Classification of causes of death
The cause of death and relevant pathological findings in the twenty-five brown bears studied is shown in Table 1 and Figure 1. In four (4/25, 16%) cases it was not possible to determine the cause of death due to inadequate preservation of collected specimens or insufficient tissue availability. The pathological findings on six animals revealed secondary bacterial infection followed by primary traumatic event caused either by human activities or natural causes. Three causes of death (Clostridium sordellii infection in one bear, infectious canine hepatitis caused by canine adenovirus type 1 (CAdV-1) in three bears and cholangiocarcinoma in one bear) had been already described in previous papers by our group [5-7].
Mortality cases were classified both caused by (i) "human intervention" or "natural causes" and based on (ii) "infectious" or "non-infectious" etiology.
Based on caused by "human intervention" or "natural causes", seven out of the 21 (7/21, 33.3%) brown bears in which the cause of death could be determined died as a consequence of "human intervention" due to illegal hunting [wire snare hunting (n=3, 14.3%) and shooting n=2 (9.5%)], handling (n=1, 4.8%) and strychnine poisoning (n=1, 4.8%). In contrast, fourteen (14/21, 66.7%) brown bears died by "natural causes" due to traumatic lesions (n=9, 42.9%) [fights (n=4, 19%), unknown traumas (n=3, 14.3%) or infanticide (n=2, 9.5%)], infectious canine hepatitis (n=3, 14.3%), neoplasia (n=1, 4.8%) or mushroom poisoning (n=1, 4.8%).
Mortality data was also stratified based on "infectious" or "non-infectious" etiology. Nine out of the 21 (9/21, 42.9%) brown bears died due to "infectious" diseases, namely gangrenous myositis (n=5, 23.8%; four of which were proven to be from clostridiosis), infectious canine hepatitis (n=3, 14.3%) or septicemia (n=1, 4.8%). The remaining twelve (12/21, 57.1%) animals died due to "non-infectious" causes, which included handling (n=1, 4.8%), traumatic lesions (n=8, 38.1%) such as shooting, snare, fighting or infanticide, strychnine poisoning (n=1, 4.8%), mushroom poisoning (n=1, 4.8%) or neoplasia (n=1, 4.8%). Additionally, the pathological findings on six animals revealed secondary bacterial infection followed by primary traumatic event caused either by human activities (i.e. wire snare hunting) or natural causes (i.e. fights) (Figure 1). Exertional myopathy was observed in two bears, a handled animal and in one bear found in a snare.
Description of causes of death
Cases of death caused by traumas such as shooting (metallic fragments were found), wire snares, fighting or infanticide were easier to determine not only based on necropsy and histopathological studies but also using complementary diagnostic techniques (i.e. radiography) or knowledge of the behavior of this species. Ante mortem traumatic lesions always showed vascular reactions such as hemorrhages, hyperemia, congestion or edema. In fighting and infanticide cases lesions in skin, muscles or bones showed signs compatible with bites and scratches. In two animals the origin of traumas could not be determined. Other causes were more difficult to discern and are reported below.
The five bears (bears number 1, 8, 10, 15 and 23) with secondary gangrenous myositis (Table 1, Figure 1), as a consequence of wire snare hunting, infanticide, fights or traumas, showed serohemorrhagic edema in the abdominal cavity, thorax, pericardium, and skeletal muscle, and hemorrhages in heart, skeletal muscles, stomach, intestine, liver, spleen, and kidney. The case description of bear 8 was previously reported . Microscopically, vascular damage and hyperacute myodegeneration consisted of myonecrosis, edema, gas, extravasation of fibrin into the interstitial spaces, and lacunar dissolution of myofibers in skeletal muscles were observed using hematoxylin-eosin and Masson´s trichrome stains. Clostridium sordellii was identified by Gram stain and culture  as the etiological agent of the lesions in four brown bears. In two of these four bears each, the C. bifermentans and C. septicum were also isolated. The presence of C. sordellii was always associated with previous muscle damage (i.e. traumas) that triggered its proliferation. In bear number 15 Clostridium spp could not be isolated. An additional bear (number 12) showed septicemia secondary to fighting although the etiological agent could not be determined.
Gross lesions in the three bears (bears number 11, 14 and 20) with infectious canine hepatitis caused by CAdV-1 was previously described  and consisted of congestion and hemorrhages in several organs; hemorrhagic fluid in thoracic and abdominal cavities; friable and yellow-tinged liver; hepatomegaly; thickening of the gall bladder due to edema . Microscopically the main pathological findings appeared in liver and gall bladder. The liver showed mild centrolobular multifocal degeneration and necrosis of hepatocytes, with the presence of intranuclear inclusions bodies and low inflammatory infiltration mainly of lymphocytes. The gall bladder showed edema of the wall. Additionally, non-purulent encephalitis was observed. CAdV-1 was confirmed by quantitative polymerase chain reaction (qPCR) and immunohistochemistry .
Exertional (degenerative) myopathy
In two animals exertional myopathy was diagnosed. One bear (bear number 1) additionally exhibited signs of clostridiosis after being captured in a snare and died after one week. A female cub also died due to exertional myopathy after handling (bear number 9). The cub was found alone in the wild when it was two months old and it was kept in captivity until it was nine months old. The cub died during transit in an attempt to reintroduce it back into the wild. Both animals showed gross lesions consisting of dry and pale cardiac and some skeletal (mainly intercostals and femoral) muscles (Figures 2a and 2b). Microscopically severe segmental degeneration of muscles was observed and consisted of hypercontracted fibers, extensive Zenker´s hyaline degeneration and coagulative necrosis of myofibers (Figures 2c to 2h). Bear number 1 also showed an intensive infiltrate mainly consisting of lymphocytes and macrophages, as well as mineralization in the affected muscles (Figure 2f). In that animal necrotic myofibers with surviving satellite cells, invading macrophages and elongating myoblasts, all indicative of events of regeneration, were also observed (Figure 2d). The cub also showed hypoplasia of adrenal glands (1.7 grams, 0.004% relative weight; physiological relative weight 0.03% ).
In the bear affected by strychnine poisoning (bear number 6) extensive hemorrhages were found in several organs (heart, lungs, liver, kidney, spleen, stomach and intestine) often showing hemothorax, hemopericardium and hemoperitoneum. Microscopically vascular damage and diffuse necrosis in those organs were the most common findings. Strychnine was identified by chromatography from hair samples in a private company.
Cholangiocarcinoma was observed in the liver of an old female (bear number 13). A complete description of this case was already published . Briefly, microscopically liver tumor tissue showed tubular, acinar or pseudoglandular structures in the area facing a large cavity of necrosis with a thick trabecular growth pattern. Multiple small nodules were also present in the gall bladder. Metastatic encapsulated foci of cholangiocarcinoma were located in lung parenchyma, adrenal glands and articulation of the left elbow.
Bear number 19 showed hemorrhagic gastritis and diffuse hepatic and renal necrosis compatible with mushroom poisoning, likely due to ingestion of poisonous Amanita spp, although that could not be confirmed by toxicological analysis.