Abstract
In safety studies, distinguishing subtle drug-associated tissue changes from spontaneous or incidental changes and postmortem or tissue processing induced artifactual changes is critical but often very challenging. A number of factors have been associated with the formation of non-drug-induced changes in tissues. These include the time between death, necropsy, and fixation as well as external factors (e.g., ambient temperature, humidity, light intensity, and trauma) and internal factors (e.g., physical or nutritional status and body temperature) (Trowell, 1946; Nunley et al., 1972; Splitter and McGavin, 1974; Sykes et al., 1976; Micozzi, 1986; Farnell, 1991; Kimura and Abe, 1994; Li et al., 2003). In the liver, it has been reported that anoxia and high intra-hepatic blood pressure are two critical factors in the formation of postmortem hepatocytic vacuolation in rats and it was shown that the vacuoles were formed by intracytoplasmic influx of plasma secondary to sinusoidal congestion (Sykes et al., 1976; Li et al., 2003).
We recently observed hepatocytic vacuolation similar to that reported in rats in a cynomolgus monkey model of extra-corporeal blood circulation (ECC). In the required procedures for this model, each animal underwent anesthesia, surgical placement of catheters and cannulae, and heparinization to achieve activated clot time (ACT) 2–3 times baseline followed by 2 hours of ECC. Following the 2-hour ECC, protamine was administered to reverse the anticoagulation by heparin. When heparin effects were reversed, test material or placebo was administered and each animal was observed for an additional 6 hours while remaining under anesthesia. At the end of the 6-hour observation period, each animal was euthanized under deep anesthesia via exsanguination, perfused by gravity flow with physiological saline followed by 10% neutral buffered formalin and necropsied. Tissues were collected and preserved in 10% neutral buffered formalin, routinely processed to paraffin, then sectioned and stained with hematoxylin and eosin.
Microscopically, the liver from each animal had diffuse hepatocellular vacuolation that was characterized by the presence of small, multivesicular intracellular vacuoles. This vacuolation, commonly associated with lipid metabolism, varied as would be expected depending on the metabolic state at the time of death. However, many of the hepatocytes contained a single, large perinuclear vacuole that frequently distorted the nucleus into a crescent at the margin of the vacuole. These vacuoles varied from being void of content to containing homogenous basophilic material (Figure 1A and B).
In this study, we believe the perinuclear hepatocytic vacuolation was the result of tissue anoxia and high intrahepatic blood pressure. During the 2-hour ECC procedure and during the 6-hour observation that followed, each animal was exposed to variable states of anoxia as well as fluctuations in blood pressure. At the end of the observation period, each animal was disconnected from the respirator, transported from the surgical suite to the necropsy room, and exsanguinated. The time required to accomplish these tasks as well as the amount of time required to set up and perfuse with 2 separate fluids was variable among animals. This variation may have been reflected in the variable incidence and severity of the vacuolation observed among the animals. Observed in control and treated animals, the hepatocyte vacuolation was often accompanied by severely distended vascular spaces suggestive of increased vascular pressure in this tissue.
Several investigations of the origins of these hepatocytic vacuoles suggest that not only their presence but the size and how rapidly they form is dependent on the degree of anoxia and the intrahepatic blood pressure at the time of death, both of which may be elevated if the animal is not exsanguinated and necropsied immediately upon death (Trowell, 1946; Sykes et al., 1976; Kimura and Abe, 1994; Li et al., 2003). In our study, increased pressure applied during whole body perfusion and rapid tissue fixation possibly contributed to or exacerbated the formation and appearance of these vacuoles. The absence of any obvious hepatocellular changes indicating cell degeneration or necrosis in the presence of these vacuoles further suggests that their formation is an acute event occurring at or near the time of death.