Influence of hydrogen ion concentration versus carbon dioxide tension on pulmonary vascular resistance after cardiac operation Disturbances of respiratory acid-base status are common in patients supported with mechanical ventilation of the lungs after cardiac operations. This study was conducted with two protocols. The purpose was to determine whether respiratory acid-base status influences pulmonary vascular resistance in adults after cardiac operations and whether the influence is mediated by hydrogen ion concentration or carbon dioxide tension. Patients were studied while under general anesthesia immediately after aorta-coronary bypass. In the first protocol, with seven patients, arterial carbon dioxide tension was manipulated by the addition of 5 % carbon dioxide to the breathing circuit. Pulmonary vascular resistance index was determined as arterial carbon dioxide tension rose from 30 mm Hg to 50 mm Hg and back to 30 mm Hg. In the second protocol, with 10 different patients, hydrogen ion concentration was manipulated by the addition of 0.2N hydrochloric acid, sodium bicarbonate, or both as arterial carbon dioxide tension was held constant. We used analysis of variance for statistical data. The results of the first protocol showed that pulmonary vascular resistance index rose by 44 % (p < 0.05) as arterial carbon dioxide tension rose from 30 to 50 mm Hg. The results of the second protocol showed that changes in pulmonary vascular resistance index were parallel to changes in hydrogen ion concentration as arterial carbon dioxide tension was held constant (p < 0.05). These data demonstrate that respiratory acid-base status is an important determinant of pulmonary vascular resistance in the adult after cardiac operations. Furthermore, these data suggest the effect is mediated by hydrogen ion concentration, not carbon dioxide tension. (J THORAC CARDIOVASC SURG 1993;106:528-36)
David A. Fullerton, MD, Lyle E. Kirson, DDS, John A. St. Cyr, MD, PhD, Theresa Kinnard, MD, and Glenn J. R. Whitman, MD, Denver, Colo.
Athough the importance of hypoxemia in the determination of pulmonary vascular resistance is widely known, considerably less attention has been paid to the role of acid-base status. Evidence generated from experimental animals suggests that hypercarbic acidemia induces pulmonary vasoconstriction.l" whereas hypocarbic alkalemia results in pulmonary vasodilation.s 4 These concepts have proved useful in the clinical management of neonates with elevated pulmonary vascular From The University of Colorado Health Sciences Center, Denver, Colo. Received for publication July 7, 1992. Accepted for publication Nov. 4,1992. Address for reprints: David A. Fullerton, MD, Department of Surgery, Campus Box C-31O,University of Colorado Health Sciences Center, 4200 E. Ninth Ave., Denver, CO 80262. Copyright
w
1993 by Mosby-Year Book, Inc.
0022-5223/93 $1.00 +.10
528
12/1/44184
resistance as a result of persistent pulmonary hypertension. Hyperventilation has been used in the management of such neonates to lower pulmonary vascular resistance and thereby improve both pulmonary blood flowand right ventricular function.! Although the influence of respiratory acid-base status on neonatal pulmonary vascular resistance is well appreciated, it is widely believed that respiratory acid-base status has little, if any, effect on pulmonary vascular resistance in the adult. 6 Yet, because our anecdotal clinical experience suggested that respiratory acid-base status may indeed influence respiratory acid-base status in adult cardiac surgical patients, this study was undertaken. Few studies have attempted to examine the influence of acid-base status on the adult pulmonary vasculature in the absence of cardiopulmonary bypass. Furthermore, the published results have been inconsistent.H" These studies have examined the influence of hypercarbia in awake,
The Journal of Thoracic and Cardiovascular Surgery Volume 106, Number 3
spontaneously breathing subjects and have therefore lacked control of other factors known to influence pulmonary vascular resistance, such as respiratory rate, tidal volume, and alveolar or arterial oxygen tension (Po z).11 In addition, these published clinical studies have lacked the ability to accurately measure the pulmonary outflow pressure (left atrial pressure).'! In this regard, the anesthetized patient supported with mechanical ventilation of the lungs after cardiac operations offered a unique and clinically relevant opportunity. This study was conducted with two protocols. The first protocol was used to examine the influence of respiratory acid-base status on adult pulmonary vascular resistance. The second protocol was designed to distinguish the role of hydrogen ion concentration ([H+]) from that of carbon dioxide tension (Pco-) in the control of pulmonary vascular resistance.
Methods To examine a safe and clinically relevant range of respiratory acid-base statuses, we reviewed the results of arterial blood gas samples from the surgical intensive care unit at our institution over a 2-month period. During this time, approximately 50 arterial bloodgas samples were taken daily. We found that 95% of these samples fell within the pH range of 7.27 and 7.63. Eighty percent fell within the pH range of 7.30 and 7.50. Thus, the pH range of 7.30 and 7.50 was judged to be a very common clinical occurrence and was therefore chosen as the range of arterial pH for this study. Patients with valvular heart disease or left ventricular dysfunction considered sufficient to influence pulmonary vasomotor tone were excluded from this study. Both protocols were approved by the Human Subjects Review Committee of the University of Colorado Health Sciences Center and the Research and Development Committee, Human Subjects Subcommittee, of the Denver Veterans Administration Medical Center. Informed consent was obtained from each participant. Protocoll
Roleofrespiratory acid-base status in the controlofpulmonaryvascular resistance. Seven consecutive male patients with
severe three-vessel coronary artery disease undergoing aortacoronary artery bypass participated in the study. All patients had normal preoperative pulmonary pressures and left ventricular ejection fractions of greater than 50%. All patients underwent complete coronary artery revascularization, including reversed saphenous vein grafts to the right coronary artery system. Patients received preoperative medication (morphine sulfate, 0.1 mg/kg body weight) and scopolamine, 0.4 mg) intramuscularly I hour before arrival in the operating room. Ongoing drug therapy for concomitant medical problems was continued as deemed appropriate by the anesthesiologist. Each patient was monitored with a five-lead electrocardiogram, a radial arterial line, and a pulmonary artery thermodilution catheter introduced through the right internal jugular vein. In addition, a left atrial pressure line was introduced through the right superior pulmonary vein after the patient had been weaned from cardiopulmonary bypass, after completion of
Fullerton et al.
529
data collectionand before chest closure. Anesthesia consisted of a high-dose narcotic (fentanyl citrate) relaxant (vecuronium bromide) technique supplemented with intravenous Medazolam and, if necessary, an inhalation agent. Anesthetic agents were administered only before cardiopulmonary bypass. Any influenceof anesthesia on smooth muscle function was therefore minimized and believedto be constant during the period of data collection. Data were collectedafter the completion of cardiopulmonary bypass and before chest closure. After being weaned from bypass and undergoing protamine administration, all patients were hemodynamically stable and demonstrated normal coagulation. No patients required postbypass cardiac pacing, antiarrhythmic therapy, or inotropic or vasoactive drug administration. No inhalational anesthetics were used. The protocol for collection of data proceeded as follows: Tidal volume was set at approximately 10 cc/kg, and respiratory rate was adjusted to establish an arterial Pco- of 30 mm Hg. To avoid changes in pulmonary hemodynamics as a result of changes in ventilatory patterns, we maintained constant ventilator settings during the study period. Fraction of inspired oxygen was maintained at a mean of 0.97 (range 0.94 to 0.99), and no patient had application of positiveend-expiratory pressure at any point during the study period. Arterial Poz was therefore maintained at a level greater than 150 mm Hg throughout the study period. Arterial Pcoj was elevated from 30 nun Hg to 40 mm Hg to 50 mm Hg by the addition of 5% carbon dioxide to the breathing circuit. Carbon dioxide was withdrawn from the breathing circuit to lower Pco- to baseline values before measurement of hemodynamic variables. Arterial blood gas samples were obtained at each point of data collection. The hemodynamic variables measured and recorded were heart rate, systemic mean arterial blood pressure, mean pulmonary arterial pressure, pulmonary capillary wedge pressure, central venous pressure, left atrial pressure, and thermodilution cardiac output (mean of three values). Pulmonary vascular resistance index was calculated with the following formula: PVRI=
PAP-LAP CI X 80
where PVRI is pulmonl!!:Lvascular resistance index (dynes. sec· em"? . m- Z) , PAP is mean pulmonary artery pressure, LAP is left atrial pressure, and CI is cardiac index. Cardiac index was derived as cardiac output divided by body surface area. CI RVSWI = HR X PAP X 0.0144 where RVSWI is right ventricular stroke work index (gm-m/
mZ) and HR is heart rate.
Baseline hemodynamic variables were those initial determinations made at an arterial Pco- of 30 mm Hg. Standard oneway classification analysis of variance in conjunction with the Student-Newman-Keuls multiple comparisons procedure and two-sidedstatistical evaluation were used. A p value of less than 0.05 was accepted as statistically significant. Protocol 2 Roles of[H+J and Pco, in the controlofpulmonary vascular resistance. Ten consecutive male patients with severe threevessel coronary artery disease undergoing aorta--coronary artery bypass participated in the study. All patients had normal
5 30
The Journal of Thoracic and Cardiovascular Surgery September 1993
Fullerton et al.
Table I. Protocol 1: demographics No. of patients Age (yr) Body surfacearea (m2) Tidalvolume (cc/kg) Respiratory rate (breaths/min) Aorticcrossclamp time (min) Cardiopulmonary bypass time Values are expressed as mean ± standard error
7 64.9 ± 2.03 ± 10.3 ± 9.6 ± 90.6 ± 134.4 ±
2 0.05 0.2 0.3 8 13
of the mean.
preoperative pulmonary pressures and left ventricular ejection fractions of greater than 50%. All patients underwent complete coronary artery revascularization, including reversed saphenous vein grafts to the right coronary artery system. The anesthetic management and intraoperative monitoring were the same as those of the first protocol. As in the first protocol, data were collected after the completion of cardiopulmonary bypass and before chest closure. Patients who were in hemodynamically stable condition and demonstrated normal coagulation after being weaned from bypass and undergoing protamine administration were considered acceptable for study. No patients required postbypass cardiac pacing, antiarrhythmic therapy, or inotropic or vasoactive drug administration. No inhalational anesthetics were used. The protocol for collection of data proceeded as follows: Tidal volume was set at approximately 10 cc/kg and respiratory rate was adjusted to establish an arterial Pco- of 30 mm Hg. To avoid changes in pulmonary hemodynamics as a result of changes in ventilatory patterns, we maintained constant ventilator settings during the study period. Fraction of inspired oxygen was maintained at 0.93 (fraction of inspired nitrogen at 0.07) and no patient had application of positive end-expiratory pressure at any point during the study period. Arterial Pco- was maintained at a level greater than 150 mm Hg throughout the study period. [H+ j was determined by converting the measured pH value to [H+j with a standard conversion chart.P The initial point of data collection was at an arterial Pcoj of 30 mm Hg and [H+j of 33.3 nmol/L (pH 7.49). Arterial Pcowas then elevated to 50 mm Hg by the addition of 5% carbon dioxide at the expense of a decreased fraction of inspired nitrogen to the anesthesia breathing circuit. This was accompanied by a rise in [H+ j without changing arterial Pco-; sodium bicarbonate was administered intravenously over a period of 10 minutes in a dose of approximately 2 mfiq/kg body weight. When approximately one half of the calculated dose was in, an arterial blood gas sample was obtained to assure that no unanticipated fluctuation in arterial pH had occurred. The remainder of the sodium bicarbonate was then administered. During the administration of bicarbonate, the amount of carbon dioxide introduced into the circuit was altered to maintain an arterial Pco- of 50 mm Hg. This dosage of sodium bicarbonate lowered [H+j to approximately 43 nmoljL (pH 7.37). To raise the [H+j again to approximately 50 nmol/L, we administered O.2N hydrochloric acid in a dosage of approximately 5 ml/kg through a central venous catheter over a l O-minute period. Again, after one half of the calculated dose was infused, an arterial blood gas sample was obtained to assure that no untoward change in arterial pH had occurred. The remainder of the hydrochloric acid was then administered to return [H+j to approximately 50 nmol/L (pH 7.30).
The final point of data collection was obtained by removal of carbon dioxide from the anesthesia breathing circuit to return to an arterial Pco- of 30 mm Hg and [H+j of 33 nmol/L (pH 7.49). Each level of [H+j was maintained in a steady state for at least 10 minutes before measurement of hemodynamic variables. The measured and derived hemodynamic variables were the same as in the first protocol. Mixed venous Paz, Pcoj, and [H+j were determined as well. Standard one-way classification analysis of variance in conjunction with the Student-NewmanKeuIs multiple comparison procedure and two-sided statistical evaluation were used. A p value of less than 0.05 was accepted as statistically significant.
Results Protocol 1. The purpose of the first protocol was to discern whether respiratory acid-base status influenced pulmonary vascular resistance in adult patients undergoing cardiac operations. The study population comprised seven consecutive male patients, with subject demographics listed in Table I. All patients had a history of cigarette smoking; however, none demonstrated clinical or radiographic evidence of significant chronic pulmonary disease. No subjects had preoperative pulmonary hypertension. All subjects were taking at least one aspirin daily in the preoperative period. Table II lists the arterial blood gas values at each point of data collection and the hemodynamic variables determined at each level of PC02. Heart rate, central venous pressure, and left atrial pressure remained remarkably consistent throughout the study period. With values obtained at an arterial PC02 of 30 mm Hg established as baseline, pulmonary vascular resistance index rose by 44% (p < 0.05) as PC02 rose to 50 mm Hg. Pulmonary vascular resistance index returned to baseline as PC02 returned to 30 mm Hg (Fig. 1). There was no difference in pulmonary vascular resistance index between the start and finish of the protocol (p = 0.80). At a PC02 of 40 mm Hg, pulmonary vascular resistance index increased by 24%; however, this increase was not statistically significant. As shown in Fig. 2, hypercarbic acidemia produced a significant (p < 0.05) rise in mean pulmonary arterial pressure with no significant change in either left atrial pressure or cardiac index. This implies that the rise in pulmonary arterial pressure and pulmonary vascular resistance was produced by pulmonary vasoconstriction. The increased pulmonary vascular resistance index produced by increasing Pco- from 30 mm Hg to 50 mm Hg was associated with a significant increase (p < 0.05) in right ventricular stroke work index as shown in Fig. 3. Protocol2. The purpose of the second protocol was to distinguish whether the effects of the respiratory acidbase status were mediated by Pco- or [H+]. This study
The Journal of Thoracic and Cardiovascular Surgery Volume 106, Number 3
Fullerton et al.
53I
460
*
440 420 400 380 II:
> D.
360 340 320 300 280
*
260 pC02
30
pH [H+]
7.47
45 7.37
34.4
42.7
50 7.31 50.1
p < 0.05
40
30
7.39 40.7
34.2
7.47
Fig. 1. Pulmonaryvascularresistanceindex (PVRI) as a function of arterial respiratoryacid-basestatus in protocol I. PVRI increased by44%as PC02 increased from 30 mm Hg to 50 mm Hg. Vertical bars representmean ± standard error of mean. *p < 0.05 comparedwith baseline. PVRI is expressed in dynes· sec . cm- 5 • m- 2. [H+] is expressed in nmol/L. PC02 is expressed in millimeters of mercury.
Table II. Protocol 1: hemodynamic data as a function of respiratory acid-base status Arterial blood gas values pH [H+l Pcoz Poz Hemodynamic variables Temperature (OC) Hemoglobin (gm/dl) HR (beats/min) CVP (rnm Hg) PCWP (mrn Hg) LAP (mm Hg) MAP (mm Hg) mPAP (mm Hg) CI (L/m/m Z) RVSWI (gm-m/rn-) PVRI (dynes. sec . cm" . m-)
7.47 34.4 30.1 256
± ± ± ±
35.2 ± II ± 78 ± 10.6 ± 10.4 ± 9.6 ± 64 ± 18.2 ± 2.54 ± 8.53 ± 294 ±
± ± ± ±
0.01 0.9 0.2 40
7.38 42.7 39.9 247
0.2 05 4 0.9 1.0 1.2 3
35.2 ± 0.2
l.l
0.3 0.6 25
72 ± 10.0 ± 9.9 ± 9.0 ± 65 ± 19.0 ± 2.30 ± 8.74 ± 370 ±
0.01 0.6 0.4 40
4 0.7 0.9 0.1 4 1.2 0.2 0.5 32
7.31 50.1 50.6 236
± ± ± ±
0.01 0.9 0.4 34
35.2 ± 0.2 74 ± 10.1 ± 9.9 ± 9.3 ± 72 ± 22.0 ± 2.42 ± 10.36 ± 424 ±
4 0.6
1.1 l.l 5 1.3* 0.2 0.4* 32*
7.39 40.7 39.9 254
± ± ± ±
0.01 0.6 0.2 38
35.2 ± 0.3 75 9.1 9.4 8.7 76 19.3 2.28 8.52 384
± ± ± ± ± ± ± ± ±
4 0.9 0.9 l.l 5* 1.5 0.1 0.5 33
7.47 34.2 30.8 217
± ± ± ±
0.01 0.6 0.7 36
35.2 ± 0.2 72 8.7 9.9 9.4 75 18.4 2.27 8.46 320
± ± ± ± ± ± ± ± ±
4 0.6 l.l 1.3 5 1.3 0.5 0.6 40
Values are expressed as mean ± standard error of the mean. HR, Heart rate; CVP, central venous pressure; PCWP, pulmonary capillary wedge pressure; LAP, left atrial pressure; MAP. mean arterial pressure; mPAP. mean pulmonary artery pressure; CI, cardiac index; RVSWI, right ventricular stroke work index; PVRI, pulmonary vascular resistance index.
•p < 0.05 versus baseline value.
population consisted of 10 consecutive male patients. Subject demographics are listed in Table III. All patients received long-term aspirin therapy in the preoperative period, and no patients had preoperative pulmonary hypertension. The arterial and mixed venous blood gas values at each point of data collection are listed in Table IV. As in the
first protocol, heart rate, central venous pressure, and left atrial pressure showed no significant changes throughout the study period. Changes in pulmonary vascular resistance index as a function of changes in arterial Pco- and arterial [H+] are illustrated in Fig. 4. Pulmonary vascular resistance index increased from 199 ± 21 dynes. sec . cm- 5 • m- 2 to
The Journal of Thoracic and Cardiovascular Surgery September 1993
5 3 2 Fullerton et al.
480
26
440
*
420
~II
22
400
:i~
ii: > Q.
~l c .§ ~ 18 " !..
12
360 340
.
11
320
Do. cQ. II
::Ii
380
14
'" <
300
10 III
280
~
260
Cii
:%:
14 13
iiE 'i:E ~-
12
~:
11
lID 2
=~
10
45
50
40
9
e
pH
7.37
7.31
IH+J
7.39
8
7.47
42.7
50.1
40.7
34.2
Q.
3.5 N
3.0
:: E ...
2.S
.E!
.. .-" u c
~~ a~
30
7 8
-
30
2.0 1.5 pC02
pH [H']
30 7.47 34.4
40 7.37 42.7
50 7.31 50.1
4 7.39 40.7
30 7.47 34.2
Fig. 2. Changes in mean pulmonaryarterial pressure,cardiac index, and left atrial pressure as a function of respiratory acidbase status in protocol 1. Mean pulmonary arterial pressure increased significantly (p < 0.05), whereas left atrial pressure and cardiac indexdid not change.Thisimpliesthat the increased pulmonary arterial pressure resulted from pulmonary vasoconstrictionas Pco- increasedfrom 30 mm Hg to 50 mm Hg. Mean pulmonary arterial pressure returned to baseline as Pco, returned to 30 mm Hg. Vertical bars represent mean ± standard error of mean. *p < 0.05 compared with baselinevalue. Table III. Protocol 2: demographics No. of patients Age (yr) Body surface area (m 2) Tidal volume (co/kg) Respiratory rate (breaths/min) Aortic crossclamp time (min) Cardiopulmonary bypass time
10 60.1 ± 1.94 ± 10.5 ± 10.4 ± 82.11 ± 1]6.7 ±
3.4 0.05 0.2 0.4 6.4 8.2
Values are expressed as mean ± standard error of the mean.
277 ± 25 dynes. sec . cm- 5 . m- 2 (mean ± standard error of the mean; p < 0.05) as arterial Pco- increased from 29 mm Hg to 52.8 mm Hg (p < 0.05) and [H+] increased from 33.3 nm/L to 52.0 nm/L (p < 0.05).
Fig. 3. Pulmonary vascular resistanceindex (PVRIj and right ventricular stroke work index (RVSWlj as a functionof arterial respiratoryacid-basestatus in protocol I. The increasedpulmonary vascularresistanceindexwasaccompaniedbya significant increase in right ventricular stroke work index (p < 0.05) as Pco, increased from 30 mm Hg to 50 mm Hg. Vertical bars representmean ± standard error of mean. *p < 0.05 compared with baseline value. Pulmonary vascular resistance index is expressed in dynes . sec· cm- 5 • m- 2.Rightventricularstroke work index is expressed in grammeters per meter squared. When arterial Pco- was held constant at 53.9 mm Hg and [H+] was reduced to 42.8 nm/L by the addition of sodium bicarbonate, resistance index returned to baseline. This reduction in [H+] was statistically significant (p < 0.05). Again, when arterial Pco- was held constant at 53.2 mm Hg, pulmonary vascular resistance index returned to 268 ± 27 dynes. sec . cm- 5 • m- 2 with the addition of hydrochloric acid (p < 0.05). Because of the withdrawal of carbon dioxide from the anesthesia breathing circuit, arterial Pco- returned to 32.3 mm Hgand [H+] returned to 33.7 nmol/L, This was accompanied by a return to baseline of pulmonary vascular resistance index. Thus, when arterial Pco, was held constant, pulmonary vascular resistance index was modulated by changes in [H+]. Fig. 5 illustrates the changes in pulmonary vascular resistance index as a result of the changes in mixed venous Pco- and [H+]. These changes correlated with changes in [H+] as mixed venous Pcoj was held constant. At the final point of data collection, mixed venous Pco- returned to 40.1 mm Hg, which was significantly higher than the baseline value of 35.5 mmHg(p < 0.05). However, [H+] returned to baseline, as did pulmonary vascular resistance index.
The Journal of Thoracic and Cardiovascular Surgery Volume 106, Number 3
Fullerton et al.
533
Table IV. Protocol 2: hemodynamic data as a function of[ W } and Pco, Arterial blood gas values pH [H+] PC02 P02 Mixed venous blood gas values pH [H+] PC02
P02 Hemodynamic variables Temperature (0C) Hemoglobin (gm/dl) HR (beats/min) CVP (mmHg) PCWP(mm Hg) LAP (mmHg) MAP (mmHg) mPAP (mmHg) CI (L/m/m 2) PVRI (dynes. sec . em"? . m- 2)
7.49 33.3 29.0 220
± ± ± ±
0.01 0.8 0.6 32
7.29 52.0 52.8 275
± ± ± ±
0.01 1.0 1.3 36
7.37 42.8 53.9 250
± ± ± ±
0.01 0.8 1.0 30
7.30 49.8 53.2 280
± ± ± ±
0.01 1.0 1.0 39
7.49 33.7 32.3 310
± ± ± ±
7.44 36.8 35.5 41.3
± ± ± ±
0.01 0.7 l.l l.l
7.28 52.4 54.5 53.9
± ± ± ±
0.02 l.l 1.5 1.6
7.36 44.5 57.4 55.0
± ± ± ±
0.01 0.9 1.3 1.2
7.29 52.1 58.4 57.2
± ± ± ±
0.01 1.2 0.8 1.3
7.42 38.4 40.1 44.5
± 0.01 ± 0.9 ± 0.6 ± 0.8
35 9 75 10 13.8 13 65 22 3.02 277
± ± ± ± ± ± ± ± ± ±
1.6 0.3 3 1.6 1.0 1.7 2.7 1.3* 0.2 25*
35 9 78 II 16 15 75 22 3.10 195
± ± ± ± ± ± ± ± ± ±
0.2 0.3 4 1.3 1.6 1.6 4.0 1.6* 0.3 23
35 9 80 II II.8 II 79 24 3.30 268
± ± ± ± ± ± ± ± ± ±
0.2 0.3 4 1.0 0.6 1.0 6.0 1.3* 0.2 27*
35 9 80 10 13.7 13 86 19 2.99 198
± ± ± ± ± ± ± ± ± ±
35 ± 0.2 9 ± 0.3 75 ± 3 9 ± 1.0 12.5 ± 1.3 12 ± 1.6 66 ± 3.0 19 ± 1.3 2.97 ± 0.2 199 ± 21
0.01 0.9 l.l 35
0.2 0.3 4 0.6 1.3 1.6 6.0* 1.0 0.2 35
Values are expressed as mean ± standard error of the mean. For abbreviations, see Table II. • p < 0.05 versus baseline value.
Discussion The influence of respiratory acid-base status on neonatal pulmonary vascular resistance is well appreciated. However, its influence in the adult has received considerably less attention. Previous studies in adults have shown either no change? or an increased pulmonary vascular resistance'" 10 when awake subjects inhaled carbon dioxide, However, with the examination of awake, spontaneously breathing subjects, these studies lacked the ability to control other factors that influence pulmonary vascular resistance, such as minute ventilation, functional residual capacity, tidal volume, and oxygen. II Also, without accurate determination of pulmonary outflow pressure (left atrial pressure), it is difficult to accurately determine pulmonary vascular resistance. 11 When observed in adult patients undergoing cardiac operations, hypercarbia increased pulmonary vascular resistance after cardiopulmonary bypass.P However, pulmonary vascular resistance did not change with alterations in respiratory acid-base status in pediatric patients undergoing cardiac operation.!" Although it is apparent that respiratory acid-base status may significantly influence pulmonary vascular resistance, some animal data have attributed the effect to PC02,15 whereas other studies have implicated [H+] .16 Studies with human patients have shown an elevation in pulmonary vascular resistance in response to intravenous acid infusion.'? However, these studies again comprised
awake, spontaneously breathing patients whose minute ventilation increased in response to acid infusion. The studies were therefore limited by an inability to control PC02, oxygen, and the mechanics of ventilation, and to measure pulmonary outflow pressure. In this regard, the patient undergoing cardiac operation offered a unique opportunity to examine the influence of PC02 and [H+] on the adult pulmonary vascular bed. Surgical access allowed determination of left atrial pressure, which, in turn, allowed for accurate determination of pulmonary outflow pressure and pulmonary vascular resistance. A constant rate of ventilation and tidal volume was maintained in the anesthetized patient to avoid mechanical alterations of pulmonary vascular resistance. Furthermore, P02 was controlled well in these patients, and hypoxia was avoided. Evaluation of the influence of respiratory acid-base status on pulmonary vascular resistance in patients supported with mechanical ventilation almost certainly requires anesthesia. In our protocols, a standard cardiac anesthetic technique was used. Anesthetic agents were administered before cardiopulmonary bypass only, and the data were collected after bypass. Therefore, any influence of anesthesia on pulmonary vascular resistance was held reasonably constant during the period of data collection. Although anesthesia mayor may not influence the response of the pulmonary vasculature to acid-base changes, the influence was consistent during the period of
The Journal of Thoracic arid Cardiovascular Surgery September 1993
5 3 4 Fullerton et al.
c;l: 300
N
l:
":280 fi280
. ~
'If
*
p c 0.05
240
l:
l:
200
~
:2- 180 ii: >160
Q.
55
:i s
50
+'
45
=. ..J c(
ii:
iii:
........--~--..--
=---~-----.---
*
p c 0.05
c(
§.
o o
44
W
40
X
3.
32
~
60
§.
55
p c 0.05
~
50
~
45
ifi
40
ii:
35
fa
35
c(
30
::IE
30
~
+---.-------,.-----r----.----.------,
*
N
*
40
Q. ..J c(
*pcO.05
48
E
*
50
180
~
W
55
200
52
i 60
le
220
~
Q
3O+-----.-----,--r-----,.---,-----,
240
.s
!
*pcO.05
260
160
> 35
260
~
oZ
40
w
lll:
...,;lii U
•>-
";220
•>-
E
300
45
o
>
~
*pcO.05
* +-_ _.---_---.-_ _-.-_ _..--_---,._ _-t-t #1
25.1---.-------,.-----.---........---.-----. o # 5 #' #2 #3 #4 Points of
Data
Collection
Fig. 4. Changes in pulmonary vascular resistance index (PVRJj as a functionof changesin arterial [H+], whenarterial Pco, isheldconstant.Changesinpulmonaryvascularresistance indexcorresponded to changesin [H+]. Vertical bars represent mean ± standard error of mean. *p < 0.05 compared with baseline value.
data collection. In addition, to optimize clinical relevance, we examined patients in the early postoperative period. The present study demonstrates an important influence of respiratory acid-base status on adult pulmonary vascular resistance after cardiac operation. This finding is consistent with those by Salmonpera and Heinoven.P The present study goes beyond this, however, and demonstrates that in changing [H+] while Pco- remained constant, it is apparent that the influence is mediated by [H+] rather than by Pco-. The hemodynamic data of this study may offer some insight into the area of action of [H+]. Experimental data have demonstrated that both the arterial and venous sides of the pulmonary capillary bed may vasoconstrict in response to a rise in [H+] .18 In the present study, the relationship between pulmonary capillary wedge pressure and left atrial pressure remained constant as pulmonary
#2
Points
#3
01 Data
#4
#5
Collection
Fig. 5. Changes in pulmonary vascular resistance index (PVRJj as a functionof mixedvenous [H+] and Pco-, Changes in pulmonaryvascularresistance indexcorresponded tochanges in [H+]. Vertical bars represent mean ± standard error of mean. *p < 0.05 compared with baseline value.
vascular resistance increased. This suggests that under the conditions of this study, the rise in resistance occurred as a result of increased pulmonary vascular tone proximal to the capillary level at the pulmonary arterioles. Possible explanations for the influence of [H+] on pulmonary vascular resistance include local mediation by prostanoids.l" thromboxane.l? and catecholamines.P or local changes in ionized calcium.I' Perhaps the most likely influence of [H+] is on pulmonary vascular smooth muscle tone. Increased intracellular [H+] leads to smooth muscle contraction. Although the exact mechanism is unknown, [H+] does influence both the activity of tropomyosin-? and intracellular calcium concentration in smooth muscle cells.23, 24 Because the mixed venous Pee, and [H+] so closely mirrored the arterial values, it is not possible to distinguish the influence of arterial from venous values in this study; but, because the mixed venous blood actually perfuses the pulmonary circulation, it seems logical to infer that the
The Journal of Thoracic and Cardiovascular Surgery Volume 106, Number 3
mixed venous [H+] may be the active variable. Nonetheless, except in states of low cardiac output" the mixed venous Pco- and [H+] are predictably related to their respective arterial valuea" Therefore, in most clinical settings, the arterial values may be used. We believe that these data have important clinical relevance, especially in the management of patients supported with mechanical ventilation after cardiac operation. Such patients are at particular risk for the development of respiratory acid-base disturbances. Because pulmonary vascular resistance is the primary clinical determinant of right ventricular afterload, a significant increase in pulmonary vascular resistance induced by hypercarbic acidemia may increase right ventricular work by producing right ventricular afterload mismatch.l? In fact, the results of the present study demonstrate a significant increase in right ventricular stroke work index caused by increased pulmonary vascular resistance during hypercarbic acidemia. Furthermore, right ventricular ejection fraction has been shown to be reduced as a result of increased pulmonary vascular resistance induced by hypercarbic acidemia. 28, 29 These undesirable influences on right ventricular function may in turn produce hemodynamic compromise. Therefore, respiratory acidemia should be avoided in the perioperative period with patients undergoing cardiac operation. On the other hand, these data suggest that respiratory alkalemia may be effectively used to minimize pulmonary vascular resistance in adults and neonates and thereby reduce right ventricular afterload. In conclusion, this study demonstrates an important role for respiratory acid-base status in the determination of pulmonary vascular resistance in the adult patient undergoing cardiac operation. These data also suggest that the influence is mediated by [H+] rather than by Pco-, Therefore, in the clinical management of respiratory acid-base status in patients supported with mechanical ventilation after cardiac operation, attention should be focused on pH rather than on Pco-.
REFERENCES I. Barer GR, Howard P, Marlow JR. The effectof hypercapnia and blood pH on the pulmonary circulation. J Physiol 1966;182:29-30. 2. Bergofsky EH, Haas F, Porcelli R. Determination of the sensitive vascular sites from which hypoxiaand hypercapnia elicit rises in pulmonary arterial pressure. Fed Proc 1968;27:1420-5. 3. Linde LM, Simmons DH, Lewis N. Pulmonary hemodynamicsin respiratory acidosisin dogs. Am J Physiol 1963; 205:1008-12. 4. Morin FC, Lindgren L, Marshall BE. Metabolic and res-
Fullerton et al.
535
piratory hydrogen ion effects on hypoxic pulmonary vasoconstriction. J Appl Physiol 1984;57:545-50. 5. Peckam GJ, Fox WW. Physiologic factors affecting pulmonary artery pressure in infants with persistent pulmonary hypertension. J Pediatr 1978;93:1005-10., 6. Dawes GS. Fetal and neonatal physiology. Chicago: Year Book, 1968:98. 7. Bergofsky EH, Lehr DE, Tuller MA, Rigatto M, Fishman AP. The effectsof acute alkalosisand acidosison the pulmonary circulation. Ann N Y Acad Sci 1962;92:627-34; 8. Figueras J, Stein L, Diez V, Wei! NM, Shubin H. Relationshipbetweenpulmonary hemodynamics and arterial pH and carbon dioxidetension in critically ill patients. Chest 1976;70:466-72. 9. FishmanAP, Fitts HW, Cournand A. Effects of breathing carbon dioxide on the pulmonary circulation. Circulation 1960;22:220-5. 10. Kilburn KH, Asmundson T, Britt RC, Cardon R. Effects of breathing 10 percent carbon dioxide on the pulmonary circulationof human subjects.Circulation 1969;39:639-53. 11. Fishman AP. Pulmonary circulation. In: Fishman AP, ed. Handbook of physiology: the respiratory system. Vol. 1. Bethesda, Md: American Physiological Society, 1985:93165. 12. Davenport HW. The ABC of acid-base chemistry. 6th ed. Chicago: The University of Chicago Press, 1974. 13. Salmonpera M, HeinovenJ. Pulmonary vascular responses to moderate changes in paCO z after cardiopulmonary bypass. Anesthesiology 1986;64:311-5. 14. Morray JP, Lynn AM, Mansfield PB. Effect of pH and pCOz on pulmonary and systemic hemodynamics after surgery in children with congenital heart disease and pulmonary hypertension. J Pediatr 1988;113:474-9. 15. Malik AB, Kidd BSL. Independent effects of changes in [H+] and COz concentration on hypoxic pulmonary vasoconstriction. J Appl PhysioI1973;34:483-8. 16. Schreiber MD, Heymann MA, Soifer SJ. Increased arterial pH, not decreased pCOz attenuates hypoxia induced pulmonaryvasoconstriction in newbornlambs. Pediatr Res 1986;20:113-7. 17. Housley E, Clark SW, Hedsworth-Whitty B, Bishop JM. Effect of acute and chronic acidemia and associated hypoxia on the pulmonary circulation of patients with chronic bronchitis. Cardiovasc Res 1970;4:482-9. 18. Koyama T, Horimoto M. Pulmonary microcirculatory response to localized hypercapnia. J Appl Physiol 1982; 53:1556-64. 19. Shams H, Peskar BA, Scheid P. Acid infusion elicits thromboxane Az-mediated effects on respiration and pulmonary hemodynamics in cats. Respir Physiol 1988; 71:169-83. 20. Si!ove ED, Inone T, Grover RF. Comparison of hypoxia, pH and sympathomimeticdrugs on bovinepulmonary vasculature. J Appl PhysioI1968;24:355-65. 21. Marshall C, Lindgren L, Marshall BE. Metabolic and respiratory hydrogen ion effects on hypoxic pulmonary vasoconstriction. J Appl Physiol 1984;57:545-50.
5 3 6 Fullerton et al.
22. Yamaguclin M, Ver A, Carlos A, Seidel JC. Modulation of the actin-activated adenosine triphosphatase activity of myosin by tropomyosin from vascular and gizzard smooth muscle. Biochemistry 1984;23:774-9. 23. Busa WB, Nuccitelli R. Metabolic regulation via intracellular pH. Am J PhysioI1984;246:R409-38. 24. Wray S. Smooth muscle intracellular pH: measurement, regulation and function. Am J PhysioI1988;254:C213-25. 25. Wei! MH, Rackon EC, Trevion R, Grundler W, Falk JL, Griffel MI. Difference in acid-base state between venous and arterial blood during cardiopulmonary resuscitation. N Engl J Moo 1986;315:153-6. 26. Steinbert JJ, Harken AH. The central venous catheter in
The Journal of Thoracic and Cardiovascular Surgery September 1993
the assay of acid base status. Surg Gynecol Obstet 1981; 152:221-2. 27. Ross J Jr. Afterload mismatch in aortic and mitral valve disease and implications for surgical therapy. J Am Coli Cardiol 1985;5:811-26. 28. Rose CE, Van Ben Thuysen K, Jackson JT, et al. Right ventricular performance during increased afterload impaired by hypercapnic acidosis in conscious dogs. Circ Res 1983;52:76-84. 29. Viitanen A, Salmonpera M, Heinoven J. Right ventricular response to hypercarbia after cardiac surgery. Anesthesiology 1990;73:393-400.
Bound volumes available to subscribers Boundvolumes of THE JOURNAL OF THORACIC AND CARDIOVASCULAR SURGERY are availableto subscribers (only)for the 1993 issues from the Publisher,at a costof $75.00 for domestic, $98.25 for Canadian, and $93.00 for internationalsubscribers for Vol. 105 (JanuaryJune) and Vol. 106 (July-December). Shippingcharges are included. Each bound volumecontains a subject and author index and all advertisingis removed. Copiesare shippedwithin 60 days after publication of the last issueofthe volume. The bindingis durable buckram with the JOURNAL name, volumenumber, and year stamped in goldon the spine. Paymentmust accompany all orders. Contact Mosby, SubscriptionServices, 11830 WestlineIndustrialDrive,St. Louis,Missouri63146-3318, USA; phone I (800) 453-4351 or (314) 453-4351. Subscriptions must be in force to qualify. Bound volumes are Dot available in place of a regular JOURNAL subscription.