INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

WHEN EXPOSED TO SEVERE COLD, mammals increase their heat production to counter heat loss to maintain a constant body temperautre (Tb). If heat loss surpasses maximum heat production, hypothermia results and, unless aided by external heat sources, death ensues. To develop proper treatments that would enhance hypothermia survival, it is of importance to understand what physiological changes may lead to death during hypothermia. Although the failure of respiratory and/or cardiovascular functions, renal functions, acid-base regulation, and ion regulatory mechanisms have been suggested as critical (4, 15, 23), there is no consensus as to what physiological deterioration actually limits survival during prolonged severe hypothermia.

A major problem in many previous studies on hypothermia survival is the instability of the hypothermic animal preparation, which often only lasts a few hours, rendering the evaluation of limiting factors for survival difficult to establish (14). Because of this inherent difficulty, we have developed an animal model in which stable hypothermia could be established for prolonged periods (24-120 h) (16). This model thus offers the investigation of various time-related deterioration in fuctions during prolonged hypothermia. Our previous findings indicated that a steady decrease in both the turnover and oxidation of glucose (14) and a time-dependent decrease in venous PO₂ in conjunction with an increase in plasma lactate were consistently observed during the progression of prolonged hypothermia (19). Taken together, these observations suggest that a gradual failure in circulatory insufficiency appears to be the limiting factor for survival. Supporting this contention are the marked declines in arterial blood pressure, heart rate, and cardiac output that would result in a marked depression of tissue blood flow during hypothermia (30). To further complicate the situation, changes in rheological characteristics of the blood can additionally impede tissue perfusion. For example, increases in whole blood viscosity (21, 24) and aggregation of erythrocytes and platelets (3) in hypothermia will further exacerbate an already compromised microcirculation.

Whereas it may be difficult to enhance hypothermia survival by overcoming the suppressive effect of low Tb on cardiovascular parameters, it may be possible to institute remedial measures to minimize the deleterious changes in rheological characteristics of the blood. This seems to be the right approach as we have previously shown that survival in hypothermia can be significantly prolonged by reducing platelet aggregation with EGTA (18). In the present study, we attempted to improve microcirculation with plasma volume expanders of long (hetastarch)- or short-lasting (mannitol) characteristics, which, by effects of hemodilution, may counteract increased cell aggregation and thus prolong survival in hypothermia.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

All experimental procedures have received prior approval of the University of Alberta Animal Use Committee and followed the guidelines of the Canadian Council on Animal Care. Three- to six-month-old male Sprague-Dawley rats, housed individually at an ambient temperature (Ta) of 22 ± 1°C under a 12:12-h light-dark photoperiod, were fed lab chow rationed to maintain the body weight near 400 g and given water ad libitum. All rats were maintained under this condition for 1-2 wk before starting the experiment. As the results were combined from groups of rats acquired at different times of the year, all treatments (i.e., 3 and 20 h) were performed randomly within the same group to eliminate the possibilities of age and seasonal differences. Cannulation (right jugular vein) and induction of hypothermia were as previously described (16, 19). Briefly, after cannulation, the halothane-anaesthetized rat was transferred to a water-jacketed Plexiglas chamber at Ta 0°C under helium-oxygen (79% He; 21% O₂) in an open-flow system to facilitate heat loss. The halothane concentration was reduced by ~0.1% per each 1°C drop in Tb, and the anaesthetic was removed when Tb reached 25 ± 1°C. When Tb reached 22°C, the He-O₂ was replaced by normal air. In hypothermia, the Tb was clamped at 19°C by a computerized negative-feedback loop using the rectal temperature as the set-point for adjustment of Ta. The heart rate of the rat was monitored by electrocardiogram throughout the hypothermic bout. The endpoint of survival was determined by a lack of heartbeat and a difference between Tb and Ta of <0.3°C. Five milliliters of either saline, hetastarch (6%), or mannitol (6%) were infused continuously through the cathether in a flow rate of 41.67 µl/min (i.e., over 2 h) via a CMA100 pump (Carnegie Medicin, Sweden) at either 3 or 20 h after the Tb had reached 19°C. All results are expressed as means ± SE. Comparisons between saline control and treated animals at specific time points during hypothermia were performed using the Student's t-test, and significance was set at P < 0.05 unless otherwise stated. Linear regression and correlation were analyzed by using SPSS/PC+ program.

Blood samples (300 µl) were taken immediately after surgery, before (at either hour 3 or 20 after Tb had reached 19°C) and 2 h after either saline, hetastarch, or mannitol infusion (i.e., hours 5 and 22), and immediately after the rat had died. All rheological assays were carried out immediately after blood sampling at room temperature (22 ± 1°C). Hematocrit of the venous blood was measured using a 75-µl heparinized capillary tube centrifuged at 10,000 rpm for 5 min. Plasma osmolality was measured with a vapor pressure osmometer (Wescor 5100, Utah). Cell fragility was measured according to Chanarin (5), and the results were expressed as the median osmotic fragility, i.e., the concentration of saline at which 50% hemolysis occurred. In another series of experiments, blood samples (300 µl) were taken at the same time points for determination of plasma volume by dye dilution (13). Briefly, 0.1 ml of Texas Red covalently bound to albumin (0.5 mg/ml) was injected intravenously and the catheter flushed with 0.3 ml saline. To ascertain the validity of the measurement in hypothermia, the rate of disappearance of the fluorescent-labeled dye after injecting into the euthermic (Tb = 37 ± 1°C) and hypothermic (Tb = 19 ± 1°C) rats was compared and found to be similar within the first 20 min (Fig. 1). Subsequent samplings were done at min 10 after dye injection to minimize the temperature effect. As a validation, our measured plasma volume of the euthermic control rat (Tb = 37 ± 1°C) was 43.2 ± 3.3 ml/kg (n = 5), similar to those reported earlier using various dye-dilution methods (11, 17, 30).

View larger version (15K): [in this window] [in a new window]   Fig. 1.   Relative intensity of control [body temperature (Tb) = 37 ± 1°C] and hypothermia (Tb = 19 ± 1°C) rat plasma at various time periods after intravenous injection of Texas Red bound to albumin. Values are means ± SE for 5 rats.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

The average survival time of the control rats treated with saline intravenously at either hour 3 or 20 after hypothermia was ~23-27 h, about the same as we reported earlier (16, 19). As shown in Table 1, treating the animals with 6% hetastarch 3 h after hypothermia significantly (P < 0.05) increased survival time to 33.5 h. Interestingly, treating the rats with the same concentration of hetastarch at hour 20 after hypothermia also increased survival time from 26.7 to 32.5 h; however, this difference did not achieve statistical significance (P = 0.1; Table 1). In contrast, infusion of mannitol (6%), which is isosmotic to hetastarch, at either 3 or 20 h after hypothermia did not elicit any beneficial effects in prolonging survival time. In fact, treating the rats with mannitol at the later stage of hypothermia (i.e., 20 h after hypothermia) appeared to have a serious deleterious effect as two of six rats receiving the infusion died within 15-30 min thereafter.

Figure 2 shows the changes in hematocrit, plasma osmolality, plasma volume, and cell fragility in rats treated with hetastarch at hour 3 after hypothermia. In control rats, hematocrit did not change significantly until hour 5 in hypothermia and then increased significantly with time throughout the remainder of the hypothermic bout (Fig. 2A). Immediately after infusing hetastarch, hematocrit at hour 5 decreased significantly (P < 0.05) compared with the control value. The hematocrit value of the hetastarch-treated rats, although showing a progressive increase with time, remained significantly lower than that of the control rats at hour 20. The plasma osmolality also increased steadily throughout the hypothermic bout in the control rats (Fig. 3), similar to that observed with hematocrit. Treating the animal with hetastarch curtailed the rise in osmolality, resulting in a rightward shift of the curve (Fig. 3), which was marginally different from the controls (x coefficients: 1.42 ± 0.16 and 1.03 ± 0.14 for control and hetastarch, respectively, P = 0.09, n = 8). The plasma volume of the control rats decreased drastically (~30%) 3 h after hypothermia and then decreased continuously with time throughout the hypothermia bout (Fig. 2C). Treating the animal with hetastarch significantly (P < 0.05) reversed the decline in plasma volume at hour 5 (Fig. 2C). Although it did not achieve any significant difference, the plasma volume of the hetastarch-treated group remained higher than the control values at hour 20 (Fig. 2C). The fragility of the red blood cells also increased with time in the control rats (Fig. 2D). Treating the rats with hetastarch significantly (P < 0.05) reduced the fragility of the red blood cells at hour 20 (Fig. 2D). At expiration, all rheological parameters were about the same in all treatment groups (Fig. 2).

View larger version (46K): [in this window] [in a new window]   Fig. 2.   Changes in hematocrit (A), osmolality (B), plasma volume (C), and cell fragility (D) in hypothermic rats after intravenous pretreatment at hour 3 with either 5 ml of saline (open bars) or 6% hetastarch (hatched bars). Blood samples were collected immediately after surgery (B), before (hour 3) and after (hour 5) infusion, at hour 20 after hypothermia, and immediately after rat had expired (Ex). Values are means for 8 rats. SE of means are represented by vertical bars. Arrow indicates starting time of infusion. * Significantly different from control value at same time point (P < 0.05). Hr, hours.

View larger version (19K): [in this window] [in a new window]   Fig. 3.   Time course of changes in osmolality in hypothermic rats after intravenous pretreatment at hour 3 with either 5 ml of saline () or 6% hetastarch () or saline () or 6% mannitol (). Values are means for 6-8 rats. For clarity, SE of means, indicated by vertical bars, are shown only for initial and final values. Arrow indicates starting time of infusion.

The effects of infusing hetastarch at hour 20 after hypothermia on the rheological changes were summarized in Fig. 4. The patterns of rheological changes after infusing hetastarch at hour 20 were similar to those observed after infusing at hour 3; that is, slight reductions against the rising trend of hematocrit and cell fragility as well as a slight reversion of the decrease in plasma volume observed in the control rats. However, none of these changes achieved statistical significance.

View larger version (48K): [in this window] [in a new window]   Fig. 4.   Changes in hematocrit (A), osmolality (B), plasma volume (C), and cell fragility (D) in hypothermic rats after intravenous pretreatment at hour 20 with either 5 ml of saline (open bars) or 6% hetastarch (hatched bars). Blood samples were collected immediately after surgery (B), at 3 h after hypothermia, before (hour 20) and after (hour 22) infusion, and immediately after rat had expired (Ex). Values are means for 8 rats. SE of means are represented by vertical bars. Arrow indicates starting time of infusion.

Figure 5 shows the changes in hematocrit, osmolality, plasma volume, and cell fragility in rats treated with 6% mannitol at hour 3 after hypothermia. Within 2 h, the decrease in plasma volume was significantly reversed (P < 0.05; Fig. 5C); however, the values of both the control and the mannitol-treated rats were about the same at hour 20 (Fig. 5C). Similar to that observed in rats treated with hetastarch at the same time (hour 3), mannitol reduced the increases in hematocrit and osmolality immediately after perfusion (i.e., at hour 5). However, the increases in hematocrit and osmolality were about the same as the control rats at hour 20 after hypothermia (Figs. 5A and 3). Treating the animal with mannitol at hour 3 after hypothermia did not cause any change in cell fragility (Fig. 5D).

View larger version (45K): [in this window] [in a new window]   Fig. 5.   Changes in hematocrit (A), osmolality (B), plasma volume (C), and cell fragility (D) in hypothermic rats after intravenous pretreatment at hour 3 with either 5 ml of saline (open bars) or 6% mannitol (hatched bars). Blood samples were collected immediately after surgery (B), before (hour 3) and after (hour 5) infusion, at hour 20 after hypothermia, and immediately after rat had expired (Ex). Values are means for 6 rats. SE of means are represented by vertical bars. Arrow indicates starting time of infusion. * Significantly different from control value at same time point (P < 0.05).

Figure 6 shows the rheological changes in rats treated with mannitol at hour 20 after hypothermia. In contrast to those observed with mannitol given at hour 3, infusion of mannitol at hour 20 did not cause any change in all rheological parameters examined.

View larger version (50K): [in this window] [in a new window]   Fig. 6.   Changes in hematocrit (A), osmolality (B), plasma volume (C), and cell fragility (D) in hypothermic rats after intravenous pretreatment at hour 20 with either 5 ml of saline (open bars) or 6% mannitol (hatched bars). Blood samples were collected immediately after surgery (B), at hour 3 after hypothermia, before (hour 20) and after (hour 22) infusion, and immediately after rat had expired (Ex). Values are means for 6 rats. SE of means are represented by vertical bars. Arrow indicates starting time of infusion.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Although the survival time of the control rats receiving intravenous saline at either hour 3 or 20 after hypothermia varied from 23 to 27 h (Table 1), there was no significant difference among the four control groups, indicating that infusion of saline at different hypothermic stages did not affect the duration of survival. This suggestion is further supported from our previous findings that the survival time (~24 h) of rats receiving no treatment (16, 19) is about the same as those observed in the present study. Collectively, these results indicate the stability and reproducibility of the animal model used. Our observed time-dependent increases in hematocrit and osmolality and a decrease in plasma volume strongly suggest that deleterious changes in the rheological characteristics of the blood were limiting tissue microcirculation and probably hypothermia survival. One of the most striking observations in the present study was the drastic (30%) decrease in plasma volume within the first 3 h of hypothermia (Figs. 2C, 4C, 5C and 6C). This decrease is unlikely due to the effect of low temperature on the method of our plasma volume measurement, because we have validated the technique against temperature effect alone (Fig. 1). A similar decrease in plasma volume has also been reported in rats (30), hamsters (28), and dogs (25) 2-6 h after the animal had become hypothermic. Classically, cold-induced hypovolemia is explained by an increased diuresis due to inhibition of antidiuretic hormone release (20). However, recent studies have added the shift of water from vascular to interstitial space due to increased blood pressure (cold-induced peripheral vasoconstriction) as another factor in the reduction of plasma volume in early hypothermia (7, 31).

Associated with the decrease in plasma volume, progressive increases in hematocrit and osmolality were also observed in hypothermic rats. It has been reported that a 3% increase in hematocrit without changes in temperature can result in a 10% increase in blood viscosity and a decrease of Tb from 37°C to 25°C can further increase the blood viscosity by another 38% (6). Thus a steady increase in hematocrit during hypothermia will lead to a continuing and substantial increase in blood viscosity. When coupled with low blood volume and cardiac output (30), the increasing viscosity is likely to further impede microcirculation in hypothermia. In addition, the fragility of the red blood cells was observed to increase with time in hypothermia (Fig. 2D). A decrease in cell deformability, aided by cell debris from damaged cells, is expected to further exacerbate an already deteriorated microcirculation in hypothermia. It is thus reasonable to suspect that this progressive deterioration in tissue perfusion will lead to poor tissue oxygenation, cellular substrate delivery, and utilization (14, 19), leading to expiration of the animal.

As a countermeasure, it is hypothesized that the hypothermic animal should survive longer if the deleterious changes in microcirculation can be minimized. One of the simplest treatments in improving microcirculation is to dilute the whole blood by increasing plasma volume, which will decrease hematocrit and lower blood viscosity. However, infusion of fluid does not always reverse cold-induced hemoconcentration. It has been shown previously that saline infusion had minimal lasting effects and did not enhance cardiovascular recovery from hypothermia (11, 25). This may be due to the rapid clearence of saline from the intravascular space. Conversely, infusing the hypothermic dog with a 10% low molecular weight dextran increased plasma volume and decreased the sludging of blood (11). However, the duration of survival was not addressed in this study. Furthermore, dextran may not be a suitable option for enhancing hypothermia survival because of its relatively short-lasting effect (2-4 h) (26) and its antigenicity due to its bacterial origin (22). In the present study, we used hetastarch, another volume expander, but with much longer duration of action (>24 h) (26) and excellent clinical safety record (25) to enhance hypothermia survival. Treating the animal with hetastarch at the early stage of a hypothermic bout (3 h after hypothermia) significantly prolonged the survival time by almost 10 h (41%). The major difference between the hetastarch- vs. saline-treated rats is the reduction of the observed increase in hematocrit at hour 5 and beyond (Fig. 2A). Because of the long-lasting effect of hetastarch, the hematocrit remained significantly lower than the control value even after 20 h in hypothermia (Fig. 2A). The importance of maintaining a near-normal hematocrit to survive hypothermia is demonstrated in our previous study that in the Richardson's ground squirrels, in which hematocrit does not change during most of a hypothermia bout, can survive at a Tb of 7°C approximately three times as long as can the rat at a Tb of 19°C (19). In addition to delaying the rise of hematocrit, hetastarch also reduced the increases in cell fragility (Fig. 2D). As hetastarch has been shown to improve erythrocyte deformability and prevent erythrocyte aggregation both in vitro and in vivo (9, 12, 29), it may have some direct effect in minimizing cell fragility. Furthermore, hetastarch may also retard the increases in hematocrit and cell fragility by better maintaining the plasma volume than saline.

It is of great interest to note that when survival time is correlated with plasma volume taken at hour 20, having received various treatments at hour 3, a positive correlation (r² = 0.61, P < 0.001, n = 28) was observed (Fig. 7A). In fact, all animals survived >30 h typically associated with a plasma volume >20 ml/kg, and animals that received hetastarch survived the longest. A similar conclusion can also be drawn by comparing the survival time with plasma volume obtained at hour 22 having received various treatments at hour 20 (Fig. 7B, r² = 0.82, P < 0.001, n = 28). Again, a plasma volume >20 ml/kg appears to segregate the long survivors versus the short survivors, and those receiving hetastarch survived the longest (Fig. 7B). Thus regardless of the timing of the hetastarch treatment, the critical consideration appears to be whether a critical plasma volume (~20 ml/kg) can be maintained to ensure better survival. Indeed, the few individuals that received either saline or mannitol and were able to maintain a high plasma volume, also survived longer (Fig. 7, A and B).

View larger version (16K): [in this window] [in a new window]   Fig. 7.   Scatter diagram showing relationship between ultimate survival time and plasma volume taken at either hour 20 (A; for animals with various treatments at hour 3; r2 = 0.61, P < 0.001, n = 28) or hour 22 (B; for animal with various treatments at hour 20; r2 = 0.82, P < 0.001, n = 28) after hypothermia.

Although it is difficult to quantify the proportional importance of hetastarch on enhancing rheological characteristics versus plasma volume per se, the longer survival in hypothermia attested to the functional benefit of this treatment. The patterns of rheological changes after treating the animal with hetastarch at hour 20 were similar to those observed at hour 3, but to a lesser extent (Fig. 4). Consequently, the survival time increased by ~6 h. The inability to achieve statistical significance for this treatment could be due to the longer survival time of the control group (26.7 vs. 23.8 h for the hour 3 control group). However, it is also possible that 32-33 h is the optimal survival time that can be achieved by the concentration of hetastarch used in the present study. Collectively, our results demonstrated that treating the animal with hetastarch can prolong hypothermia survival by delaying the progressive deterioration of microcirculation.

Mannitol (6%) has also been shown to increase plasma volume after administration (1, 33). When given at the early stage of hypothermia, similar effects on rheological changes as those observed with hetastarch were seen (Fig. 5). However, in contrast to those observed with hetastarch, the changes in hematocrit, plasma volume, and osmolality after mannitol were short-lived and by hour 20, became indistinguishable as those seen in the controls. Consequently, the mannitol-treated rats survived about the same length of time as the saline controls. The failure of mannitol to maintain a long-lasting effect could be due to its relative short elimination half-life (<2 h) (2, 8). Thus a critical consideration in using plasma expanders to enhance hypothermia survival is the duration of its action; a longer-acting one is significantly more beneficial than a shorter-acting one. Unexpectedly, infusing mannitol (6%) at the later stage of hypothermia appeared to have a serious deleterious effect as two of six animals died shortly after infusion. Although not certain why this should be so, the osmotic effect of mannitol, which can cause water to be drawn from erythrocyte (8, 32) and further decrease cell deformability, could exacerbate an already impeded microcirculation. In addition, mannitol-induced circulatory overload due to expansion of the vascular volume has been reported to cause pulmonary edema and congestive heart failure in patients with diminished cardiac reserve (10, 32). As the cardiac function is markedly reduced in the later stage of hypothermia, these combined adverse effects could further impair survival of the animal.

In conclusion, during hypothermia, a high hematocrit and increased blood viscosity, coupled with a concomitant decrease in cardiac output and an intense peripheral vasoconstriction, are conducive for the aggregation of blood cells in capillary beds leading to poor tissue perfusion. In particular, hemoconcentration may reduce cerebral blood flow and compound the depressant effects of low Tb on central regulation of cardiovascular and respiratory functions. Because only hetastarch, but not mannitol, prolongs survival at low Tb, it is of importance to employ a long-lasting volume expander to maintain the patency of microcirculation in hypothermia. Because of the excellent clinical safety of hetastarch, our observed results may encourage further studies in improving microcirculation and prolonging long-term clinical hypothermia survival.

Perspectives