/A N/A N/A 6/6y mild to moderate 6/6x minimal to mild 1/6x mild 5/6x mild to severe 6/6z severe N/A N/A N/A 5/6y mild 4/6x mild to moderate 3/6x minimal to moderate 6/6x moderate to severe N/A N/A N/A N/A N/A 4/12x minimal to mild 4/12x minimal to mild 5/12x minimal to mild
N/A 5/6 mild to moderate 1/6z moderate 3/6 mild 6/6y minimal to mild 0/6 3/6x minimal 6/6x mild to moderate
4/6* Mild 6/6y minimal to mild N/A 6/6* minimal to moderate 6/6y mild to moderate N/A 6/6* mild to severe 6/6y moderate to severe N/A 6/6* moderate to severe 6/6y severe N/A Miscellaneous finding * 12/12 y minimal 12/12 minimal
Adhesion Fibrosis/Fibroplasia Infiltrate of low numbers of macrophages and to lesser extent neutrophils Inflammation, granulomatous Inflammation, chronic Inflammation, chronic-active Foamy/pigmented macrophages Neovascularization Necrosis (Intrarenal) Hemorrhage (Intrarenal) Tubular degeneration/ regeneration (Intrarenal) Resection surface completely or partially covered by hematoma
Day 56 conventional Day 14 Fibrin pad Day 14 conventional Day 2 Fibrin pad Microscopic changes
Microscopic Changes in Right Kidney (Tissue in Contact with Implant)
TABLE 5
Day 56 Fibrin pad
HUTCHINSON ET AL.: FIBRIN PAD EFFECTIVENESS AFTER PARTIAL NEPHRECTOMY
e297
parenchymal organ hemorrhage. They are not as effective in the control of some forms of active or severe bleeding where the topically applied agents can be simply carried away from the bleeding site by actively flowing blood or require a dry field for application [11, 12]. Timely control of bleeding has obvious benefits for patients and for surgeons. Minimization of blood loss is important to help avoid hypovolemia, hemodynamic and metabolic deterioration, and anemia [13]. Minimization of exposure to blood component transfusion and the attendant risks has become increasingly important to patients and to surgeons [14]. Control of bleeding also facilitates surgeons’ visualization of the surgical site, enabling precise and timely conduct of the surgical procedure. The present study includes a new hemostatic product being developed to meet a current unmet surgical need, to rapidly and reliably control bleeding that is not amenable to treatment with standard methods of ligation or energy based coagulation and for which currently available products are not satisfactorily effective. Johnston, et al. (2006) [20] compared seven hemostats (CoSeal, CoStasis, Tisseel, FloSeal, Dermabond, Gelfoam/Tisseel, and Bolster/Floseal) in a porcine partial nephrectomy model where several sizes of lesion were made; the largest was an amputation of the kidney at the level of the hilus. Transient hypertension was created using dopamine, and the hemostatic agents were applied while the renal hilus was clamped. The study concluded that most of the hemostatic agents tested were effective for small target application sites, but only sutured bolsters were consistently effective for large partial nephrectomies. The present study explores the efficacy of test hemostats without clamping of the renal hilus, and allows for assessing the durability of the hemostatic repair by surviving animals until absorption of the product is complete. Dalpiaz et al. (2008) [21] reviewed hemostat usage in 15 animal models and 11 clinical trials. This review concludes that the efficacy of fibrin sealants was demonstrated in animal partial nephrectomy models and has been confirmed in clinical trials. Nine of the animal models reported required some form of hilar control. The present study demonstrates efficacy without the need for hilar control. Overall, FP was as effective as CTR in establishing and maintaining hemostasis in a swine partial nephrectomy model. The use of FP eliminated the need for occlusion of the renal artery and reduced the overall treatment time to achieve hemostasis by half, thereby demonstrating both safety and performance advantages. Intraoperative hemostasis was achieved and was maintained in all animals, and there was no evidence for loss of hemostatic effectiveness in any animal during follow-up. Clinical pathology, necropsy, and histopathologic assessments
e298
JOURNAL OF SURGICAL RESEARCH: VOL. 167, NO. 2, MAY 15, 2011
did not demonstrate the presence of any adverse response to Fibrin pad. These data demonstrate that FP is an effective primary and sole hemostatic treatment in this severely challenging bleeding model, avoids the need for renal artery occlusion, and the sequential use of cautery, TAH with Thrombin, ORC, and suture. We painstakingly looked for evidence for either a local or systemic propensity for thrombogenicity (gross, histologic, and clinical pathology) and found none. CONCLUSIONS
Successful completion of the safety and efficacy endpoints included in this model support the advancement of FP into clinical evaluation in challenging soft tissue and parenchymal bleeding. ACKNOWLEDGMENTS The authors thank the staff at MPI research, in particular, Keith Nelson, DVM, who served as the study pathologist at MPI. The authors also acknowledge Alexia McKnight, DVM, who interpreted the MRI results (University of Pennsylvania School of Veterinary Medicine, Kennett Square, Pennsylvania). The authors thank Drs. Xiao-Yu Song, M.D., Ramy Mahmoud, M.D., and Larry Johnson, DVM, Ph.D. for their generous support and helpful discussions during the preparation of this manuscript.
REFERENCES 1. Erdogan D, van Gulik TM. Evolution of fibrinogen-coated collagen patch for use as a topical hemostatic agent. J Biomed Mater Res Part B: Appl Biomater 2008;85B:272. 2. Kheirabadi BS, Acheson EM, Deguzman R, et al. The potential utility of fibrin sealant dressing in repair of vascular injury in swine. J Trauma 2007;62:94. 3. Finley DS, Lee DI, Eichel L, et al. Fibrin glue-oxidized cellulose sandwich for laproscopic wedge resection of small renal lesions. J Urolol 2005;173:1477. 4. Kheirabadi BS, Acheson EM, Deguzman R, et al. Hemostatic efficacy of two advanced dressings in an aortic hemorrhage model in swine. J Trauma 2005;59:25. 5. Cornum RL, Morey AF, Harris R, et al. Does the absorbable fibrin adhesive bandage facilitate partial nephrectomy? J Urol 2000;164:864.
6. Kheirabadi BS, Sieber J, Holcomb JB. Assessment of the thrombogenic effect of fibrin sealant dressing in a vascular surgery model in rabbits. J Invest Surg 2006;19:387. 7. Birth M, Figueras J, Bernardini S, et al. Collagen fleece-bound fibrin sealant is not associated with an increased risk of thromboembolic events or major bleeding after its use for hemostasis in surgery: A prospective multicenter surveillance study. Patient Safety Surg 2009;3:13. 8. Delgado AV, Kheirabadi BS, Fruchterman TM, et al. A novel biologic hemostatic dressing (fibrin patch) reduces blood loss and resuscitation volume and improves survival in hypothermic, coagulopathic Swine with grade V liver injury. J Trauma 2008; 64:75. 9. Pusateri AE, Holcomb JB, Kheirabadi BS, et al. Making sense of the preclinical literature on advanced hemostatic products. J Trauma 2006;60:674. 10. Institute of Laboratory Animal Research, Commission on Life Sciences, National Research Council. Guide for the Care and Use of Laboratory Animals. Washington, DC: National Academy Press, 1996. 11. de la Torre RA, Bachman SL, Wheeler AA, et al. Hemostasis and hemostatic agents in minimally invasive surgery. Surgery 2007; 142(4 Suppl):S39. 12. Msezane LP, Katz MH, Gofrit ON, et al. Hemostatic agents and instruments in laparoscopic renal surgery. J Endourol 2008; 22:403. 13. Shander A. Financial and clinical outcomes associated with surgical bleeding complications. Surgery 2007;142:S21. 14. Boucher BA, Hannon TJ. Blood management: a primer for clinicians. Pharmacotherapy 2007;27:1394. 15. Aziz O, Athanasiou T, Darzi A. Haemostasis using a ready-to-use collagen sponge coated with activated thrombin and fibrinogen. Surg Technol Int 2005;14:35. 16. Acheson EM, Kheirabadi BS, Deguzman R, et al. Comparison of hemorrhage control agents applied to lethal extremity arterial hemorrhages in swine. J Trauma 2005;59:865. 17. Palm MD, Altman JS. Topical hemostatic agents: A review. Dermatol Surg 2008;43:431. 18. Recinos GA, Inaba KA, Dubose JA, et al. Local and systemic hemostatics in trauma: A review. Ulusal Travma ve Acil Cerrahi Dergisi 2008;14:175. 19. Gaarder C, Naess PA, Frischknecht Christensen E, et al. Scandinavian guidelines: The massively bleeding patient. Scand J Surg 2008;97:15. 20. Johnston WK, Kelel KM, Hollenbeck BK, et al. Acute integrity of closure for partial nephrectomy: Comparison of 7 agents in a hypertensive porcine model. J Urol 2006;175:2307. 21. Dalpaiz O, Neururer R, Bartsch G, et al. Hemostatic sealants in nephron-sparing surgery: What surgeons need to know. BJU International 2008;102:1502.
Hemostatic Effectiveness of Fibrin Pad After Partial Nephrectomy in Swine Richard W. Hutchinson, D.V.M., Ph.D.,*,1 Duan Broughton, D.V.M.,* Thomas A. Barbolt, Ph.D.,*,** Thomas Poandl, M.S.,* Tim Muench, D.V.M., Ph.D.,* Rick Rockar, V.M.D.,* Mark Johnson, M.S.,† and James Hart, M.D.* *Ethicon, Inc., a Johnson and Johnson Company, Somerville, New Jersey; and †MPI Research, Mattawan, Michigan
Submitted for publication September 10, 2009
Background. Current management of severe surgical or traumatic bleeding is often achieved by manual tamponade or occlusion using devices such as tourniquets or ligatures. There are some clinical scenarios where these options are either marginally effective or impractical. The present study evaluates a new combination device (Fibrin pad) consisting of biologically active components (human thrombin and fibrinogen) delivered to the targeted site by an absorbable synthetic matrix (oxidized regenerated cellulose and polyglactin 910) in a swine severe bleeding model. In this model, severe bleeding can be managed by concurrent use of several currently available treatments, or a more convenient option that offers performance and safety advantages. Materials and Methods. Partial nephrectomies were performed on swine and treated with either Fibrin pad (FP) or conventional therapy (CTR)-temporary occlusion of renal artery, electrocautery, SURGIFLO, EVITHROM, SURGICEL NU-KNIT, and PDS II suture). After intraoperative hemostasis was confirmed, the animals were closed and recovered, then survived for 2, 14, or 56 d. Results. Hemostasis was achieved at surgery and maintained in all FP and CTR treated animals. FP was as effective as CTR at establishing durable hemostasis. Treatment with FP did not require temporary occlusion of the renal artery and decreased the total treatment time by half. No animals in either group had complications related to postoperative bleeding at any time during the study. There was no evidence of pulmonary thrombi or evidence of thrombotic complications. No biologically significant adverse local **
Current Address: TAB Consulting, Cambridge, New York To whom correspondence and reprint requests should be addressed at Ethicon, Inc., a Johnson and Johnson Company, P.O. Box 151,Somerville, NJ 08876-0151. E-mail: [email protected]. 1
tissue response was present in association with the Fibrin pad at any study interval, and no biologically relevant or consistent changes in blood parameters were identified. Conclusions. Fibrin pad was as effective as CTR for the primary management of severe bleeding without occlusion of the renal artery and a shorter surgical time. No evidence of a systemic or local adverse response was identified due to exposure to the Fibrin pad. Ó 2011 Elsevier Inc. All rights reserved. Key Words: severe bleeding; animal model; partial nephrectomy; EVITHROM; SURGIFLO; SURGICEL.
INTRODUCTION
The delivery of fibrin glue components using a backing pad has been described using either a horse-derived backing [1], VICRYL* Mesh backing [2], and with Oxidized Regenerated Cellulose [3]. These products have been shown to be capable of maintaining shortterm hemostasis in severe aortic injuries [4] and renal models of severe bleeding [5]. These products have also been shown to have no excess risk of thrombogenicity [6, 7]. Fibrin Pad (FP) described in this report has been previously shown to be effective in achieving hemostasis in severe parenchymal trauma models in severe coagulopathic states [8]. The objective of this study was to compare the effectiveness of a new combination product (FP) to conventional therapy of sequential use of temporary renal artery occlusion, electrocautery, SURGIFLO þ EVITHROM, SURGICEL NU-KNIT, and PDS II suture for the primary management of severe bleeding (defined as more intense pulsatile and/or rapidly
e291
0022-4804/$36.00 Ó 2011 Elsevier Inc. All rights reserved.
e292
JOURNAL OF SURGICAL RESEARCH: VOL. 167, NO. 2, MAY 15, 2011
flowing bleeding that could come from one or more artery or vein). The animal model has been used for similar evaluations of hemostats [5], and because the kidney is one of the most perfused organs, it is considered a worst-case model of parenchymal bleeding. Successfully demonstrating durable hemostasis in a model of this severity is intended to support a variety of other clinical applications. The conventional therapy described above includes both active and inactive hemostatic modalities that induce clotting, and physically stop bleeding. Similarly, the test article also contains pharmaceutical and device components that contribute to hemostais by both active and passive mechanisms, however it can be applied without clamping the renal artery. Because the two options rely on similar dual mechanisms, the study objectives are logical. MATERIALS AND METHODS Test Article FP is an absorbable combination product consisting of a flexible, easy to conform, matrix consisting of layer of oxidized regenerated cellulose (ORC: similar to SURGICEL topical absorbable hemostat and INTERCEED adhesion prevention barrier) and polyglactin 910 fibers (similar to VICRYL suture and mesh products) with a layer of lyophilized human plasma-derived fibrinogen and thrombin (similar to EVICEL Fibrin sealant). Fibrin pad is intended to be used when standard surgical methods of hemostasis are impractical or ineffective.
Test System Thirty-six healthy female pigs, approximately 12–13 wk of age (weighing between 34 and 56 kg), were obtained from USDA approved vendors and underwent a partial nephrectomy of the caudal pole of the right kidney. The swine model was used based on its similarity to human coagulation biochemistry and gross renal anatomy [9]. This established swine model has been used for similar evaluations of hemostatic agents [5].
pre-weighed laparotomy sponges were placed in the abdominal cavity surrounding the right kidney. For treatment groups 1 and 2, the caudal pole of the right kidney was resected at a 1.0 6 0.5 cm depth, creating an approximate 2.5 6 1 cm diameter resection surface using a scalpel. The nephrectomy was intended to include the full depth of the renal cortex without intrusion into the renal pelvis. As such, several arcuate arteries are typically involved. An approximately 2 3 3 in. section of Fibrin pad was applied directly to the resection surface with the ORC backing up. Manual tamponade was applied for 3 min using pre-weighed laparotomy sponges. The time from amputating the kidney to hemostasis was recorded (including the 3 min for the manual tamponade after article application). Improperly applied articles were removed and another application of Fibrin pad was applied. All visible blood was collected, and the pre-weighed laparotomy sponges were removed and re-weighed. Five of 12 animals required reapplication in group 1; also, in group 2, 5 of 12 animals required reapplication. In treatment group 3, the renal artery was temporarily occluded by the use of atraumatic vascular clamps. The caudal pole was resected as previously described. Electrocautery was utilized to stop bleeding from larger vessels. The resection surface was dried as much as possible using pre-weighed laparotomy sponges. SURGIFLO and EVITHROM were then applied to the cut surface with moist gauze sponges. Manual tamponade was applied for approximately 3 min. A double layer of 2 3 3 in. SURGICEL NU-KNIT was then applied and sutured in place with overlying PDS bolstering sutures fixed to the renal capsule. The vascular clamp was released and then the resection surface was evaluated for bleeding. If necessary, an additional 3 min compression was applied to achieve hemostasis. The time from clamping the renal artery to release of the clamp and the observation of no active bleeding was considered the time to hemostasis. All visible blood was collected and the pre-weighed laparotomy sponges were removed and re-weighed as a measure of blood loss. For all animals, the mean arterial pressure (MAP) was measured via indirect blood pressure cuff and maintained within 620 mm Hg of the pre-nephrectomy MAP value by intravenous infusion of lactated Ringer’s solution. The capsule surrounding the right kidney, primary abdominal wall, fascia, subcutaneous layers, and skin were closed with appropriately sized suture and suture patterns. Following surgery, the animals were closely monitored during anesthetic recovery for physiologic disturbances, including cardiovascular/respiratory depression, hypothermia, and hypovolemia.
Clinical Pathology Study Design Treatment group 1 consisted of 12 pigs with the Fibrin pad applied to the resection surface and a 48 (62) h study interval. Treatment group 2 consisted of 12 pigs treated with Fibrin pad, and the animals were evaluated at 10 and 28 d with MRI, and necropsy and histopathology of selected tissues at 14 (62) or 56 (62) d post-surgery (6 animals at each time). The resection surface of treatment group 3 was treated with temporary renal artery occlusion, electrocautery, SURGIFLO, EVITHROM, SURGICEL NU-KNIT, and PDS II, and was evaluated similar to treatment group 2.
Surgical Model Following induction of anesthesia, a ventral midline incision was made from the xyphoid to just cranial of the pubis. Routine clamping and electrocautery was used to control cutaneous hemorrhage. The abdominal organs were inspected for evidence of past or present pathology. Because of the potential for the porcine spleen to sequester a variable red cell mass during hemorrhage, a splenectomy was performed. The spleen was weighed and an intravenous bolus injection of warm lactated Ringer’s solution at three times the weight of the explanted spleen was administered. Following the splenectomy,
For treatment group 1, standard hematology evaluations and a coagulation profile (prothrombin time (PT), activated partial thromboplastin time (aPTT), and fibrinogen) were conducted on all animals prior to surgery and at 1 and 2 d postoperatively. For treatment groups 2 and 3, a coagulation profile (PT and aPTT), and standard hematology, blood serum chemistry and urinalysis evaluations were conducted on all animals prior to surgery and at 10 6 2 and 28 6 2 d postoperatively, and prior to necropsy (14 or 56 d post-surgery). All blood samples were collected from the jugular vein or other suitable vein. The samples were collected into tubes containing EDTA for evaluation of hematology parameters, citrate for evaluation of coagulation parameters, and clot tubes for serum biochemistry samples. Urine samples were collected via cystocentesis.
MRI Evaluations The animals from treatment groups 2 and 3 were anesthetized and MRI evaluations of the resection sites (Northern Biomedical Research, Muskegon, MI) were conducted pre-surgery, and at 10 and 28 d. An independent radiologist evaluated the resection surface images and measured fluid accumulations when present.
HUTCHINSON ET AL.: FIBRIN PAD EFFECTIVENESS AFTER PARTIAL NEPHRECTOMY
Scheduled Euthanasia Treatment group 1 animals were euthanized at 48 6 2 h. Necropsies were conducted and the thoracic and abdominal vasculature was examined for thrombosis. A section from each lung lobe and the resection site were collected for microscopic evaluation. Animals from treatment groups 2 and 3 were divided into two. Half of the animals from each treatment group were necropsied at 14 (62) or 56 (62) d post-surgery. The lung (three sections from each lobe), liver, right and left kidneys, heart, ureters, bladder, pancreas, and adrenals were collected for microscopic evaluation. In addition, the arterial and venous vessels were traced to the abdominal aorta and vena cava, respectively, and examined for thrombosis. All harvested tissues were fixed in 10% neutral buffered formalin (NBF) for at least 48 h. Tissues were embedded, sectioned, mounted, and stained with hematoxylin and eosin according to standard procedures.
Statistical Comparisons Total treatment times and blood loss values for groups 2 and 3 were compared because these two groups were conducted together with the same surgical team in a randomized order. Group 1 was conducted at a later time, and although the same procedure was used, the surgical staff available at the time was different. Because the hemostasis time and blood loss evaluation is somewhat dependant on the surgeon, the inclusion of group 1 in the statistical comparison would add an additional variable that is not relevant when only groups 2 and 3 are compared. Groups 2 and 3 were compared using a t-test (Graphpad Prism ver. 5.02; GraphPad Software, San Diego, CA). When comparing peri-renal fluid, body weight, hematology (except leukocyte count), coagulation parameters, clinical chemistry, and urinalysis values, Levene’s was used to assess homogeneity of group variances for each specified endpoint and for all collection intervals. If Levene’s test was not significant (P > 0.01), a pooled estimate of the variance (mean square error or MSE) was computed from a one-way analysis of variance (ANOVA) and utilized by a Dunnett’s comparison of each treatment group with the CTR group. If Levene’s test was significant (P < 0.01), comparisons with the CTR group were made using Welch’s t-test with a Bonferroni correction. Historic data indicate that total and differential leukocyte counts are not normally distributed; therefore a log transformation was performed and then group pair-wise comparisons were conducted using Welch’s t-test with a Bonferroni correction (SAS ver. 8.2; SAS Institute, Cary, NC).
e293
animals gained weight over the study period for treatment groups 2 and 3. Intraoperative Hemostasis
Assessment of intraoperative bleeding demonstrated that FP was as effective as the CTR that included renal artery occlusion at establishing hemostasis. Twelve animals in the FP group (group 2) did not require occlusion of the renal artery, and hemostasis was obtained with between one and three applications of FP with an average manual tamponade time of 5.9 6 2.6 min, and was statistically significant between the Fibrin pad and CTR treatment groups (P ¼ 0.0013; Fig. 1). Similar results were repeated with another group (group 1) of 12 animals with one to two applications of FP and an average manual tamponade time of 7.1 6 4.2 min. The CTR group (group 3) required occlusion of the renal artery for an average of 10.8 6 2.6 min. Groups 2 and 3 were conducted in random order by the same surgical team, and group 1 was conducted by a different surgical team, which may account for slightly different hemostasis times for groups 1 and 2. A similar amount of blood loss was found in each treatment group with an average of 176.7 6 80.9 mL blood loss in animals receiving Fibrin pad (no renal artery occlusion) and an average of 139.9 6 63.0 mL of blood loss in animals receiving the conventional therapy (with renal artery occlusion). The difference in blood loss was not statistically significant (P ¼ 0.2272), and
Animal Welfare Compliance The in vivo portions of this study were conducted in facilities accredited by the Association for the Accreditation and Assessment of Laboratory Animal Care International (AAALAC). The guide for the care and use of laboratory animals was followed [10].
RESULTS Size of Partial Nephrectomy
The portion of kidney removed was on average 3.0 (60.19) cm long, 1.6 (60.05) cm wide, 1.1 (60.08) cm deep, and weighed 1.92 (60.51) g. Clinical Finidngs and Body Weights
Clinical findings were consistent for the age of the animals as well as for animals that had undergone a recent significant abdominal surgical procedure. All
FIG. 1. Box and whiskers plot comparing conventional treatment to FP treatment. The total treatment times for both groups are compared, the two groups were statistically different (P ¼ 0.0013). For the FP group, treatment time is measured from the time that the amputation occurs until the time that manual tamponade is completed and hemostasis is confirmed. For the CTR group, the time is begun at renal hilus clamping (immediately prior to renal tissue amputation) until hemostasis is confirmed after removal of the renal clamp. This time therefore reflects the renal ischemia time avoided by the use of FP (n ¼ 12/group). In this plot, the box indicates the 25th and 75th percentiles (with the median), and the whiskers indicate the minimum and maximum values.
e294
JOURNAL OF SURGICAL RESEARCH: VOL. 167, NO. 2, MAY 15, 2011
biologically trivial when one considers the potential hemorrhage from a partial nephrectomy wound. MRI Evaluations
MRI evaluations on d 10 and 28 demonstrated increased edema and significantly increased perirenal fluid in the FP group compared with the CTR group at d 10 (P < 0.001), but no significant difference at d 28 (P > 0.05). Edema and perirenal fluid accumulations were not present at necropsy for the animals necropsied at 14 or 56 d, so the significance of the MRI findings are unclear. The average area of fluid in the perirenal space was found to decrease over time in both treatment groups. At 10 d, the average area in the perirenal space for the FP group was 5.45 6 1.40 cm2 and for the CTR group it was 1.63 6 1.07 cm2. At 28 d, the average area in the perirenal space for the FP group was decreased to 1.98 6 1.05 cm2, and for the CTR group it was 0.22 6 0.34 cm2 (Fig. 2). Fluid accumulations were interpreted as transudates with blood products by MRI; however, necropsy findings and microscopic evaluation in the animals necropsied at 14 d do not correlate to the MRI interpretation (i.e.; no evidence of fluid accumulation with blood components). The necropsy findings and microscopic evaluation from treatment group 1 animals (48 6 2 h post-implantation) suggest that the edema fluid accumulation was within the adjacent surgical traumatized tissues, and when perirenal fluid was present, it was consistent with an ultrafiltrate of blood rather than a transudate with blood components.
FIG. 2. Accumulation of perirenal fluid. Calculated volumes of perirenal fluid accumulation as observed during MRI evaluations at 10 and 28 d. At 10 d the two treatments were statistically different (P < 0.001); at 28 d there was no significant difference (P > 0.05). The volume of fluid decreased in both groups, but was delayed in the FP group (compared with CTR). The mean and standard error (SEM) are plotted for each group.
Intraparenchymal edema fluid and hemorrhage was identified in the MRI evaluation. This change was not measured due to the inability to reliably define the edges of the alteration. The intraparenchymal changes were present in 10 of 11 Fibrin pad animals at 10 d postsurgery, and in 3 of 5 animals at 28 d post-implantation. Similar alterations were present in the CTR group for 11 of 11 animals at 10 d post-implantation, and 6 of 6 animals at 28 d post-implantation. Intraparenchymal hemorrhage, occasional clots in small vessels damaged by the resection procedure and necrosis was present in group 1 animals at 48 h and corresponded to the MRI reported alterations for treatment group 2 and 3 animals at 10 d. Macroscopically and microscopically the intraparenchymal areas were centrally located in the resection area and were cone-shaped but did not have a vascular orientation. The areas most likely represent damage due to blood from the resection surface forced along a path of least resistance (center of resection surface close to renal pelvis). Clinical Pathology
Hematology parameters were within the normal range for swine for treatment groups 2 and 3 (Table 1). A few parameters were noted to be statistically different between treatment groups 2 and 3; however, since the values were within the normal range for swine these differences were not considered biologically relevant and were believed to represent normal biological variation. No consistent hematology changes were observed in treatment group 1 (taken pre-surgery, 1 and 2 d post-surgery) except increased white blood cell counts 1 day post-implantation (data not shown). This change reflects a response to the surgical procedure and is not considered treatment related. Mean values for the clinical chemistries were within the normal range except for alkaline phosphatase, which was increased for treatment groups 2 and 3 at all time points (Table 2). The total protein was low at all time points for both treatment groups 2 and 3. Statistically different mean values were detected occasionally between treatment groups 2 and 3; however, the values were still within the normal range and the differences were not considered biologically significant. Prothrombin time (PT) and activated partial thromboplastin time (APTT) for all animals were within the expected range and were not statistically different between the treatment groups (Table 3). The urinalysis parameters were considered normal for treatment groups 2 and 3 (Table 4). Necropsy and Histopathology
At 2, 14, and 56 d post-implantation there was no evidence of biologically significant post-surgical
e295
HUTCHINSON ET AL.: FIBRIN PAD EFFECTIVENESS AFTER PARTIAL NEPHRECTOMY
TABLE 1 Selected Hematology Results (Mean ± SD). Results Significantly Different Than Control (P < 0.05) Are Marked with *
Leukocytes (103/uL) Erythrocytes (106/uL) Hematocrit (%) Platelets (103/uL) Absolute Reticulocytes (103/uL)
Control FP Control FP Control FP Control FP Control FP
Pretest
Day 10
Day 14
Day 28
Day 56
18.88 (4.689) 19.58 (4.661) 6.623 (0.597) 6.436 (0.7529) 34.17 (3.775) 33.71 (5.402) 467.2 (92.85) 425.6 (127.01) 57.19 (28.918) 50.28 (38.725)
19.48 (5.347) 22.33 (4.62) 6.489 (0.455) 6.358 (0.7151) 32.25 (1.659) 32.91 (3.767) 677.3 (242.45) 654.8 (101.34) 68.62 (17.887) 102.24 (46.107)*
18.95 (5.256) 22.33 (4.036) 6.617 (0.3625) 6.313 (0.4439) 32.97 (1.703) 31.43 (1.247) 504.3 (54.35) 604.2 (149.82) 88.82 (42.335) 61.53 (33.548)
19.73 (6.066) 22.7 (4.481) 6.483 (0.702) 6.512 (0.4527) 31.5 (2.505) 33.33 (2.343) 574.2 (208.43) 475.7 (43.81) 65.82 (17.633) 76.43 (32.829)
20.02 (4.386) 18.7 (2.433) 6.715 (0.3769) 6.44 (0.3301) 33.53 (1.877) 32.8 (1.219) 441.5 (81.97) 473.4 (52.3) 73.2 (38.516) 63.14 (22.082)
hemorrhage in any of the animals examined. No thrombi were identified in the major vasculature of the thoracic or abdominal cavities or in examined sections of lung for any of the animals evaluated. At necropsy, the expected findings indicative of abdominal surgery in the pig were present, including adhesion formation and variable fluid accumulations. There were no biologically relevant differences in adhesion formation between the Fibrin pad and the CTR treatment groups, or any evidence of a macroscopic adverse response to the treatment received. In treatment group 1, thin hematomas were occasionally located between the Fibrin pad and the resection surface. The hematomas merged with the resection surface and the Fibrin pad and were not considered clinically relevant.
At 48 h post-implantation (treatment group 1), occasional clots involving small blood vessels and glomeruli damaged by the resection were observed along the resection surface. These clots were considered normal and necessary for the cessation of hemorrhage from a resection surface. No other thrombi were present within the sections of kidney examined. Intrarenal hemorrhage, necrosis, and degeneration were present, and were the result of mechanical trauma associated with the surgical and implant procedure. Inflammation associated with the Fibrin pad was limited at 48 h. At 2 weeks, FP group (group 2) had decreased severity of inflammation (3 of 6 moderate, 3 of 6 severe), fibrosis/fibroplasia (2 of 6 moderate, 4 of 6 severe), and tubular degeneration/regeneration (1 minimal, 1 mild,
TABLE 2 Selected Clinical Chemistry Results (Mean ± SD). Results Significantly Different Than Control (P < 0.05) are marked with *
Chloride (mEq/L) Alkaline Phosphatase (U/L) GGT (U/L) AST (U/L) ALT (U/L) Sorbitol Dehydrogenase (U/L) Urea Nitrogen (mg/dL) Creatinine (mg/dL) Total Protein (g/dL) Albumin (g/dL) Globulin (g/dL)
Control FP Control FP Control FP Control FP Control FP Control FP Control FP Control FP Control FP Control FP Control FP
Pretest
Day 10
Day 14
Day 28
Day 56
102.7 (1.3) 103.4 (2.78) 114.6 (18.61) 122 (54.94) 33.4 (13.23) 36.4 (12.49) 33.3 (11.69) 41.8 (17.4) 36.2 (9.93) 39.2 (9.56) 5.43 (6.278) 3.14 (2.015) 9.1 (2.78) 8.3 (2.19) 1.11 (0.116) 1.13 (0.13) 6.61 (0.425) 6.56 (0.54) 3.52 (0.266) 3.39 (0.348) 3.09 (0.355) 3.17 (0.623)
100.9 (2.35) 102.4 (2.11) 94.5 (29.64) 100.5 (36.2) 32.7 (7.1) 36.5 (12.46) 29.5 (10.94) 28.2 (7.9) 37.9 (11.15) 38.9 (8.28) 2.03 (1.709) 3.48 (2.15) 8.8 (2.09) 8.5 (3.29) 1.03 (0.15) 1.09 (0.261) 6.28 (0.409) 6.37 (0.464) 3.29 (0.287) 3.27 (0.227) 2.98 (0.484) 3.1 (0.447)
100.7 (3.01) 102.7 (2.8) 101.8 (36.92) 84.2 (17.93) 32.8 (9.2) 31.8 (6.74) 37.3 (21.2) 23.2 (3.87) 39.5 (5.21) 38.8 (7.17) 2.27 (2.185) 2.83 (2.408) 7 (2.37) 9.2 (5.49) 1.07 (0.103) 1.2 (0.126) 6.45 (0.409) 6.32 (0.417) 3.28 (0.343) 3.35 (0.281) 3.17 (0.717) 2.97 (0.501)
100.7 (3.33) 101.2 (1.72) 96 (17.46) 112.7 (22.84) 30.3 (7.74) 36.3 (10.76) 28.3 (8.26) 26.2 (5.78) 39.2 (10.07) 41.2 (5.98) 1.03 (0.361) 1.42 (1.137) 9.7 (3.78) 8.2 (1.94) 1.28 (0.133) 1.2 (0.167) 6.55 (0.476) 6.73 (0.505) 3.5 (0.283) 3.47 (0.333) 3.05 (0.468) 3.27 (0.565)
101.5 (1.97) 103.7 (1.03)* 89 (16.27) 100.3 (16.99) 33.5 (10.25) 37.3 (8.57) 23.5 (4.51) 23.8 (7.05) 40.7 (9.48) 40.8 (4.36) 1.9 (1.066) 2.4 (0.663) 11.7 (4.68) 13.7 (3.08) 1.47 (0.175) 1.4 (0.21) 6.57 (0.393) 6.5 (0.329) 3.62 (0.286) 3.72 (0.172) 2.95 (0.575) 2.78 (0.397)
e296
JOURNAL OF SURGICAL RESEARCH: VOL. 167, NO. 2, MAY 15, 2011
TABLE 3 Coagulation Profile (Mean ± SD)
APTT (sec) Prothrombin Time (sec) Fibrinogenx
Control FP Control FP FP
Pretest
Day 1
Day 2
Day 10
Day 14
Day 28
Day 56
13.83 6 3.7* 13.63 6 2.0y 11.62 6 0.5* 14.12 6 0.7z 166.4 6 46.64
NC 16.9 6 3.1 NC 14.98 6 0.9 375.3 6 63.6
NC 14.43 6 2.8 NC 14.97 6 1.5 311.3 6 82.96
15.30 6 4.7 11.52 6 6.3 11.49 6 0.5 11.30 6 0.8 NC
15.44 6 1.4 15.78 6 1.0 11.26 6 0.5 11.82 6 0.5 NC
12.78 6 4.9 15.22 6 1.54 11.28 6 0.5 11.44 6 0.8 NC
13.08 6 5.8 15.58 6 1.18 11.44 6 0.6 11.56 6 0.3 NC
NC¼ Not Collected. Pretest value for groups 2 and 3, data not collected for group 1. y Pretest results for group 1 are presented, pretest APTT for groups 2 and 3 was 13.0865.1 sec. z Pretest results for group 1 are presented, pretest PT for groups 2 and 3 was 11.3560.6 sec. x Test only conducted for group 1. *
1 moderate, and 2 severe), and increased neovascularization (4 mild and 2 moderate) at d 14 compared with the CTR group (group 3). This may reflect the differences in the surgical manipulation and article application procedures between the CTR group versus the Fibrin pad group. Over the course of the study, the character of the inflammation seen in FP treated animals was observed to change from a primarily involving macrophages (6 of 6) response at d 14 to a more mature response involving lymphocytes, plasma cells, and macrophages at d 56. These changes reflect normal healing response and are not considered adverse effects. Some of the noted inflammation (i.e., the lymphoplasmacytic aggregates) associated with the Fibrin pad is suggestive of an immune response to the foreign (human) proteins in the Fibrin pad. The results of this study support the safety and biocompatibility of FP (Table 5). FP was considered essentially absorbed because less than 10% of the FP material remained at 56 d (data not shown). The material remaining represented the matrix (VICRYL) component of the Fibrin pad. DISCUSSION
Considering the morbidity, mortality, and extended hospitalization due to uncontrolled severe traumatic or surgical bleeding, the timely control of bleeding has obvious benefits for patients and for surgeons, and is an active area of research [11-14]. Surgeons encounter
bleeding from a variety of tissue types, intensity, and etiologies. The selection of appropriate methods or products to control bleeding is dependent upon many factors, which include but are not limited to: 1. Bleeding severity 2. Anatomic location of the source and the proximity of adjacent critical structures 3. Whether the bleeding is from a discrete source or from a broader surface area 4. Visibility and precise identification of the source 5. Access to the source 6. Tissue type and fragility/friability 7. Coagulation system status 8. Patient stability Bleeding of mild, moderate, or even severe intensity can frequently be quickly and successfully treated with traditional mechanical methods (i.e., suture, ligature, clips, and staples), or with various energy-based coagulation systems, (i.e., electrocautery, argon beam lasers, and ultrasound) [15, 16]. When these conventional measures are ineffective or impractical, adjunctive hemostasis techniques and products are typically utilized. These include topical absorbable hemostats (TAH), such as oxidized regenerated cellulose, gelatin, or collagen [17] and active hemostats, such as topical thrombin or fibrin sealants [18, 19]. These products have proven most efficacious in controlling slowly bleeding foci, diffuse oozing, bleeding from needle puncture sites, and diffuse
TABLE 4 Urinalysis (Mean ± SD)
Volume (ml) Specific Gravity pH
Control FP Control FP Control FP
Pretest
Day 10
Day 14
Day 28
Day 56
9.50 (1.7) 10 (0) 1.0100 (0.004) 1.0103 (0.004) 6.33 (0.86) 6.54 (1.1)
9.17 (1.9) 9.58 (1.4) 1.0104 (0.004) 1.0116 (0.004) 7.17 (0.72) 6.88 (0.9)
10 (0) 10 (0) 1.0125 (0.007) 1.0198 (0.009) 6.42 (0.74) 6.25 (1.3)
10 (0) 10 (0) 1.0053 (0.004) 1.0085 (0.004) 7.33 (0.75) 6.83 (0.7)
10 (0) 10 (0) 1.0082 (0.009) 1.0120 (0.012) 6.75 (0.69) 6.83 (0.4)
y
*
Not treatment related (attributed to major abdominal surgery). Within acceptable level as represents part of normal healing. z Defined as presence of macrophages. x Mechanically induced due to blood flow from resection surface along the path of least resistance (central resection site area closest to the renal pelvis).
N/A N/A N/A 7/12 minimal to mild
N/A
N/A 6/6 minimal to mild 0/6 6/6 minimal to mild 4/6y minimal to mild 1/6x Mild 0/6 6/6x minimal to mild 6/6z moderate to severe N/A N/A N/A 6/6y mild to moderate 6/6x minimal to mild 1/6x mild 5/6x mild to severe 6/6z severe N/A N/A N/A 5/6y mild 4/6x mild to moderate 3/6x minimal to moderate 6/6x moderate to severe N/A N/A N/A N/A N/A 4/12x minimal to mild 4/12x minimal to mild 5/12x minimal to mild
N/A 5/6 mild to moderate 1/6z moderate 3/6 mild 6/6y minimal to mild 0/6 3/6x minimal 6/6x mild to moderate
4/6* Mild 6/6y minimal to mild N/A 6/6* minimal to moderate 6/6y mild to moderate N/A 6/6* mild to severe 6/6y moderate to severe N/A 6/6* moderate to severe 6/6y severe N/A Miscellaneous finding * 12/12 y minimal 12/12 minimal
Adhesion Fibrosis/Fibroplasia Infiltrate of low numbers of macrophages and to lesser extent neutrophils Inflammation, granulomatous Inflammation, chronic Inflammation, chronic-active Foamy/pigmented macrophages Neovascularization Necrosis (Intrarenal) Hemorrhage (Intrarenal) Tubular degeneration/ regeneration (Intrarenal) Resection surface completely or partially covered by hematoma
Day 56 conventional Day 14 Fibrin pad Day 14 conventional Day 2 Fibrin pad Microscopic changes
Microscopic Changes in Right Kidney (Tissue in Contact with Implant)
TABLE 5
Day 56 Fibrin pad
HUTCHINSON ET AL.: FIBRIN PAD EFFECTIVENESS AFTER PARTIAL NEPHRECTOMY
e297
parenchymal organ hemorrhage. They are not as effective in the control of some forms of active or severe bleeding where the topically applied agents can be simply carried away from the bleeding site by actively flowing blood or require a dry field for application [11, 12]. Timely control of bleeding has obvious benefits for patients and for surgeons. Minimization of blood loss is important to help avoid hypovolemia, hemodynamic and metabolic deterioration, and anemia [13]. Minimization of exposure to blood component transfusion and the attendant risks has become increasingly important to patients and to surgeons [14]. Control of bleeding also facilitates surgeons’ visualization of the surgical site, enabling precise and timely conduct of the surgical procedure. The present study includes a new hemostatic product being developed to meet a current unmet surgical need, to rapidly and reliably control bleeding that is not amenable to treatment with standard methods of ligation or energy based coagulation and for which currently available products are not satisfactorily effective. Johnston, et al. (2006) [20] compared seven hemostats (CoSeal, CoStasis, Tisseel, FloSeal, Dermabond, Gelfoam/Tisseel, and Bolster/Floseal) in a porcine partial nephrectomy model where several sizes of lesion were made; the largest was an amputation of the kidney at the level of the hilus. Transient hypertension was created using dopamine, and the hemostatic agents were applied while the renal hilus was clamped. The study concluded that most of the hemostatic agents tested were effective for small target application sites, but only sutured bolsters were consistently effective for large partial nephrectomies. The present study explores the efficacy of test hemostats without clamping of the renal hilus, and allows for assessing the durability of the hemostatic repair by surviving animals until absorption of the product is complete. Dalpiaz et al. (2008) [21] reviewed hemostat usage in 15 animal models and 11 clinical trials. This review concludes that the efficacy of fibrin sealants was demonstrated in animal partial nephrectomy models and has been confirmed in clinical trials. Nine of the animal models reported required some form of hilar control. The present study demonstrates efficacy without the need for hilar control. Overall, FP was as effective as CTR in establishing and maintaining hemostasis in a swine partial nephrectomy model. The use of FP eliminated the need for occlusion of the renal artery and reduced the overall treatment time to achieve hemostasis by half, thereby demonstrating both safety and performance advantages. Intraoperative hemostasis was achieved and was maintained in all animals, and there was no evidence for loss of hemostatic effectiveness in any animal during follow-up. Clinical pathology, necropsy, and histopathologic assessments
e298
JOURNAL OF SURGICAL RESEARCH: VOL. 167, NO. 2, MAY 15, 2011
did not demonstrate the presence of any adverse response to Fibrin pad. These data demonstrate that FP is an effective primary and sole hemostatic treatment in this severely challenging bleeding model, avoids the need for renal artery occlusion, and the sequential use of cautery, TAH with Thrombin, ORC, and suture. We painstakingly looked for evidence for either a local or systemic propensity for thrombogenicity (gross, histologic, and clinical pathology) and found none. CONCLUSIONS
Successful completion of the safety and efficacy endpoints included in this model support the advancement of FP into clinical evaluation in challenging soft tissue and parenchymal bleeding. ACKNOWLEDGMENTS The authors thank the staff at MPI research, in particular, Keith Nelson, DVM, who served as the study pathologist at MPI. The authors also acknowledge Alexia McKnight, DVM, who interpreted the MRI results (University of Pennsylvania School of Veterinary Medicine, Kennett Square, Pennsylvania). The authors thank Drs. Xiao-Yu Song, M.D., Ramy Mahmoud, M.D., and Larry Johnson, DVM, Ph.D. for their generous support and helpful discussions during the preparation of this manuscript.
REFERENCES 1. Erdogan D, van Gulik TM. Evolution of fibrinogen-coated collagen patch for use as a topical hemostatic agent. J Biomed Mater Res Part B: Appl Biomater 2008;85B:272. 2. Kheirabadi BS, Acheson EM, Deguzman R, et al. The potential utility of fibrin sealant dressing in repair of vascular injury in swine. J Trauma 2007;62:94. 3. Finley DS, Lee DI, Eichel L, et al. Fibrin glue-oxidized cellulose sandwich for laproscopic wedge resection of small renal lesions. J Urolol 2005;173:1477. 4. Kheirabadi BS, Acheson EM, Deguzman R, et al. Hemostatic efficacy of two advanced dressings in an aortic hemorrhage model in swine. J Trauma 2005;59:25. 5. Cornum RL, Morey AF, Harris R, et al. Does the absorbable fibrin adhesive bandage facilitate partial nephrectomy? J Urol 2000;164:864.
6. Kheirabadi BS, Sieber J, Holcomb JB. Assessment of the thrombogenic effect of fibrin sealant dressing in a vascular surgery model in rabbits. J Invest Surg 2006;19:387. 7. Birth M, Figueras J, Bernardini S, et al. Collagen fleece-bound fibrin sealant is not associated with an increased risk of thromboembolic events or major bleeding after its use for hemostasis in surgery: A prospective multicenter surveillance study. Patient Safety Surg 2009;3:13. 8. Delgado AV, Kheirabadi BS, Fruchterman TM, et al. A novel biologic hemostatic dressing (fibrin patch) reduces blood loss and resuscitation volume and improves survival in hypothermic, coagulopathic Swine with grade V liver injury. J Trauma 2008; 64:75. 9. Pusateri AE, Holcomb JB, Kheirabadi BS, et al. Making sense of the preclinical literature on advanced hemostatic products. J Trauma 2006;60:674. 10. Institute of Laboratory Animal Research, Commission on Life Sciences, National Research Council. Guide for the Care and Use of Laboratory Animals. Washington, DC: National Academy Press, 1996. 11. de la Torre RA, Bachman SL, Wheeler AA, et al. Hemostasis and hemostatic agents in minimally invasive surgery. Surgery 2007; 142(4 Suppl):S39. 12. Msezane LP, Katz MH, Gofrit ON, et al. Hemostatic agents and instruments in laparoscopic renal surgery. J Endourol 2008; 22:403. 13. Shander A. Financial and clinical outcomes associated with surgical bleeding complications. Surgery 2007;142:S21. 14. Boucher BA, Hannon TJ. Blood management: a primer for clinicians. Pharmacotherapy 2007;27:1394. 15. Aziz O, Athanasiou T, Darzi A. Haemostasis using a ready-to-use collagen sponge coated with activated thrombin and fibrinogen. Surg Technol Int 2005;14:35. 16. Acheson EM, Kheirabadi BS, Deguzman R, et al. Comparison of hemorrhage control agents applied to lethal extremity arterial hemorrhages in swine. J Trauma 2005;59:865. 17. Palm MD, Altman JS. Topical hemostatic agents: A review. Dermatol Surg 2008;43:431. 18. Recinos GA, Inaba KA, Dubose JA, et al. Local and systemic hemostatics in trauma: A review. Ulusal Travma ve Acil Cerrahi Dergisi 2008;14:175. 19. Gaarder C, Naess PA, Frischknecht Christensen E, et al. Scandinavian guidelines: The massively bleeding patient. Scand J Surg 2008;97:15. 20. Johnston WK, Kelel KM, Hollenbeck BK, et al. Acute integrity of closure for partial nephrectomy: Comparison of 7 agents in a hypertensive porcine model. J Urol 2006;175:2307. 21. Dalpaiz O, Neururer R, Bartsch G, et al. Hemostatic sealants in nephron-sparing surgery: What surgeons need to know. BJU International 2008;102:1502.