The pericardium treated with glutaraldehyde is one of the biological materials more widely used in cardiovascular surgery [3-5]. Despite the good results obtained in clinical trials, degeneration of biological tissue is frequently observed, especially calcification, which is the major cause of dysfunction of these tissues that may cause mechanical failure and degeneration [6-8].

Pires et al [9,10] described the profile of bovine pericardium preserved in glutaraldehyde, while employed as a vascular patch. They reported large functional differences between the smooth (membranous) and rough (fibrosis) face of the pericardium. These differences depended on the location. The authors suggested a possible correlation between calcification and the contact of the rough face of the pericardium with the bloodstream. Several experiments were conducted in order to avoid or reduce the calcification of glutaraldehyde-treated pericardium [9,10], but none was able to completely inhibit the calcification of the material in question.

We implanted the standardized retails of bovine pericardium preserved in glutaraldehyde, obtained from the Gravataí industry in 14 young Large White breed pigs, weighing between 10.3 kg and 18.4 kg, six females and eight males, We implanted three remnants in each pig, two in the abdominal aorta and one was juxtaposed to peritoneum, with a smooth face (SF), the other with the rough face (RF) facing the lumen and one remnant was juxtaposed to peritoneum (Figure ​(Figure1).1). We followed the following standardized procedures:

Animals were kept in the immediate postoperative period in the vivarium of our University, where they received antibiotic prophylaxis with intramuscular penicillin G benzathine (Sigma^®) at a dose of 600,000 units, analgesia with tenoxicam (Sigma^®) in the first 24 hours after surgery, and all the necessary postoperative care including feeding, vaccinations and other medications, supervised by a veterinarian. They remained on average until the seventh postoperative day, depending on their clinical conditions. They were then transferred to a farm specializing in raising pigs where they received usual care as needed.

The animals were divided into three groups according to their postoperative period, ranging from 4.5 to 8 months, and weighing between 75 to 109 kg. They were sacrificed using a lethal dose of sodium pentobarbital (Sigma^®). Segments of the aorta with the location of the implants were removed and longitudinally opened opposite to the implant. The aorta segment which was opened was fixed on a cork board in order to keep its shape, with the inner side facing down and fixed in aqueous 10% formaldehyde (Sigma^®). We evaluated the deformation of the retail which is characterized by its expansion in relation to the adjacent aorta.

The removed parts were analyzed at the Department of Pathology of the Universidade Federal de São Paulo. The pieces were cut along the longest axis of the retail and embedded in paraffin. Fifteen sections (8-15 μm) were cut from each tissue sample. Slides were stained by the hematoxylin - eosin method and the method of Verhoeff.

The removed parts were macroscopically and microscopically evaluated. We used criteria from previous studies to conduct a quantitative analysis of each slide as follows [13-15]:

Calcification: characterized by deposition of calcium in the intimacy of the retail and the adjacent aorta. The presence of one to two foci was considered to be mild calcification (1st degree), three foci were considered moderate (2nd degree), and more than three foci or confluent areas of calcification were considered intense (3rd degree).

Integration tissue: composed of fibrous connective tissue responsible for the integration of the bovine pericardium retail to the aorta.

Internal apposition fibrosis: characterized by the presence of fibrous tissue in the face of the pericardium toward the interior of the aorta. Internal apposition fibrosis was assesses according to the extent and thickness. The extent was considered mild (1st degree) when covering less than half of the retail area, moderate (2nd degree) when covering half or more of the retail area and intense (3rd degree) when covering the entire area of the retail. The thickness was considered mild (1st degree) when its thickness was lower than the retail thickness, moderate (2nd degree) when its thickness was equivalent to the retail thickness and intense (3rd degree) when its thickness exceeded the retail thickness.

Bone metaplasia - cartilaginous: characterized by the formation of bone tissue - cartilage in the fibrous layer of the internal apposition. It was considered mild (1st degree) in the presence of one to two foci, moderate (2nd degree) in the presence of three foci, and intense (3rd degree) in the presence of more than three foci or confluent areas.

The experiments lasted a total of 14 months and the surgical procedures were performed by the same surgical team. Data were reviewed by two independent readers.

We observed dilation of the surgical patch area in relation to the surrounding aorta in both groups. In the SF group dilation occurred in all animals and was mild in almost half. On the other hand, in the RF group we did not observe dilation in most animals (Table ​(Table11).

The calcification of the surgical patch occurred in both groups. In the SF group it was absent in more than a third of the animals, while in the RF group it was absent in over half of the animals (Table ​(Table2).2). The calcification of the aorta adjacent to the vascular patch was observed in nine animals in the SF group (Table ​(Table3)3) and in six animals in the RF group (Table ​(Table4).4). We did not observe calcification in the retail implanted in the peritoneum in approximately 40% of the animals (Tables ​(Tables33 and ​and44).

The coaptation between the bovine pericardial patch and the aortic wall was noted by apposition or slight invagination of edges in 12 cases. In two cases one side presented a slight overlap the edges. The integration between the retail and the aortic wall was observed in a thin band of fibrous connective tissue that also involved the suture.

In the group in which the flat edge of the retail was turned into the aorta, this phenomenon was more intense on the suture line. In the RF group the extent of coverage by internal apposition fibrosis of the retail area was moderate (2nd degree) in most animals (Table ​(Table5).5). In the group in which the flat edge of the retail was turned into the aorta, this phenomenon was more intense on the suture line and it occurred in both groups. In the SF group the thickness of the internal apposition fibrosis in relation to the retail area was slight (1st degree) in most animals while in the RF group we found intense thickness in all patches (Table ​(Table66).

In both groups we found a fibrous layer internal apposition that consisted of dense fibrous tissue predominantly modeled, and the cells and fibromas were compactly arranged parallel to each other, acquiring the pattern of tendons and ligaments tissue (Figure ​(Figure33).

There was only one animal (7.1%) in the RF group that presented osteo-cartilaginous metaplasia in the internal apposition fibrous layer.

It is important to note the presence of an expansion of the retail in the aortic wall. There was dilation of the material with the formation of an aneurysm wall at the implant site, which was significantly higher when the smooth face was turned to the light of the aorta. Furthermore, we observed that the bovine pericardium patch was well integrated into the wall of the aorta, confirming what Pires et al [9] observed. The structure of the patch provided by surgical retail, when implanted in the aorta of dogs, promotes clear and precise limits in preserved shape, and it is already well defined at the end of the first month after surgery [9,10]. Also, this type of procedure is vulnerable to infections [16-18].

In the group where the rough face was turned into the aorta, we observed the formation of a covering layer on the inner surface with more plasticity, modeling elastic tendon and bone formation, which provided greater tension resistance to the surgical patch, preventing its expansion. Pires et al [9,10] suggested that the internal apposition fibrosis originates from the fibrous organization of blood components deposited on the surface of the retail, because this plan was not observed in any case in retail implanted in the pericardium. In the group in which the flat edge of the retail was turned into the aorta this phenomenon was more intense on the suture line and was related to the roughness of the area and local release of thromboplastin by damaged tissues. Our results suggest that the wrinkled face of the retail would facilitate the capture and adhesion of blood components across its surface, with subsequent assemblage by the release of platelet factors and secondarily by thromboplastin released from damaged tissues [19]. We also verified that in this group there was the formation of a thin inner cover layer, resulting in dilatation of the patch area of surgery. These results may suggest that fibrosis internal apposition would support retail and strengthen the structure of the surgical patch. Its histological structure reveals a dense fibrous connective tissue shaped pattern tendon with great tensile resistance.

Researchers have offered different hypotheses to explain why glutaraldehyde helped preserve bovine pericardium from changes in its structure when deployed with different faces to the vessel lumen. Ishihara et al [11] reported that bovine pericardium treated with glutaraldehyde loses the mesothelial cells of serosal surface, exposing the submesotelial layer of connective smooth tissue, while Pires et al [9] observed that the surface of the pericardium was covered by a layer of pavement cells and fibrous connective tissue, which they called internal apposition tissue. On the other hand Schoen et al [8] reported that platelet aggregation, on the pericardial membrane, would be a factor for accelerating the calcification process. We observed that in both sides of the retail implanted in animals, an inner surface composed of endothelium was formed. However, when the retail was implanted with the rough side facing the lumen surface, it was isolated from bovine pericardium through blood, thus reducing the intensity and incidence of aortic dilatation in relation to the adjacent aorta. Without this coverage, we would expect an increase in calcification. Additionally, we observed that this layer of tissue was differentiated from the other sides and better trained when it had the wrinkled face toward the lumen. Furthermore, modeling elastic tendon and bone formation, formed a surgical patch level with the retail bovine pericardium which had good resistance and was well integrated into the aortic wall, probably because of the uniformity resulting from the molding process.

The presence of osteo-cartilaginous metaplasia in the fibrous layer of internal apposition was observed in only one case as an isolated focus, occurring in the group in which the rough face was turned into the aorta. These data are not in agreement with the findings of Pires et al [9] who described a large amount of calcium present in the inner layer of apposition. The authors hypothesized that this calcification represented an osteo-cartilaginous metaplasia and ossification of fibrous tissue covering the graft, indicating the transformation of fibroblast into osteoblasts. Nonetheless, we feel confident that our findings are valid. Our methodology was validated by Sucu et al [20] who demonstrated that the use of microscopic sections merged for evaluation of calcification was more precise than the use of the chemical method of extraction of calcium from the patch.

The calcification of biological tissue is the main cause of bioprostheses dysfunction. According to Schoen et al [7,8], Chanda et al [21], Vasudev et al [22] and Pires et al [9] this calcification is a multifactorial phenomenon, but it is not well defined and varies in different animal models. The authors pointed out that calcification in bovine pericardium used as a vascular patch presented a different connotation of calcification than valvular prostheses, because it implied a reduction in leaflet mobility, resulting in dysfunction of the prosthesis, which was not observed in vascular grafts.

We observed calcification of the retail irrespective of the side facing the lumen; the same was true regarding the retail implanted in the peritoneum of the animal. However, these finding are not consistent with findings of Schoen et al [6] and Rossi et al [23] who found calcification of retail implanted in rats. The authors described calcification of bovine pericardium from 24 hours of implantation and progressively increased with time, but did not report the presence of internal tissue apposition. We did not find this tissue in the patches implanted in the peritoneum (without direct contact with the bloodstream). These findings are consistent with findings by Gabbay et al [24], Bortolotti et al [25] and Pires et al [10], who indicated the major influences of the implant site of bovine pericardium treated with glutaraldehyde and its direct contact with the bloodstream.

We used a bovine pericardium produced by Braile Biomedica. However, several other industries also produce the same patch. Some industries use special procedures to reduce the prevalence of calcification. For example, Synovis, who produces Vascu-Guard [26], follows Apex-Processing. Briefly, in this process, the levels of residual glutaraldehyde are below the limits of detection by the sophisticated analytical methods now available (< 0.5 ppm) and products undergoing Apex-Processing have levels of cellularity that are four times lower than a variety of competitive materials including products conventionally treated.

According to our data, there was no calcification in the retail implanted in the peritoneum in almost half of the animals. Peritoneal calcification is a rare condition developed in uremic patients on continuous ambulatory peritoneal dialysis. Once peritoneal calcification is detected, it is essential to assess whether encapsulating peritoneal sclerosis develops [27].

Bovine pericardium is also implanted for the vascular reconstruction in the femoral or carotid artery as patchplasty and venous patch angioplasty [28]. Moreover, the bovine pericardium is not only used in cardiovascular surgery but also in hernia [29] or thoracic surgery [30]. Based on our data, we confirm the use of the rough face in those types of surgery. We propose future studies to investigate this possibility.