Many ectotherms regularly experience considerable short-term variations in environmental temperature, which affects their body temperature. Here we investigate the cardiovascular responses to a stepwise acute temperature increase from 10 to 13 and 16°C in rainbow trout (Oncorhynchus mykiss). Cardiac output increased by 20 and 31% at 13 and 16°C, respectively. This increase was entirely mediated by an increased heart rate (fH), whereas stroke volume (SV) decreased significantly by 20% at 16°C. The mean circulatory filling pressure (MCFP), a measure of venous capacitance, increased with temperature. Central venous pressure (Pven) did not change, whereas the pressure gradient for venous return (MCFP-Pven) was significantly increased at both 13 and 16°C. Blood volume, as measured by the dilution of 51Cr-labeled red blood cells, was temperature insensitive in both intact and splenectomized trout. This study demonstrates that venous capacitance in trout decreases, but cardiac filling pressure as estimated by Pven does not change when cardiac output increases during an acute temperature increase. SV was compromised as fH increased with temperature. The decreased capacitance likely serves to prevent passive pooling of blood in the venous periphery and to maintain cardiac filling pressure and a favorable pressure gradient for venous return.
- cardiac performance
- cardiac preload
- stroke volume
- venous capacitance
many fish live in a variable water environment where temperature may fluctuate considerably, both on a seasonal scale and on a daily basis. For example, pelagic juvenile sockeye salmon (Oncorhynchus nerka) undertake diel vertical migrations, exposing them to temperature differences as large as 8–10°C (30), and stream-dwelling rainbow trout (O. mykiss) in the high-desert region of southeastern Oregon are subjected to diel fluctuations in stream water temperatures in excess of 10°C during the summer months (40).
Since most fish breathe water, and oxygen uptake takes place via a highly efficient countercurrent arrangement between blood and water at the gills, changes in ambient temperature are rapidly mirrored by the body temperature of the fish (13, 39, 52). Temperature is positively correlated with aerobic metabolic rate in fish and, perhaps not surprisingly, acute temperature changes are associated with several cardiovascular changes. Cardiac output often increases to meet the increased metabolic demand following an acute increase in ambient water temperature (6, 10, 18, 19, 22, 28, 29, 32, 50). In a recent study on Atlantic cod (Gadus morhua) from Newfoundland, routine cardiac output increased continuously from 21.5 ml/min at 10°C to a maximum value of 52.6 ml/min at ∼21°C, when temperature was gradually increased by ∼1.7°C/h (22). This increase in cardiac output was entirely mediated by an increased heart rate (fH) as stroke volume (SV) did not change. Only when the fish lost equilibrium at its critical thermal maximum (22.2°C) and displayed bradycardia and severe cardiac arrhythmias, did SV increase (22). Increased cardiac output through tachycardia with maintained SV during acute temperature increase has also been documented in lingcod (Ophiodon elongates) (50) and winter flounder (Pseudopleuronectes americanus) (10). However, in another study on cod from the North Sea, fH increased exponentially, whereas cardiac output leveled off with increasing temperature, indicating that SV was compromised in these animals (29). Similarly, in trout, SV appeared to be unchanged when temperature increased from 12 to 15°C, but tended to drop as temperature was increased further (6). Reduced SV with increasing fH during acute temperature increase has also been reported for yellowfin tuna (Thunnus albacares) and various Antarctic species (2, 28, 32). Whether these studies represent true interspecific differences or simply differences in experimental protocols is unknown.
Isolated perfused fish hearts respond to increased temperature with tachycardia, and routine SV may decrease when the diastolic filling time and the force of contraction is reduced at high temperatures (4, 5, 18, 19, 23, 47). This indicates that additional mechanisms may operate in vivo in some species to maintain SV as fH increases. The reduced SV can be compensated for by raising the input pressure, which is equivalent to the central venous blood pressure (Pven) in vivo, suggesting that an increased cardiac filling pressure at high temperatures may be one possible mechanism by which SV is maintained (4, 5, 19, 23). In the intact animal, this could be accomplished through constriction of the venous capacitance vasculature, which mobilizes blood to the central venous compartment and increases cardiac filling pressure (37, 41).
Vascular capacitance is the relationship between contained blood volume (Vbl and pressure, and depends on vascular compliance and smooth muscle tone (37, 41). At least in mammals, compliance and tone are controlled by apparently unrelated mechanisms and can change independently (37). The venous circulation contains the main part of the circulating blood volume, and it is much more compliant than the arterial circulation. Thus, changes in total vascular capacitance are more or less exclusively related to changes in the venous vasculature. This has been assumed to hold true for fish, as well (11, 12, 26, 36, 42–46, 55, 56). Vascular capacitance can be estimated in vivo by measuring the mean circulatory filling pressure (MCFP), which is the venous pressure during a transient circulatory arrest. If total blood volume is unchanged, an increased MCFP is an indication of a reduced capacitance due to an increased tone, a decreased compliance, or both (37, 38, 41). Furthermore, Poiseuille's law describes the relationship between pressure and flow in tubes: flow = πΔPr4/8 μl, where π = 3.14, ΔP = the pressure difference, r = vessel radius, μ = viscosity, and l = vessel length. During exercise, active changes in venous tone and compliance are believed to be important to prevent blood from passively pooling (because ΔP increases) in the compliant veins as cardiac output increases (45, 46). The venous changes observed during exercise in fish are also likely to be important during an increase in cardiac output with high ambient temperatures, although a decreased blood viscosity with increased temperature and blood velocity may serve to offset some of this effect (8, 31). Moreover, in mammals, temperature has a direct effect on the vasculature, such that hypothermia decreases compliance (24, 48). Thus, if compliance is temperature sensitive in ectothermic animals as well, there seems to be a conflict between an increased cardiac output and the expected increase in compliance at high temperatures. Altogether, these data suggest that venous capacitance must be actively reduced during an acute temperature increase in ectotherms to prevent blood from pooling in the venous periphery, and possibly, also to increase cardiac filling pressure. However, to our knowledge, no experimental evidence is presently available to either support or refute these ideas.
In the present study, cardiovascular variables, including Pven and MCFP, were measured in rainbow trout during a stepwise increase in temperature to investigate the role of the venous vasculature in the cardiovascular responses to an acute temperature increase in an ectothermic vertebrate.
MATERIALS AND METHODS
Rainbow trout, Oncorhynchus mykiss, with a size range of 290–1,060 g were purchased from a local hatchery (Antens Laxodling). Fish were kept in tanks supplied with recirculating water at 10°C. Newly arrived fish were allowed an acclimation period of at least 3.5 wk prior to any experimentation. Fish were held at a 12:12-h light-dark photoperiod and where fed a maintenance diet of commercial trout pellets. Swedish laws on animal care and ethical permits on physiological experimentation on fish covered all procedures reported here (no. 64/2004).
Fish were netted from the holding tanks, anesthetized in water containing NaHCO3 buffered MS-222 (300 and 150 mg/l, respectively) and placed on a surgery table covered with wet foam rubber. The gills were continuously irrigated via the mouth with aerated water at ∼10°C, containing NaHCO3 buffered MS-222 (150 and 75 mg/l, respectively). The dorsal aorta was cannulated through the roof of the buccal cavity with tapered polyethylene (PE)-50 catheter. By using a sharpened steel wire guide, the sinus venosus/ductus of Cuvier was nonocclusively cannulated with a PE-50 catheter through the skin (33). To optimize patency of the venous catheter, 2–3 side holes were cut in the first 5–10 mm. In addition to low and pulsating pressures and easy withdrawal of blood, correct placement of the venous catheter was verified post mortem in several specimens. Both catheters were filled with heparinized (∼100 IU/ml) saline (0.9%) and secured to the skin with several 3-0 silk sutures. The ventral aorta was exposed at the base of the first gill arch by using sharp and blunt dissection. Care was taken not to damage the pericardium. A combined Doppler-flow/occlusion probe was fitted around the ventral aorta. The probe was custom-made from Perspex and equipped with an inflatable latex balloon that allowed for recordings of cardiac output and transient occlusion of blood flow when the MCFP was measured. This type of probe has been described in detail elsewhere (43). The probe lead was sutured to the skin with several 3-0 silk sutures. Following surgery, fish were revived in fresh water, transported to opaque experimental chambers, and left to recover for at least 24 h prior to experimentation.
Water to the experimental chambers was supplied from the departmental recirculating water system and was directed via a titan countercurrent heat exchanger (Tranter, Vänersborg, Sweden) before emptying into the experimental chambers. A counterflow to the heat exchanger was provided by a pump submerged in an external water container equipped with immersion heaters. By connecting heaters to the power supply, temperature in the experimental chambers could be increased to the desired temperature. When cardiovascular variables were stable, routine values at 10°C (10.5 ± 0.2) were first recorded before the temperature was increased to 13 (13.1 ± 0.1) and 16°C (15.9 ± 0.1). This temperature interval was chosen since it is well within the normal thermal range of these fish and was considered unlikely to evoke a general thermal stress response of the animals. Finally, the power supply to the heating was turned off, and cardiovascular variables were again recorded at 10°C. After a change in temperature, the new temperature was attained in less than ∼15 min. MCFP was measured at the end of each recording period by occluding the ventral aorta for ∼10 s. MCFP was taken as the central venous pressure between the 5th and 7th s of the occlusion. Longer occlusion times may overestimate MCFP, since a baroreflex-mediated constriction of the venous vasculature is initiated (43). Between 45 and 90 min at each temperature was allowed for cardiovascular variables to stabilize before recordings began. This period is well beyond the time necessary for body temperature in similarly sized fish to equilibrate with ambient water temperature (13, 39).
Acquisition and Calculation of Cardiovascular Data
Central venous (Pven) and dorsal aortic (Pda
Relative changes for cardiac output (Q), SV, and systemic resistance (Rsys) are reported with the routine value at 10°C set to 100%. The fH was derived from pulsatile pressure traces, and SV was calculated as SV = Q/fH. Rsys was calculated from the pressure drop across the circulation as Rsys = (Pda−Pven)/Q. Although it has not been verified experimentally for fish, MCFP was assumed to give a good approximation of the pressure in the peripheral small veins and venules, as is the case in mammals (37). Hence, venous vascular resistance (Rven) was calculated as Rven = (MCFP−Pven)/Q, and the pressure gradient for venous return (ΔPven) was calculated as ΔPven = MCFP − Pven. For the calculations of SV and resistances, raw data for cardiac output were used before subsequent transformation to relative values.
Blood Volume Determination
Total circulating Vbl was determined at 10 and 16°C using 51Cr-labeled red blood cells (RBCs). The experimental protocol was slightly modified from those described by Gingerich et al. (21) and Duff and Olson (17). Two groups of fish, cannulated in the dorsal aorta as described above, were used in these experiments. Fish in one of the groups were also splenectomized, as previous studies have demonstrated that 51Cr-labeled RBC are sequestered by the spleen, which may lead to an overestimation of the total blood volume (16). The spleen was removed by making an incision in the ventral midline starting rostral to the pelvic fins and running 3 cm rostrally. The vessels running to and from the spleen were ligated before the spleen was dissected free and removed. The incision was closed with uninterrupted 3-0 silk sutures, and the fish were recovered as described above.
On the day of experimentation, ∼1 ml/kg body mass of blood was withdrawn from each fish and spun at 5,500 rpm for 3 min. The plasma was reinjected into the respective fish and the RBC were washed in ∼10 volumes of trout Ringer (TR; 27). RBC were resuspended in ∼10 volumes of TR and 2 μCi of 51Cr as sodium chromate (Amersham, Buckinghamshire, UK) was added to the suspension. The radioactive cell suspension was incubated and gently agitated for 4 h at 10°C. Following incubation, the 51Cr-labeled cells were washed 5 times in ∼10 volumes of TR and suspended to their approximately initial volume and hematocrit (Hct) (i.e., ∼1 ml/kg body mass; mean Hct ∼23%). Duplicate Hct samples and duplicate 50-μl samples for measurement of total activity were taken before the cell suspension was injected into the respective fish. To clear the catheter dead space, 0.5 ml of TR was injected after the RBC suspension. The exact volume of the injected RBC suspension was determined by weighing the syringe before and after injection. For both 10 and 16°C, a circulation time of 30 min was used as this is assumed to be sufficient for complete intravascular mixing in trout (17). With this circulation time, blood volume could be relatively quickly measured after the acute temperature change, thus coinciding with the period when cardiovascular variables were recorded. Following circulation, blood was withdrawn and duplicate Hct samples and 50-μl duplicates for subsequent determination of total activity were taken from the final circulating blood. Activity of samples was determined by using a Wallac Wizard 1470 Automatic gamma counter (Perkin Elmer, Wellesley, MA). Red cell space for whole blood (RCSb) was calculated as RCSb = Vinj*SAinj/SA30, where Vinj is the volume of injected packed RBC, SAinj is the specific activity (i.e., cpm/ml) of injected RBC, and SA30 is the specific activity of the RBC in the final circulating blood after 30 min. Total Vbl was calculated as Vbl = RCSb/Hct, where Hct is the Hct of the final circulating blood. Vbl was determined at 10 and 16°C in individual groups of fish. Essentially the same protocol as that described for the cardiovascular recordings was followed. In other words, for the 16°C measurement, the temperature was first increased to 13°C and maintained at that temperature for 1 h before it was increased to 16°C and maintained for 1 h. The sample at 16°C was taken at the end of the 1-h period to coincide with the period when cardiovascular recordings were made.
Statistical comparisons for percentage values were made using raw data but are for clarity presented as relative changes from the initial routine value at 10°C set to 100%. Cardiovascular data were analyzed using repeated-measures ANOVA followed by Dunnett's post hoc test to investigate significant changes from the baseline with increasing temperature. Hematological variables at 10 and 16°C were compared with a two-tailed Mann-Whitney U-test. Values were considered statistically significant at P < 0.05.
Cardiovascular Responses to Acute Temperature Increase
Cardiac output increased by 20 and 31% with a stepwise increase in temperature to 13 and 16°C, respectively. The increase in cardiac output was entirely mediated through tachycardia as fH increased from 46.8 ± 3.9 beats/min at 10°C to 61.6 ± 5.0 beats/min at 13°C and 75.6 ± 5.8 beats/min at 16°C, whereas SV tended to decrease with increasing temperature. At 16°C, SV had decreased significantly by 20% (Fig. 1 and Table 1). At 10°C, Pda was 3.2 ± 0.1 kPa, and the increase in cardiac output with temperature resulted in a moderate hypertension that was significant at 16°C (3.5 ± 0.1 kPa). If anything, Rsys showed a tendency to drop with increasing temperature, but this was not statistically significant (Table 1). The effect of temperature changes on cardiac output and fH had the respective Q10 values of 1.3 to 1.8 and 2.0 to 2.5 (Table 2).
At 10°C, Pven and MCFP were 0.01 ± 0.02 and 0.16 ± 0.03 kPa, respectively, resulting in a ΔPven of 0.15 ± 0.03 kPa. Pven did not change with increasing temperature, but MCFP increased significantly to 0.21 ± 0.05 and 0.23 ± 0.04 kPa at 13 and 16°C, respectively. The increase in MCFP resulted in the respective increase in ΔPven to 0.20 ± 0.04 and 0.21 ± 0.04 kPa. Rven did not change with increasing temperature (Fig. 2 and Table 1). The Q10 values for the changes in MCFP and ΔPven ranged between 1.3 to 2.5 and 1.2 to 2.6, respectively (Table 2).
Blood Volume Responses to Acute Temperature Increase
Total circulating blood volume was 38.0 ± 3.1 and 31.5 ± 1.5 ml/kg body wt at 10°C in the control and the splenectomized groups, respectively. Acute temperature increase to 16°C had no significant effect on total Vbl in any of the two groups (Fig. 3).
Hematocrit was 21.1 ± 2.2% in the control group at 10°C and increased significantly to 27.3 ± 1.4% at 16°C. In the splenectomized group, temperature had no significant effect on Hct (Fig. 3).
General Cardiovascular Responses
As far as we are aware, this is the first study to investigate the responses of the venous vasculature to acute temperature changes in an ectothermic animal. Fish typically remained calm in the holding tanks throughout the experiments. Startle responses as indicated by sudden bradycardia and/or flight responses were seldom observed, indicating that the temperature regimen used in this study did not evoke an agitated state of the fish, which otherwise may have interfered with the low pressure recordings. Following the acute temperature challenge, all cardiovascular variables returned to baseline values within 1 h during the recovery period at 10°C, demonstrating that the cardiovascular temperature responses were fully reversible (Figs. 1 and 2 and Table 1).
Previous studies have revealed that when fish are subjected to changes in ambient water temperature, a number of cardiovascular responses are elicited. In accordance with our results, cardiac output typically increases, although the exact mechanism behind this response seems to differ somewhat between species and experimental conditions. In the present study on trout, cardiac output increased through tachycardia, whereas SV tended to drop with increasing temperature (Fig. 1 and Table 1). The reduced SV was likely due to the reduced cardiac filling time as fH increased (1, 20). This is consistent with another study on cardiac performance in trout (6). Reduced SV at high temperatures has also been observed in other species (2, 28, 32), while species like Atlantic cod, lingcod, and winter flounder display a maintained SV as fH increases with increasing temperature (10, 22, 50). It cannot be excluded, however, that the cardiovascular response of one species may be different depending on where in the species thermal window the study is carried out (29). Cardiac preload probably did not increase in trout when temperature was increased as judged by the unaltered Pven in the present study. Future studies will have to reveal whether species that have the capability to maintain SV, do so by increasing cardiac filling pressure. The only previous report on venous pressures at different temperatures that we are aware of is provided in the study on winter flounder by Cech et al. (10). In that study, pressure in the caudal vein did not increase although SV was unchanged as cardiac output and fH increased with temperature. This may suggest that the winter flounder has a cardiac end-systolic reserve that can be mobilized to maintain SV when the cardiac filling time is reduced. Pda increased with temperature in the present study (Table 1), which is consistent with previous studies on trout (25) and eel (Anguilla japonica) (51), but different from the winter flounder, where pressure in the caudal artery does not change (10). Our study demonstrates that the arterial pressure increase in trout is the result of increased cardiac output and not by an increased vascular resistance.
Changes in Vascular Capacitance
An increased MCFP can be interpreted as a reduced vascular capacitance if blood volume does not change (41). Given that blood volume was unaltered in trout with increasing temperature (for further discussion, see Fig. 3 and Blood Volume and Hematocrit Responses), this study clearly demonstrates that venous capacitance decreases by means of active vascular changes in trout as temperature increases. The function of this response is probably to prevent blood from passively pooling in the peripheral (venous) circulation as pressure increases due to the increased blood flow according to Poiseuille's law. This mechanism is likely also important to ensure that a favorable pressure gradient for venous return is maintained between the peripheral circulation and the heart with increasing temperature. Furthermore, if vascular compliance passively increases with increasing temperature in ectothermic animals as it does in mammals (24, 48), the importance of a compensatory active reduction in vascular capacitance at high temperatures becomes even clearer. Without a compensatory increase in vascular tone, a passive increase in compliance with temperature would result in a conflict between one of the most important factors dictating cardiac filling pressure and the return of blood to the heart (i.e., venous capacitance) and the need to increase cardiac output with temperature. This mismatch between the factors dictating venous return and cardiac output would presumably become more severe, the greater the acute temperature increase is. The direct effect of temperature on vascular compliance in ectothermic animals certainly warrants further investigation.
Although central venous pressure did not increase in the present study, the decreased venous capacitance is probably important to at least maintain Pven in the trout with increasing temperature. Without adjustments in the venous microcirculation, Pven would presumably drop as fH increases, and this would reduce SV even more and further compromise the heart's ability to increase cardiac output at high temperatures (1, 20). From Fig. 2, it seems that the increase in MCFP levels off at 16°C, and this is also suggested by the Q10 of 2.5 for the change in MCFP between 10 and 13°C, but only 1.3 for the 13–16°C range (Table 2). This could reflect an attenuated need for increasing vascular tone to prevent passive flow-induced blood pooling at higher temperatures, as blood viscosity decreases with increasing temperature and blood flow velocity (8, 31).
Adrenergic control systems have a considerable influence on venous capacitance in fish (42–44, 56). We have previously demonstrated that reduced venous capacitance through increased α-adrenergic tone is important during exercise, which is another type of cardiovascular challenge resulting in tachycardia and increased cardiac output in fish (45, 46). Preliminary experiments with α-adrenergic blockade at different temperatures in the present study proved to be very difficult as both arterial and venous blood pressures became extremely variable, displaying cyclical fluctuations when temperature was increased after blockade (data not shown). Due to these, so far unresolved problems, these experiments were not pursued further. Another possible control mechanism is that of local myogenic control mediated by increased vessel wall tension as blood flow increases. Some mammalian veins display a myogenic response (3, 14, 15), but the importance of this mechanism in fish has yet to be tested.
Blood Volume and Hematocrit Responses
Blood volume did clearly not increase with temperature (Fig. 3), demonstrating that the observed changes in MCFP were mediated by vascular capacitance changes as pointed out above. An unaltered blood volume with acute temperature increase is in agreement with a previous study on trout using dilution of Evans blue dye (34). Temperature could theoretically influence blood volume by several different mechanisms. For example, acute elevation of water temperature increases drinking rate in some fish (9, 51), and similar to the situation during exercise, the functional gill surface area probably increases with temperature to optimize oxygen uptake (49, 54). These mechanisms could, at least transiently, lead to intravascular fluid uptake in freshwater fish. However, the increase in arterial blood pressure and accumulation of intracellular metabolites, which could be associated with increased temperature, would instead lead to a net loss of intravascular fluid (53).
The method with 51Cr-labeled RBCs was used since other methods, such as plasma dyes, have been found to overestimate blood volume in fish (35). However, a potential source of error with the 51Cr method is that labeled RBCs are sequestered by the spleen, and this may lead to a slight overestimation of blood volume, especially when the circulation time is long (16). Thus, to control for potential differences is sequestration rate by the spleen at different temperatures, experiments were also undertaken in splenectomized fish. In intact trout, routine blood volume at 10°C was 38.0 ml/kg body wt. This is well within the range of previously reported values for salmonids in studies using similar methods (7, 35). In splenectomized fish, Vbl was 31.5 ml/kg body wt at 10°C and was similarly unaffected by temperature (Fig. 3).
Cardiac output only increased by 31% between 10 and 16°C in the present study (Table 1), resulting in a Q10 value for cardiac output of 1.6 (Table 2). This indicates that the arteriovenous O2 difference increased due to increased oxygen extraction or that the O2-carrying capacity of the blood increased at the higher temperature. The latter possibility is supported by the finding that hematocrit was significantly higher at 16°C compared with 10°C in control fish (Fig. 3). The increased Hct was most likely the effect of splenic release of erythrocytes and not by RBC swelling, as the hemoconcentration was absent in the splenectomized group.
In conclusion, in their daily activities, fish are likely exposed to acute changes in temperature that affects their body temperature and metabolism. This implies that a number of rapid cardiovascular adjustments must be made to cope with the altered metabolic demands. In this study, we demonstrate that an active reduction of venous capacitance occurs in the trout during an acute increase in temperature. This is likely a highly important response that serves to provide a favorable pressure gradient for venous return, which enables the heart to increase cardiac output to match oxygen supply with oxygen demand at different temperatures. It probably also counteracts any potential passive effects of temperature on vascular compliance, which otherwise would severely compromise the heart's ability to increase cardiac output at high temperatures.
Financial support was received from the Swedish Research Council (to M. Axelsson). E. Sandblom received financial support from Helge Ax:son Johnson's Foundation, Knut and Alice Wallenberg's Foundation, Paul and Marie Berghaus Foundation and Hierta-Retzius Foundation.
We are grateful to Dr. Kristina Sundell and Barbro Egnér for valuable help and suggestions regarding the work with 51Cr. Henrik Seth and Albin Gräns are also acknowledged for practical help and valuable input during the course of this study.
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2007 the American Physiological Society