Materials and Methods

Healthy mixed-breed dogs (18 males and 13 females) that were 2 to 5 years old were used in the study. The mean ± SE weight of the dogs was 26 ± 2 kg, and each was in good body condition (body score, 3/4). All animals received humane care in compliance with the Principles of Laboratory Animal Care, formulated by the National Society for Medical Research and the Guide for the Care and Use of Laboratory Animals prepared by the National Academy of Sciences.¹² Experimental protocols were reviewed and approved by the Ethics Committee for Animal Research of the Universidad de Cordoba.

After withholding of food for 12 hours, dogs were premedicated with ketamine^a (7.5 mg/kg, IM), fentanyl^b (75 μg/kg, IM), and droperidol^c (375 μg/kg, IM). In each dog, the left femoral artery, the right jugular vein, and both cephalic veins were cannulated. A bolus of sodium thiopental^d (12.5 mg/kg) was then given IV to induce anesthesia and facilitate tracheal intubation. After placement of the endotracheal tube, ventilation was controlled with a variable rate respirator.^e The oxygen concentration (FIO₂) in the inspired gas mixture was set at 27% to maintain PaO₂ at 95 to 105 mm Hg. Ventilation was set at 12 breaths/min with a tidal volume of 15 mL/kg to achieve and maintain PaCO₂ of 30 to 40 mm Hg. Anesthesia was maintained during the experiment via intermittent IV administration of fentanyl^b (2 μg/kg), midazolam^f (0.25 mg/kg), and pancuronium bromide^g (0.1 mg/kg). The arterial catheter was used for blood sampling and blood pressure monitoring.^h The cephalic vein catheters were used to infuse HCl,ⁱ EDTA,^j and 5% dextrose-saline (0.45% NaCl) solution. The jugular vein catheter was used to infuse anesthetic agents and, when required, magnesium sulfate^k to prevent development of hypomagnesemia during the EDTA infusion.

Dogs were randomly allocated to 1 of 4 groups. The experimental protocols for each group were as follows: normal acid-base status maintained and hypocalcemia induced after a 60-minute period during which dogs remained normocalcemic (group I), metabolic acidosis induced and hypocalcemia induced after a 60-minute period during which dogs remained normocalcemic (group II), metabolic acidosis induced and dogs allowed to become hypercalcemic (group III), and treatment identical to that of group II with parathyroidectomy performed during the period of hypocalcemia (group IV). In all groups, baseline blood samples were obtained at 15, 10, and 5 minutes prior to the start of the experiment and immediately before the start of the experiment (time 0). The duration of the experimental period was 120 minutes in groups I and II and 60 minutes in group III; in group IV, the experimental design was the same as in group II except that parathyroidectomy was performed at 90 minutes. Arterial blood samples were collected into tubes containing heparin every 10 minutes during the experimental period in groups I, II, and III; in group IV, samples were collected every 10 minutes during the first 60 minutes and every 5 minutes from 60 to 120 minutes, with an additional sample collected 2.5 minutes after parathyroidectomy.

Experimental protocol for group I—For the first 60 minutes, the 8 dogs allocated to group I received 5% dextrose-saline (0.45% NaCl) solution IV to match volume load in the other groups. After 60 minutes, hypocalcemia was induced via administration of an EDTA solution (280 mg of EDTA/kg of body weight of EDTA in 5% dextrose-0.45% saline solution). The rate of EDTA infusion was adjusted to produce a linear decrease in arterial Ca²⁺ concentration from 60 to 90 minutes. A hypocalcemic clamp, in which Ca²⁺ concentrations were kept at 90-minute values, was maintained from 90 to 120 minutes.^7,13 During the induction of hypocalcemia, the EDTA infusion rate was progressively increased from 0.19 to 0.57 mmol/kg/h. During the hypocalcemic clamp in the period between 90 and 120 minutes, the infusion rate was progressively decreased from 0.57 to 0.15 mmol/kg/h. The total dose of EDTA infused was 0.303 mmol/kg. To prevent hypomagnesemia during the EDTA infusion (60 to 120 minutes), a solution of magnesium sulfate was infused at an increasing rate (0.200 to 0.375 mmol/kg/h) from 60 to 90 minutes, and then the rate was reduced to 0.1 mmol/kg/h during the hypocalcemic clamp (total dose of magnesium infused, 0.225 mmol/kg).

Experimental protocol for group II—Metabolic acidosis was induced in 10 dogs via IV infusion of HCl (2.5 mEq/kg in 200 mL of distilled water) during the 120-minute study. The rate of HCl infusion was 2.25 mEq/kg/h (from 0 to 60 minutes), 1.75 mEq/kg/h (from 60 to 90 minutes), and 1.25 mEq/kg/h (from 90 to 120 minutes; total dose, 3.75 mEq of HCl/kg). During the first 60 minutes, EDTA was infused IV at 0.08 mmol/kg/h to prevent an acidosis-induced increase in arterial Ca²⁺ concentration and maintain Ca²⁺ concentration within reference limits (normocalcemic clamp). Because the infusion of EDTA resulted in hypomagnesemia, magnesium was infused at a rate of 0.05 mmol/kg/h to maintain normalcy. From 60 to 90 minutes, hypocalcemia was induced by a progressively increasing dose of EDTA. Then, to maintain a hypocalcemic clamp (90 to 120 minutes), a progressively decreasing dose of EDTA was infused. Plasma magnesium concentration was maintained within reference limits (1.1 to 1.45 mEq of magnesium/L) by use of a magnesium infusion. The total doses of EDTA and magnesium were 0.432 and 0.3 mmol/kg, respectively.

Experimental protocol for group III—For a period of 60 minutes, metabolic acidosis was induced in 10 dogs via the same experimental protocol as that used in group II, but EDTA was not infused to control the arterial Ca²⁺ concentration. As a result, the Ca²⁺ concentration increased during the induction of metabolic acidosis. After 60 minutes, the study was ended and hypocalcemia was not induced. During the 60-minute infusion of HCl, the total dose was 2.25 mmol of HCl/kg. Because EDTA was not administered, plasma magnesium concentrations did not change.

Experimental protocol for group IV—Five dogs underwent an experimental protocol identical to that used in group II for the first 90 minutes except that before starting the experiment, the thyroid-parathyroid glands were surgically exposed and sutures were placed loosely around the thyroid vessels. At 90 minutes (end of induction of hypocalcemia), a parathyroidectomy was performed.⁸ Briefly, the preplaced sutures were tightened and the thyroid-parathyroid glands were removed in < 5 minutes. In all dogs, 4 parathyroid glands were identified in the resected tissue. A hypocalcemic clamp was maintained from 90 to 120 minutes.

Assessments—Systolic and diastolic blood pressures were measured at the times of blood sample collection in all groups. Concentrations of ionized calcium, sodium, potassium, and chloride and values of pHa, PaO₂, and PaCO₂ were measured in arterial blood samples with selective electrodes^l; measurements were performed immediately after each sample was obtained. By use of an immunoradiometric assay,^m the concentration of intact PTH in arterial blood was quantified. The use of this assay for measurement of PTH concentration in dogs has been validated previously.^7,8,13,14 Parathyroid hormone values for each dog were measured in 1 assay. The arterial bicarbonate concentration was calculated from the PaCO₂ and pHa values by use of the Henderson-Hasselbach equation. Plasma phosphate and magnesium concentrations were measured via spectrophotometry.^n,o At the end of the experiments, dogs were euthanized with an overdose of sodium thiopental.

Statistical analysis—For normally distributed data (all parameters except arterial Ca²⁺ and PTH concentrations), the unpaired Student t test was used for the comparison of 2 groups at the same time interval. A 1-way ANOVA was used to compare more than 2 groups and, if the ANOVA test revealed a significant difference (significance set at a value of P < 0.05), a post hoc Scheffe test was performed to determine intergroup differences. A repeated-measures ANOVA, followed by the post hoc Scheffe test, was used to compare more than 2 means from the same experimental group. For Ca²⁺ and PTH concentrations, nonparametric testing (ie, Mann-Whitney, Friedman ANOVA, and Wilcoxon tests) was used. A Pearson test was performed to determine the correlation between systolic blood pressure and PaO₂ and between systolic blood pressure and ionized calcium values. In the metabolic acidotic dogs with and without the normocalcemic clamp (group II and group III, respectively), a direct comparison of PaO₂ values at each time interval by an unpaired t test did not identify differences between the 2 groups. However, at each time interval during the 60-minute comparison, the PaO₂ value was less in group III than in group II; therefore, the effect on PaO₂ was evaluated over the entire 60-minute period. This evaluation was performed by use of an ANCOVA in which the response variable was PaO₂, the covariates were pH and arterial Ca²⁺ concentration, and the factor analyzed was the difference between groups II and III. For all statistical tests, a value of P < 0.05 was considered significant. Results are presented as mean ± SE values.

Results

Arterial Ca²⁺ and PTH concentrations, pHa, and PaO₂ values in group I (ie, dogs in which normal acidbase status was maintained) and group II (ie, acidotic dogs in which hypocalcemia was induced after 60 minutes of normocalcemia) were compared (Figure 1). The infusion of HCl in group II dogs resulted in a progressive decrease in pHa values that reached 7.21 ± 0.02 at 60 minutes and 7.09 ± 0.02 at 120 minutes (both values were significantly [P < 0.001] different from baseline value). Because EDTA was infused to prevent increases in arterial Ca²⁺ concentration, Ca²⁺ values did not increase in group II during the first 60 minutes of the experimental period. Thus, at all time intervals from 0 to 60 minutes, arterial Ca²⁺ concentrations were not significantly different between groups I and II. From 60 to 90 minutes, hypocalcemia was induced with an EDTA infusion in both groups; arterial Ca²⁺ concentration decreased similarly in both groups (decrease of approx 0.4mM), followed by similar values during the hypocalcemic clamp (90 to 120 minutes). Parathyroid hormone concentrations did not change in group I from 0 to 60 minutes, but in group II, the values increased significantly (P < 0.01) from a baseline concentration of 27 ± 4 pg/mL to 146 ± 30 pg/mL at 60 minutes.

View Full Size

Changes in pHa, arterial Ca²⁺ concentration, PTH concentration, and PaO₂ in 10 anesthetized dogs in which metabolic acidosis was induced at time 0 and hypocalcemia was induced after 60 minutes of normocalcemic conditions (group II;squares) versus data for 8 dogs in which normal acid-base status was maintained and hypocalcemia was induced after 60 minutes of normocalcemic conditions (group I;circles). The letter a indicates that the value in group II is significantly (P < 0.05) different from that in group I at the same time point. The letter b indicates that the value in group I at this time point was significantly (P < 0.05) different from the 60-minute value.

Citation: American Journal of Veterinary Research 67, 5; 10.2460/ajvr.67.5.801

• Download Figure
• Download figure as PowerPoint slide

View Full Size

Changes in pHa, arterial Ca²⁺ concentration, PTH concentration, and PaO₂ in 10 anesthetized dogs in which metabolic acidosis was induced at time 0 and hypercalcemia was allowed to develop during a 60-minute period (group III;squares) versus data for 8 dogs in which normal acid-base status and normocalcemia were maintained during a similar period (group I; closed circles). See Figure 1 for key.

Citation: American Journal of Veterinary Research 67, 5; 10.2460/ajvr.67.5.801

• Download Figure
• Download figure as PowerPoint slide

During the first 60 minutes in group II, the induction of metabolic acidosis resulted in an increase (P < 0.01) in PaO₂ values from a baseline value of 96 ± 2 mm Hg to 108 ± 2 mm Hg. The increase in PaO₂ was evident by 10 minutes (at a pHa value of 7.33 ± 0.001). Subsequent reductions in pHa from 7.33 to 7.21 at 60 minutes did not result in any further increase in PaO₂. At all sampling times between 10 and 60 minutes, PaO₂ values in group II were significantly greater than the baseline value of that group; also, group II values were significantly greater than group I values at the same time intervals (Figure 1). During the induction of hypocalcemia (from 60 to 90 minutes) in group I, PaO₂ values increased (P < 0.02) from 95 ± 2 mm Hg to 104 ± 3 mm Hg and then remained increased during application of the hypocalcemic clamp (90 to 120 minutes). However, in contrast to the rapid increase in PaO₂ associated with acidosis in group II, the hypocalcemia-induced increase in PaO₂ in group I was slow and progressive.

In the dogs with metabolic acidosis in which arterial Ca²⁺ concentration was not clamped (group III), pHa values decreased from 7.40 ± 0.02 to 7.21 ± 0.02 at 60 minutes. At all sampling times from 10 to 60 minutes, pHa values in group III were significantly less than baseline value of that group; also, group III values were significantly lower than group I values at the same time intervals (Figure 2). Because EDTA was not administered during the induction of acidosis, arterial Ca²⁺ concentration increased progressively from a baseline value of 1.24 ± 0.02mM to a maximum of 1.36 ± 0.02mM at 60 minutes. From 20 to 60 minutes, Ca²⁺ values in group III were greater (P < 0.05) than the baseline value; also, group III values were significantly (P < 0.05) greater than group I values at the same time intervals. In group III, PTH concentration increased from baseline values (23 ± 5 pg/mL) in response to acidosis and reached a maximum of 49 ± 9 pg/mL at 30 minutes. However, as the arterial Ca²⁺ concentration continued to increase after 30 minutes, PTH concentration started to decrease, and by 60 minutes, the value was not greater than baseline or the 60-minute value in group I. An increase in PaO₂ was also detected in group III after the first 10 minutes of acidosis, and at 30 and 40 minutes, the values were significantly (P < 0.05) greater than the corresponding values in group I.

View Full Size

Changes in pHa, arterial Ca²⁺ concentration, PTH concentration, and PaO₂ in 10 anesthetized dogs in which metabolic acidosis was induced at time 0 and normocalcemia was maintained during a 60-minute period (group II; closed squares) versus data for 10 anesthetized dogs in which metabolic acidosis was induced at time 0 and hypercalcemia was allowed to develop during a similar period (group III; open squares). The letter a indicates that the value for group II was significantly (P < 0.05) different from that for group III at this time point.

Citation: American Journal of Veterinary Research 67, 5; 10.2460/ajvr.67.5.801

• Download Figure
• Download figure as PowerPoint slide

View Full Size

Changes in PTH concentration and PaO₂ in 5 dogs in which metabolic acidosis was induced at time 0, hypocalcemia was induced after 60 minutes of normocalcemic conditions, and thyroparathyroidectomy (TPX) was performed at 90 minutes (group IV; open circles) versus data for 10 anesthetized dogs in which metabolic acidosis was induced at time 0 and hypocalcemia was induced after 60 minutes of normocalcemic conditions (parathyroid glands remained intact; group II; closed squares). The letter a indicates that the value in group IV is significantly (P < 0.05) different from that in group II at the same time point.

Citation: American Journal of Veterinary Research 67, 5; 10.2460/ajvr.67.5.801

• Download Figure
• Download figure as PowerPoint slide

The comparison between the acidotic dogs with and without the calcium clamp (group II and group III, respectively) revealed that the decreases in pHa during 60 minutes in those groups were not different (Figure 3). Arterial Ca²⁺ concentrations were greater and PTH values were less in group III than in group II. Compared with group II, PaO₂ values were less in group III at every time interval but were not significantly different at any time interval. Because of the consistency of the lesser mean PaO₂ values at all time intervals in group III, an ANCOVA (which integrates the effect on PaO₂ values during the entire 60 minutes) was performed. That analysis revealed that the PaO₂ value was less (P < 0.01) in group III than in group II when integrated for the entire 60 minutes of study. However, the integrated difference in PaO₂ between groups II and III was small, being only 2 mm Hg.

In dogs that underwent parathyroidectomy at 90 minutes of the experimental period (group IV), PTH values rapidly decreased during the subsequent 30 minutes (Figure 4). Parathyroidectomy did not affect pHa or arterial Ca²⁺ concentration because these were controlled by external infusions. In group IV dogs, PaO₂ values were unaffected by parathyroidectomy and remained high from 90 to 120 minutes. In that interval (90 to 120 minutes), there were no differences in PaO₂ between the parathyroidectomized dogs and group II dogs, which were similarly acidotic and hypocalcemic.

Blood pressure (systolic and diastolic) values at baseline were similar among the groups (Table 1). In group I, hypocalcemia resulted in a significant (P < 0.05) decrease in blood pressure from 156 ± 9 mm Hg to 118 ± 10 mm Hg (systolic) and from 81 ± 5 mm Hg to 65 ± 7 mm Hg (diastolic) at 90 minutes. The decrease in systolic blood pressure during the induction of hypocalcemia (from 60 to 90 minutes) strongly correlated with the increase in PaO₂ values (r = −0.66; P < 0.001) and had a weaker correlation with the decrease in arterial Ca²⁺ concentration (r = 0.43; P < 0.01). The decreased blood pressure did not change at 120 minutes during the hypocalcemic clamp. No significant changes in systolic blood pressure were observed after 60 minutes of acidosis in either group II or group III. The blood pressure values in group II during hypocalcemia at 90 and 120 minutes were not different from baseline.

The PaCO₂ value and sodium, phosphate, and magnesium concentrations in arterial blood did not differ at any time point within the same group or among the different groups. As expected, bicarbonate concentration decreased and chloride concentration increased in groups II and III. A small but significant decrease in potassium concentration was detected in dogs during the induction of hypocalcemia in group I. At all time points, values of PaO₂ and PaCO₂ and concentrations of bicarbonate, sodium, potassium, chloride, magnesium, and phosphate for dogs in group IV (that underwent the same experimental protocol as group II except that parathyroidectomy was performed at 90 minutes to determine the role of PTH) did not differ from those in group II.

Discussion

In the present study, the induction of hypocalcemia increased the PaO₂ value in dogs. Moreover, the increase in PaO₂ correlated with a decrease in systolic blood pressure during the induction of hypocalcemia. In addition to verifying that the induction of metabolic acidosis increases PaO₂ values, our data have indicated that the increase in arterial Ca²⁺ concentration induced by metabolic acidosis attenuated the acidosis-induced increase in PaO₂.

The effect of metabolic acidosis on PaO₂ in dogs has been determined previously. Haas and Bergofsky¹¹ reported a decrease in the alveolar-to-arterial O₂ gradient in anesthetized dogs in which acidosis was rapidly induced. Those investigators attributed the increase in PaO₂ to a more homogeneous distribution of ventilation to perfusion in the lung. This hypothesis was challenged by results of subsequent studies,^9,10 which indicated that the acidosis-induced increase in PaO₂ was mainly a consequence of a rightward shift of the oxyhemoglobin dissociation curve. Our results also have indicated that metabolic acidosis increases the PaO₂ value by a magnitude similar to that previously reported.^9–11

The induction of hypocalcemia was associated with a slow, progressive increase in PaO₂ that correlated with a decrease in systolic blood pressure, which in turn correlated with a decrease in arterial Ca²⁺ concentration. In anesthetized dogs, hypocalcemia has been reported^15,16 to reduce cardiac output and decrease peripheral vascular resistance, both of which would act to decrease blood pressure. In dogs, resting pulmonary vascular tone is low and is not lowered further by potent vasodilators.¹⁷ Drop et al¹⁸ reported that pulmonary vessels are not sensitive to changes in arterial Ca²⁺ concentration. However, Farrukh and Michael¹⁹ determined that decreased Ca²⁺ concentration does attenuate hypoxemia-induced pulmonary vasoconstriction. Our data, obtained in the absence of hypoxemia, indicated that the induction of hypocalcemia correlated with an increase in PaO₂ and a decrease in systolic blood pressure in dogs. The decrease in blood pressure induced by hypocalcemia is associated with decreased cardiac output.^15,20–22 Also, because the dogs were receiving assisted ventilation with fixed FIO₂ and tidal volume and had the same value of PaCO₂, it seems unlikely that there was a change in alveolar gas concentration. Intrapulmonary shunt is also known to vary directly with cardiac output.²³ Thus, the best explanation for the increase in PaO₂ associated with hypocalcemia appears to be that the decrease in cardiac output reduced the pulmonary shunt by reducing blood flow through underventilated areas of the lung.

It is also important to note that blood pressure did not change during the induction of hypocalcemia in dogs with metabolic acidosis in the present study. Several factors associated with acidosis might be responsible for the maintenance of normal blood pressure. These include an increase in the contractile force of the left ventricle mediated by sympathoadrenal factors and a catecholamine-mediated arterial and venous vasoconstriction that enhanced blood delivery to the central circulation and counteracted the decrease in peripheral vascular resistance induced by acidosis.²⁴

In the acidotic dogs with and without hypercalcemia, a direct comparison at each time interval did not reveal a significant difference in PaO₂ values; nevertheless, at each time interval, the PaO₂ value was less in the dogs with acidosis and hypercalcemia. Consequently, an ANCOVA was performed to evaluate the effect over the 60-minute period and revealed a modest, but significant, decrease in PaO₂ in the dogs with acidosis and hypercalcemia, compared with the dogs with acidosis and normocalcemia. One possible explanation for the lower PaO₂ value in acidotic and hypercalcemic dogs is that hypercalcemia increases cardiac contractility and output,^20,21,25 which in turn has been associated with increased pulmonary shunting.^23,26–28

The lesser increase in PaO₂ during the acidosis-induced increase in arterial Ca²⁺ concentration together with the increase in PaO₂ detected during the period of hypocalcemia suggests that serum calcium concentration affects the PaO₂ value. Although the increase in PaO₂ in acidotic dogs and hypocalcemic dogs was similar (change in PaO₂ values of approx 10 mm Hg), the increase in PaO₂ was slowly progressive during the induction of hypocalcemia, whereas a small decrease in pHa resulted in a maximal increase in PaO₂. With regard to the acidosis-induced increase in arterial Ca²⁺ concentration, the small PaO₂ differences between normo- and hypercalcemic acidotic dogs would seem to have limited impact on blood oxygenation and consequently be of little clinical relevance. Increases in PaO₂ induced by metabolic acidosis or hypocalcemia may have clinical relevance in the acute care setting, but studies need to be performed to evaluate their importance in critically ill dogs in which acidosis, hypocalcemia, and reduced cardiac output frequently develop. In clinically normal or noncritically ill dogs undergoing anesthesia, the development of metabolic acidosis or reductions in arterial Ca²⁺ concentration may affect PaO₂ values and thus require adjustment of ventilatory settings.

Other factors that change in parallel with alterations in Ca²⁺ concentration might also have affected the PaO₂ value. One such factor is PTH, which increases in concentration as Ca²⁺ concentration and pHa values decrease. Parathyroid hormone has cardiovascular effects. Results of several studies^29,30 have determined that the infusion of PTH has a hypotensive effect, but chronic excess of PTH has been associated with systemic³¹ and pulmonary³² hypertension. Because both acidosis and hypocalcemia increase PTH concentration and PaO₂, it is possible that changes in PTH concentration were responsible for the PaO₂ changes detected in the dogs of the present study. However, this possibility seems unlikely because PaO₂ values continued to increase during hypocalcemia even after maximal PTH concentrations were obtained and the acidosis-induced increase in PaO₂ values was attenuated during the period of hypercalcemia induced by acidosis even though PTH values did not change. Finally and most importantly, the increased PaO₂ values in acidotic dogs were unaffected by parathyroidectomy, even though PTH values decreased rapidly as acidosis and hypocalcemia were maintained. Thus, we do not believe that PTH contributed to the changes in PaO₂.

Another factor to consider is the administration of EDTA, which was used to induce the hypocalcemia in the dogs of the present study. The progressive hypocalcemia was induced by increasing the amount of EDTA administered. However, during application of the hypocalcemic clamp, the EDTA dose was reduced 4-fold, and this did not result in a decrease in PaO₂. Thus, any effect of EDTA on PaO₂ would seem to be indirect and mediated by hypocalcemia.

In our study, maintenance of dogs on mechanical ventilation minimized changes in PaCO₂. In the absence of fixed ventilation, the induction of metabolic acidosis does result in compensatory hyperventilation with a decrease in PaCO₂, which should have 2 opposite effects on values of PaO₂; there would likely be an increase in PaO₂ mediated by the increase in alveolar PO₂ and a lesser decrease in pHa, which would act to minimize an acidosis-induced increase in PaO₂.

Our data have indicated that the induction of hypocalcemia increases the PaO₂ value of anesthetized dogs by a mechanism that is independent of PTH and is likely related to the cardiovascular effects of hypocalcemia that probably resulted in decreased pulmonary shunting. Furthermore, results of the present study in dogs have indicated that the acidosis-induced increase in PaO₂ values is modestly reduced by the increase in arterial Ca²⁺ concentration induced by metabolic acidosis. The effect of metabolic acidosis and hypocalcemia on PaO₂ should be taken into account in anesthetized dogs in which these abnormalities develop. Whether the improved oxygenation detected in healthy dogs during the induction of metabolic acidosis and hypocalcemia would occur in critically ill dogs remains to be determined.