Role of Heparin and Nitric Oxide in the Cardiac and Regional Hemodynamic Properties of Protamine in Conscious Chronically Instrumented Dogs

The protocols were approved by The University of Texas Animal Welfare Committee.

A description of the basic model has been previously described. 13Briefly, 12 mongrel dogs of either sex, heartworm free, weighing 25–34 kg, were instrumented with catheters in the iliac artery, left atrium, and pulmonary artery. A precalibrated ultrasonic flow probe was positioned around the pulmonary artery, and a miniature pressure transducer was inserted into the left ventricle. Pulsed Doppler flow probes were positioned around the vertebral, carotid, coronary, mesenteric, hepatic, and renal arteries and the portal vein. All transducer leads and catheters were tunneled subcutaneously to the dorsum of the neck and secured after the thoracotomy was closed. Analgesia and antibiotic prophylaxis were initiated before surgery and were maintained thereafter.

A detailed description of the measurement techniques has also been previously published. 13–15Phasic and mean aortic pressures, pulmonary arterial pressure, first derivative of left ventricular pressure (dP/dt), heart rate, cardiac output, and carotid, vertebral, coronary, hepatic, portal, mesenteric, and renal blood flow were continuously recorded on a 16-channel Gould brush polygraph (Gould, Cleveland, OH). Systemic vascular resistance was calculated as the ratio of mean arterial pressure (MAP) to cardiac output; regional vascular resistance was calculated as the ratio of MAP to regional blood flow.

Dogs were carefully nursed through the first 24 h postoperatively and on subsequent days were trained to lie quietly on the laboratory floor. The dogs were studied no less than 10 days after surgery, when hematocrit was greater than 30% and body temperature, appetite, and general appearance were normal. Body weight, body temperature, arterial blood gases, and hematocrit were routinely measured before each experiment. All experiments were conducted in fasted conscious dogs lying on their right sides. Although either immediate or secondary effects after protamine administration have been previously documented, 16,17no studies have assessed both changes. Therefore, this study was designed to assess combined immediate and secondary effects of protamine in conscious and chronically instrumented dogs. Dogs were equally distributed to either protocol 1 (n = 6) or protocol 2 (n = 6).

Each dose of protamine was given on a separate day at least 48 h after the previous dose. On each experimental day, control measurements were obtained in resting conscious animals before the administration of either protamine alone or protamine combined with heparin. The order of administration of the protamine dose alone or in the presence of heparin was randomized.

Protamine sulfate was administered intravenously in doses of 0.5, 1.0, 3.0, 5.0, and 8.0 mg/kg, either alone or in the presence of heparin. The dose of heparin was calculated to maintain a ratio of 100 IU/mg protamine: 50 IU/kg heparin for 0.5 mg/kg protamine; 100 IU/kg heparin for 1 mg/kg protamine; 300 IU/kg heparin for 3 mg/kg protamine; 500 IU/kg heparin for 5 mg/kg protamine, and 800 IU/kg heparin for 8 mg/kg protamine. Protamine was administered 15 min after heparin administration. Because the magnitude of cardiovascular responses to protamine administration has been shown to depend on the rate of administration, 8we chose to inject heparin and protamine as a bolus dose over a period of 10 s. Hemodynamic parameters were recorded at steady state before injection of protamine or heparin and for 30 min after protamine injection.

To study the role of the NO pathway in protamine-associated cardiac changes, N -methyl-l-arginine (L-NMA), a specific NO synthase (NOS) inhibitor, was administered at 300 μg · kg⁻¹· min⁻¹, intravenously, before protamine administration. Protamine (3 mg/kg) was administered after heparin (300 IU/kg) as described in protocol 1 in the presence or absence of L-NMA infused at 300 μg · kg⁻¹· min⁻¹on 2 separate occasions. The dose of L-NMA was chosen because it produced maximum pressor effects in conscious dogs (unpublished data). The infusion of L-NMA was started 45 min before the injection of protamine using a 0.2-μm syringe filter (Nalge Company, Rochester, NY). In the non–L-NMA-treated group, the same dose of saline was infused throughout the study. Identical experiments were also performed with protamine (8 mg/kg) in nonheparinized dogs in the absence or presence of L-NMA infusion (300 μg · kg⁻¹· min⁻¹). The dose of protamine alone and combined with heparin was chosen to obtain a similar decrease in MAP (protocol 1). Hemodynamic parameters were recorded at baseline (before L-NMA infusion), before protamine injection, and for 30 min after protamine injection.

Drugs used in all experiments were porcine intestinal mucosal heparin sodium (Elkins-Sinn, Inc., Cherry Hill, NY), protamine sulfate (Fujisawa, USA, Inc., Deerfield, IL), and L-NMA (ALEXIS Corporation, San Diego, CA).

Data obtained from protocol 1 in heparinized and nonheparinized conditions were analyzed using an analysis of variance (ANOVA) for repeated measures in each group. When differences were significant, multiple within-group comparisons to the baseline value obtained before protamine injections were performed using the Dunnett t  test. 18In addition, when changes in heparinized and nonheparinized groups were significantly different, the magnitude of change was compared using an unpaired t  test. Data obtained in protocol 2 were analyzed using an analysis of variance followed by a Dunnett t  test. An unpaired t  test was also used to determine statistical differences between control and L-NMA–treated groups in heparinized and nonheparinized conditions. Results in the figures are presented as percent changes versus  control. A P  value < 0.05 was considered significant.

Cardiac and regional hemodynamics before protamine administration are presented in tables 1 and 2. Values are consistent with those previously reported for trained and unstressed animals. 9,10Changes in cardiac and regional hemodynamic parameters after protamine alone and in the presence of heparin are presented in figures 1–4. Hemodynamic changes after protamine alone and in the presence of heparin varied according to the parameter, the time course (whether immediate or secondary responses), and the dose under investigation. By itself, heparin produced no cardiac and regional hemodynamic changes (table 2).

Protamine alone produced an initial and short-lasting increase in cardiac function and an increase in MAP followed by a secondary decrease in cardiac function and hypotension. Its effects on regional circulation were limited. Thus, we recorded immediate and transient increases in MAP at 3, 5, and 8 mg/kg, pulmonary arterial pressure and dP/dt at 5.0 and 8.0 mg/kg, and cardiac output and heart rate at 8 mg/kg, as well as mesenteric vasoconstriction and vertebral vasodilatation at 5 and 8 mg/kg.

In the presence of heparin, the protamine-induced pressor effect converted into hypotension at 3.0, 5.0, and 8.0 mg/kg. Consequently, systemic vasodilatation was also recorded. Furthermore, protamine also induced pulmonary vasoconstriction at 3.0, 5.0, and 8.0 mg/kg, tachycardia at 3.0 and 5.0 mg/kg, and an increase in dP/dt at 5 mg/kg. Regional circulation in peripheral vascular beds was also affected after protamine administration. Vasodilatation was recorded in the carotid circulation at 3 and 5 mg/kg and in the coronary and hepatic circulation at 3, 5, and 8 mg/kg. Mesenteric vascular resistance initially increased at 1 mg/kg, and vertebral vasodilatation was recorded at 1, 3, and 8 mg/kg. A decrease in portal blood flow was observed at 3 mg/kg. Finally, renal vasoconstriction was detected after protamine was administered at 8 mg/kg.

The effects of protamine alone recorded at 5 min were limited to cardiac output, heart rate, dP/dt, pulmonary and renal vascular resistance, and portal blood flow. At the highest dose, cardiac output, dP/dt, and renal vascular resistance decreased at 8 mg/kg, whereas pulmonary vascular resistance increased at 3, 5, and 8 mg/kg. In addition, a decrease in portal blood flow was recorded at 5 and 8 mg/kg. Finally, heart rate increased at 3, 5, and 8 mg/kg.

In the presence of heparin, protamine induced more pronounced cardiac and regional hemodynamic changes. Protamine at 3, 5, and 8 mg/kg produced decreases in MAP, cardiac output, and dP/dt with a renal vasodilatation and a decrease in portal blood flow. In addition, hepatic vasodilatation was recorded at 8 mg/kg, and mesenteric vasodilatation was recorded at 5 mg/kg after protamine administration. In contrast, the secondary pulmonary vasoconstriction produced by protamine in the presence of heparin was attenuated.

Cardiac and regional hemodynamics in control conditions and before L-NMA are presented in tables 3 and 4. In the saline groups, each measurement remained essentially unchanged. L-NMA infused at 300 μg · kg⁻¹· min⁻¹induced changes consistent with those previously observed using L-NMA administered intravenously at 20 mg/kg in conscious dogs. 13Hemodynamic changes after protamine alone or in the presence of heparin varied according to the presence of L-NMA, the hemodynamic parameters, the time course (whether immediate or secondary response), and the dose under investigation.

Nitric oxide synthase blockade resulted in an inhibition of protamine-induced immediate increase in dP/dt and an accentuation of the renal vasoconstriction. However, administration of L-NMA did not significantly prevent other hemodynamic changes caused by protamine.

In the presence of heparin, NOS blockade resulted in an inhibition of protamine-induced immediate decreases in MAP, portal blood flow, and systemic, coronary, mesenteric, and hepatic vascular resistance. Accentuation of the renal vasoconstriction was also detected.

Nitric oxide synthase blockade resulted in an accentuation of protamine-induced renal vasodilatation. In the presence of heparin, NOS blockade did not affect the sustained hypotension observed at 5 min.

Our data obtained from the first protocol demonstrate that the cardiac and regional vascular properties of protamine in the tissue or organ examined are altered by the presence of heparin. They also confirm that the time course and the dose used are important determinants of the cardiac and regional hemodynamic properties of protamine. Protamine alone at doses of 3 mg/kg or greater appears to produce an immediate and mostly dose-independent increase in MAP as well as cardiac stimulation that is followed by a cardiac depression. Protamine induced immediate mesenteric vasoconstriction and vertebral vasodilatation and secondary renal vasodilatation, especially at doses of 5 and 8 mg/kg. In the presence of heparin, the changes previously documented with protamine were either accentuated, attenuated, or unchanged. Thus, the secondary changes in MAP, cardiac output and dP/dt, pulmonary vascular tone, heart rate, and vertebral and renal vasodilatations produced by protamine were accentuated in the presence of heparin.

In contrast, the immediate systemic and mesenteric vasoconstriction produced by protamine was changed into hypotension and systemic and mesenteric vasodilatations. In the presence of heparin, protamine failed to produce cardiac stimulation, and the magnitude of the secondary increase in pulmonary arterial pressure was attenuated. In addition, protamine administered in the presence of heparin led to immediate and dose-independent carotid, coronary, and hepatic vasodilatations, territories that were not affected by protamine alone.

Our data obtained from the second protocol indicate that the NO pathway was involved in the immediate hemodynamic changes associated with protamine in the presence of heparin. Thus, the immediate increase in dP/dt after the administration of protamine alone was inhibited by NOS blockade, and renal vasoconstriction was accentuated. In the presence of a NOS inhibitor, the combination of heparin and protamine failed to induce immediate hypotension, portal hypoperfusion, or coronary, mesenteric, or hepatic vasodilatation, and the magnitude of the systemic vasodilatation was attenuated. In contrast, our data do not support the role of NO in the secondary hemodynamic changes produced by protamine.

In in vivo  studies, Kien et al.  1reported that in chronically instrumented dogs during halothane anesthesia, the cardiac depression induced by protamine was not affected by heparin. Our data obtained in the absence of anesthesia and surgical stress indicate that, in dogs, protamine alone secondarily decreased myocardial function in doses of 3 mg/kg and higher. Thus, cardiac output and dP/dt decreased in the absence of changes in systemic vascular resistance and in the presence of increased pulmonary vascular resistance and heart rate. 19Although in our experimental conditions heparin did not affect the threshold and maximum cardiac changes, it accentuated the myocardial depression produced by protamine, especially at 3 and 5 mg/kg. Kien et al.  1previously reported that, in anesthetized dogs, protamine produced similar effects on cardiac function. However, in the latter studies the cardiac depression appears to be of greater magnitude when given at 3 and 5 mg/kg in anesthetized animals as compared with our awake model. It is most likely that the presence of halothane was a major contributor to the accentuation of cardiac depression with protamine. We previously demonstrated that halothane may accentuate the cardiac depression induced by calcium channel blockers. 20This is of special interest because protamine properties have also been demonstrated to depend on calcium movement. 11 

In in vitro , Akata et al.  11demonstrated that protamine produced a dose-dependent vasodilatation in isolated mesenteric arteries precontracted with potassium and norepinephrine. This protamine-induced relaxation appears to be calcium-dependent and endothelium-dependent and -independent. More recently, Akata et al.  21demonstrated that pretreatment with heparin blocked this response. Further studies from the same group also revealed that the ratio of heparin to protamine is an important determinant of the magnitude of the interaction. The investigators reported that in an in vitro  experiment, heparin did not affect protamine-induced relaxation for ratios less than 1.45 IU/μg, whereas heparin completely blocked protamine responses for ratios greater than 4.7 IU/μg. 22Our data indicate that protamine in doses of 5 and 8 mg/kg produced an immediate and transient mesenteric vasoconstriction. In the presence of heparin at a ratio of 100 IU/mg, mesenteric vasoconstriction was observed at the lowest dose (1 mg/kg), whereas protamine administered at 3 and 5 mg/kg produced an immediate and sustained vasodilatation. This apparent discrepancy between in vitro  and in vivo  studies may result from alterations in local circulation caused by elimination of exogenous neural control as well as alterations in blood flow–mediated changes in endothelial function. In the absence of heparin, protamine at 1.0, 3.0, and 8 mg/kg produced an immediate and brief increase in MAP combined with a modest, transient increase followed by a more pronounced decrease in cardiac output at 8 mg/kg. Furthermore, in in vitro  studies, protamine-mediated mesenteric vasodilatory properties required a precontracted vessel with either potassium or norepinephrine. Therefore, it is likely that in the absence of high levels of norepinephrine or potassium, protamine may have failed to induce relaxation in in vivo  studies.

Pearson et al.  12demonstrated that in vitro , protamine produced a dose-related relaxation of canine coronary and renal arteries precontracted with prostaglandin F²α, and that heparin did not interfere with protamine properties. Our findings also contrasted with those obtained in the portal circulation, in which protamine had minimal effects in the absence of heparin or produced modest changes in the presence of heparin. As previously discussed, autoregulatory mechanisms play a major role in the control of vascular tone in the carotid, vertebral, and hepatic circulation. In contrast, the coronary circulation is mostly metabolically regulated. However, the profound coronary vasodilatation produced by protamine in the presence of heparin and the consequences of NOS inhibition suggest that these effects are most likely related to surface active properties of protamine through the release of NO. However, in our second protocol, a NOS inhibitor prevented coronary vasodilatation in the presence of heparin–protamine, whereas dP/dt was still depressed. Because the coronary artery was precontracted with prostaglandin F²α in the study by Pearson et al. , 12the lack of a high concentration of prostaglandin F²α in in vivo  studies may be the reason for the discrepancy between in vivo  and in vitro  studies.

In several in vivo  studies, 6,8,23–25protamine administered either alone or in the presence of heparin has been shown to produce pulmonary vasoconstriction. However, in these studies, the properties of protamine alone and in the presence of heparin have been documented in separate experimental designs. Therefore, the role played by heparin in the pulmonary vasoconstriction induced by protamine was not established. Our data indicate that protamine administered alone increased pulmonary arterial pressure, a change usually observed immediately after protamine administration. However, at the highest dose, a secondary pulmonary vasoconstriction was recorded. In the presence of heparin, the immediate increase in pulmonary arterial pressure was accentuated, whereas heparin attenuated the secondary increase in the pulmonary vascular tone.

Protamine is rich in the basic amino acid arginine, which is the precursor of endothelial cell synthesis of NO. Although in vitro  protamine has been shown to relax vascular smooth muscle through an endothelium-dependent mechanism, 11Hakim et al.  6indicated that the NO system was not involved in the heparin–protamine response in anesthetized pigs, in which local blood flows could not be obtained. Our data from the second protocol would explain these discrepancies between in vitro  and in vivo  experiments. Therefore, it seems that the importance of the NO system as the mechanism of protamine-induced vasodilatation depends on the local circulation. Furthermore, heparin potentiated the protamine-induced systemic and selective vascular NO-dependent vasodilatations in the coronary, mesenteric, portal, and hepatic circulations at 1 min. However, our data do not support the role of NO in the decrease in cardiac function and pulmonary, carotid, and vertebral vascular changes induced by protamine-heparin complex.

In conclusion, our data provide evidence that protamine itself causes biphasic changes in arterial blood pressure and cardiac function, pulmonary vasoconstriction, and limited and selective regional vasodilatation and/or vasoconstriction in conscious dogs. The presence of heparin had multiple consequences on the cardiac and hemodynamic properties of protamine; it abolished the pressor and cardiac stimulation and accentuated hypotension and cardiac depression as well as the initial pulmonary pressor response of protamine. In the presence of heparin, protamine also produced several regional hemodynamic consequences: (1) carotid, coronary, and hepatic circulation appeared to be selectively affected by the combination of heparin plus protamine and not by protamine alone; (2) the effects of protamine on renal and mesenteric circulations are dependent on the time course and the dose, respectively; and (3) the secondary pulmonary pressor effect of protamine was attenuated in the presence of heparin. Finally, our findings also indicate that NO-dependent mechanisms are involved in the systemic and selective regional protamine-mediated vasodilatation, especially in the coronary, mesenteric, and hepatic circulations in heparinized dogs.

The authors thank David Valdecanas, B.S., Research Assistant II, The University of Texas Medical School, Houston, Texas, for valuable technical assistance and Anne Starr, Staff Assistant, The University of Texas Medical School, Houston, Texas, for preparation of the manuscript.