- Email: [email protected]
Changes in plasma atrial natriuretic peptide concentration during heart transplantation
Changes
in Plasma
Atria1 Natriuretic Peptide Heart Transplantation
Concentration
During
Tetsuhiro Sakai, MD, Terry W. Latson, MD, Charles W. Whitten, MD, David N. O’Flaherty, MB, Dac Vu, MS Ill, Satya Krishnan, MS, James M. Lipton, PhD, and W. Steves Ring, MD Examination of changes in plasma atrial natriuretic peptide (ANP) concentrations during heart transplantation may provide important information about factors influencing plasma ANP in patients with severe heart failure. Serial changes in plasma ANP during heart transplantation, and atrial content of ANP in native and donor atria, were measured in 12 patients. Preoperative plasma ANP was elevated in all patients (387 f 77 pg/mL), whereas atrial content of ANP in native atria was reduced (0.36 f 0.082 Fg/mg protein). Preoperative plasma ANP did not correlate with hemodynamics, but was negatively correlated with creatinine clearance (r = -0.76, P < .Ol). lntraoperative plasma ANP prior to transplantation was strongly correlated with intraoperative plasma ANP after transplantation (r = 0.84, P < .OOl). Although postoperative plasma ANP was reduced from preoperative plasma ANP by 75%. these two measurements were also significantly correlated (r = 0.70, P < .02). Postoperative plasma ANP was not correlated with hemodynamics, but was negatively correlated with both creatinine clearance (r = -0.65, P < .05) and content of ANP in the native atria (r = -0.75, P < .Ol). Multiple linear regression analysis sug-
A
TRIAL natriuretic peptide (ANP) is a potent hormone with natriuretic, vasodilatory, and neuroregulatory actions.‘,? This hormone may play an important role in the regulation of blood pressure, blood volume, and sodium excretion. Plasma concentrations of ANP (plasma ANP) are elevated in patients with congestive heart failure (CHF).2-X The primary etiology of this increase in plasma ANP is thought to be increased atria1 secretion of ANP in response to atria1 stretch. However, several investigators have suggested that other factors may also play a role in the elevations in plasma ANP observed in patients with severe CHF. These factors include: (1) alterations in metabolic clearance; (2) cellular adaptations in atria1 myocytes to maximize ANP synthesis and secretion; and (3) recruitment of ventricular myocytes to synthesize additional ANP. Heart transplantation represents a unique clinical event during which hemodynamics, atria1 chamber size, and myocardial function are abruptly changed. Examinations of changes in plasma ANP during this event might provide additional insight into factors influencing plasma ANP. Strong correlations between pretransplant and posttransplant plasma ANP would provide important evidence that factors other than acute changes in hemodynamics and atria1 stretch play a significant role in determining plasma ANP. The goal of this study was to investigate these perioperative changes in plasma ANP in heart transplant patients. Possible correlations between plasma ANP and
From the Departments of Anesthesiology and Surgev, University of Texas Southwestern Medical Center at Dallas. Address reprint requests to Terry W. Latson, MD, CJ.T. Southwestern Medical Center, Department of Anesthesiology, 5323 Hany Hines Blvd, Dallas, TX 75235-8894. Copyright 0 I992 by W.B. Saunders Company 1053-077019210606-0009$03.00/O 686
Journalof
Cardiothoraoc
gested that up to 85% of the variability of early postoperative plasma ANP could be accounted for by the variability in these latter two parameters. The decrease in native atrial ANP content, in the context of elevated plasma ANP concentration, is consistent with prior animal studies suggesting that severe heart failure induces cellular adaptations favoring accelerated ANP synthesis and secretion (with resultant reduction in tissue content). The significant correlations between pretransplant and posttransplant measurements of plasma ANP suggest that factors that influence variability in plasma ANP between patients prior to heart transplantation remain important determinants of variability in plasma ANP in the first 24 hours after transplantation. These factors appear to be independent of the acute changes in atrial size and myocardial function that accompany heart transplantation, and may involve more slowly adapting biologic mechanisms regulating ANP synthesis, secretion, and clearance in patients with severe heart failure. Copyright 0 1992 by W. B. Saunders Company KEY WORDS: heart surgery, anesthesia, atrium, hormone
tissue concentrations of ANP in both the native and donor atria were also examined. METHODS After institutional Review Board approval, 12 heart transplant patients ranging in age from 43 to 65 years were studied. Informed written consent was obtained from all patients. The patients were sedated with midazolam, 0.03 to 0.05 mgikg, IV, upon arrival in the operating room. Following insertion of a radial artery catheter. anesthesia was induced with sufentanil, 1.5 to 8.5 kg/kg, and/or etomidate, 0.1 to 0.25 mgikg, along with vecuronium, 0.14 to 0. IX mgikg, for muscle relaxation. A pulmonary artery catheter (PAC) was inserted in all patients after induction of anesthesia. Anesthesia was maintained with sufentanil, supplemental isoflurane (0.2% to I%), and muscle relaxants (vecuronium or pancuronium). Ventilation was mechanically controlled using a rate of X to 10 breathsimin and a tidal volume of approximately 10 ml/kg. adjusted to maintain PaC02 at 35 + 5 mmHg. Blood samples for the measurements of plasma ANP levels were drawn from the arterial catheter at the following sampling points: Pl) before induction; P2) after induction; P3) after incision; P4) immediately before cardiopulmonary bypass (CPB); P.5) on CPB, immediately before removal of the old heart: P6) on CPB. immediately before reperfusion of the new heart; P7) on CPB. immediately before weaning from CPB; P8) 10 minutes after discontinuing CPB; P9) 60 minutes after discontinuing CPB; PlO) 24 hours postoperatively. Measurements of central venous pressure (CVP) and pulmonary artery pressure (PAP) were recorded at sample points P3 (the first sample point after insertion of the PAC), P8 (first sample point after CPB), and PlO (first postoperative day). All 4 mL samples of arterial blood were collected in chilled tubes containing EDTA and 100 FL of aprotinin (0.67 trypsin inhibiting units per mL of blood). The samples were immediately placed in ice and centrifuged within 10 minutes (3,000 rpm, 4”C, 15 minutes). After transferral of the platelet-rich plasma to polypropylene Eppendorf tubes (Brinkman. Westbury, NY), the samples were centrifuged in a microfuge for 1 minute. The plasma was stored at and
Vascular Anesthesia,
Vol6, No 6 (December),
1992: pp 686-691
PLASMA ANP AND HEART TRANSPLANTATION
-70°C until the assay was performed. Immunoreactive ANP was measured by radioimmunoassay (Peninsula Laboratories, Belmont, CA). Cross-reactivities of this assay with a-hANP, P-hANP, and r-hANP are lOO%, 94%, and 40%, respectively. Interassay variation with this technique was between 5% and 10%. Atria1 tissue samples were obtained from the nonauricular free wall of the right atrium for both native (n = 11) and donor (n = 8) hearts. In donor hearts, this sample was taken when the excised heart was brought into the operating room (donor hearts were perfused with cardioplegia solution before excision, and then immediately placed in iced saline). For the native heart, this sample was taken within minutes after the native heart (ie, both ventricles and approximately 25% of the atria) was excised. Tissue samples were immediately placed in iced saline, and subsequently frozen in liquid nitrogen and stored at -70°C until assays were performed. For measurements of ANP tissue concentration, the tissue samples were placed in an ice-cold solution of 0.1 mol/L acetic acid with 0.02N HCI and homogenized with a Polytron (VirTishear, Virtis Co, Gardiner, NY) for six 30-second intervals.9 Tissue homogenates were then sonicated for three 20-second intervals using a Virsonic 50 (Virtis Co). ANP in tissue homogenates was then measured using the same radioimmunoassay used for plasma ANP. ANP concentrations are reported as microgram of ANP per milligram of protein. Total protein was assessed by comparing the optical density of dyed tissue homogenates with the density of known standards (Biorad Laboratories, Richmond, CA). Significance of changes in plasma ANP measured at different sample times was assessed with the Wilcoxon signed-ranks test. Correlations between plasma ANP at different sample times, and between tissue ANP and plasma ANP, were assessed using least squares linear regression. The significance of differences in tissue concentrations of ANP in native and donor atria was assessed using the Mann-Whitney II test. For all statistical comparisons, P values less than 0.05 were considered significant. All values are reported as mean 2 SEM. RESULTS
Preinduction plasma ANP (387 ? 77 pg/mL) was elevated in all patients relative to normal values for this assay technique (10 to 50 pg/mL). However, there was considerable variation among patients (range, 177 to 940 pg/mL). No significant correlation was found between plasma ANP and hemodynamics at any of the three sample times for which simultaneous hemodynamic measurements were recorded (P3, P8, PlO). Similarly, there was no correlation between preinduction plasma ANP and hemodynamic parameters (CVP, PAP, pulmonary capillary wedge pressure [PCWP], cardiac index) recorded during prior cardiac catheterization. However, there was a significant negative correlation between preinduction plasma ANP and creatinine clearance (r = -0.76, P < .Ol; Fig 1). This correlation between creatinine clearance and plasma ANP was also present in the postoperative period (r = -0.65, P < .05). There were clear changes in plasma ANP at certain points during transplant surgery (Fig 2). Although there was no significant change in plasma ANP immediately after induction (Pl v P2), there was an approximate 35% decrease during maintenance of anesthesia prior to CPB (P3 and P4). With onset of CPB (P5), plasma ANP decreased to 23% of control level. After excision of the native heart, plasma ANP decreased to only 4% of control
PreTransplant Plasma ANP
I-=-0.76 p<.Ol
(pg/mL) 500
0 20
30
40
50
Creatinine Clearance (mL/min/m*)
60 2
Fig 1. Preoperative plasma ANP concentration plotted against preoperative creatinine clearance. The body surface area used to normalize creatinine clearance was based on ideal body weight. A recent creatinine clearance measurement was not available in one patient who required preoperative placement of an intraaortic balloon pump.
level (P6). By the end of CPB (P7), mean plasma ANP returned to approximately 60% of the control value, and remained at this level 10 minutes after CPB (P8). However, by 60 minutes after CPB (P9), plasma ANP was further reduced to 44% of the control level, and then to 25% of the control level on the first postoperative day. Despite these large variations in intraoperative plasma ANP, and despite implantation of a new heart, significant correlations were still found between pretransplant and posttransplant measurements of plasma ANP. Plasma ANP measured at the end of CPB (P7) and following CPB (P8) were both strongly correlated with prebypass plasma ANP (P4) (r = 0.84, P <:.OOl for P7 and P4; r = 0.86, P < .OOl for P8 and P4; Fig 3). Furthermore, plasma ANP measured 24 hours after transplantation (PlO) was significantly corre-
500
375
Plasma ANP
250
(p&W 125
0
Fig 2. Perioperative changes in plasma ANP concentration. See text for explanation of sample points. Brackets depict standard error of the mean. ‘P c .05; # P c .02; * P e .Ol Sample v control (point 1) +P < .Ol Sample vs initial post-CPB measurement (point 8).
SAKAI ET AL
666
500
500
400
300
300 Plasma ANP IOmin After CPB 200
Plasma ANP
at End of CPB (pg/mL)
200
(pg/mL)
0
v
600 200 400 Plasma ANP Prior to CPB (pg/mL)
o-
800
0
200
400
600
800
Plasma ANP Prior to CPB (pg/mL)
Fig 3. (A) Correlation between plasma ANP concentration measured just before termination of cardiopulmonary bypass (CPB) with plasma ANP measured prior to CPB. (B) Correlation between plasma ANP concentration measured 10 minutes after CPB with plasma ANP measured prior to CPB.
lated with preinduction plasma ANP (Pl) (r = 0.70, P < .02; Fig 4). Tissue concentrations of ANP in the native atria (0.36 k 0.082 p,g/mg protein) were significantly lower than tissue concentrations in the donor atria (2.55 + 0.84 p,g/mg protein, P < .002). There was no significant correlation between native atria1 ANP concentration and preinduction plasma ANP. However, there was a significant negative correlation between native atria1 ANP concentration and plasma ANP measured 24 hours after transplantation (r = -0.75, P < .Ol; Fig 5). There was no significant correlation between tissue concentrations of ANP in the donor atria and any post-bypass measurements of plasma ANP. The combined influence of creatinine clearance and native atria1 ANP concentration on postoperative plasma ANP was assessed using multiple linear regression (SAS analysis system). This analysis revealed a highly significant correlation of both independent parameters with postoperative plasma ANP (P < .0024 for tissue ANP; P < .0019 for creatinine clearance), with an overall r2 value of 0.854.
DISCUSSION
The elevated baseline levels of plasma AN? were expected in these patients with significant CHF.2-8 The wide variation in these baseline levels (177 to 940 pg/mL) has also been observed by other investigators studying similar patient groups. No correlation was found between plasma ANP and ventricular filling pressures. This observation is consistent with the plotted data in two recent reports by Hare et al’ and Haass et alh examining plasma ANP in patients with severe cardiomyopathy. In both reports, there is no apparent correlation between hemodynamics and plasma ANP for patients with PCWP 2 15 mmHg or CVP 28 mmHg. In the present group of 12 patients, only 2 patients had a PCWP less than 15 mmHg and/or a CVP less than 8 mmHg. These observations suggest that correlations between plasma ANP and ventricular filling pressures (which have been reported in other patient groups)‘J7 may not be present in patients with severe cardiomyopathy. The strong correlations observed in these patients bctween pretransplant and posttransplant plasma ANP also suggest that factors other than hemodynamics (and/or 160
160 .
\
Plasma ANP r=0.70 60 40
p<.oz ??
t
20
I
0
200 Pre-Induction
400
. I_j__ . .
140
600
800
Plasma ANP (pg/mL)
I
1000
Fig 4. Correlation between plasma ANP concentration measured on the first postoperative day with plasma ANP measured prior to induction.
PostOperative Plasma ANP (pg/mL)
120
r=-0.75
p<.Ol
.
8CI
-
400
800
1200
Native Atria1 ANP (pg/mg protein) Fig 5. Correlation between plasma ANP concentration measured on the first postoperative day with tissue concentrations of ANP measured in the native atria.
PLASMA
ANP AND HEART TRANSPLANTATION
atria1 stretch) are important determinants of plasma ANP in patients with severe CHF. The r2 values for the correlations between prebypass plasma ANP and plasma ANP at both the end of CPB and 10 minutes after CPB (Fig 3) suggest that up to 70% of the variation in plasma ANP between patients at these latter two times may be accounted for by similar variations in plasma ANP prior to CPB. These strong correlations were present despite the dramatic changes in hemodynamics, atria1 stretch, and atria1 size (via addition of new atria1 tissue) that occurred among these three time points. These observations suggest that factors influencing patient variability in plasma ANP in the prebypass period remained important determinants of patient variability in plasma ANP at both the end of CPB and after separation from CPB. The statistically significant correlation between baseline and postoperative plasma ANP (r = 0.70,P < .02) suggests that these factors influencing interindividual variability in plasma ANP remained influential in the early postoperative period. Both the lack of correlation between plasma ANP and hemodynamics, as well as the persistence of similar interindividual variations in plasma ANP, suggest that factors other than hemodynamics (and/or atria1 stretch) may be important perioperative determinants of plasma ANP in these patients with severe cardiomyopathy undergoing transplantation. Such factors may relate to slowly adapting cellular mechanisms controlling ANP synthesis and secretion, or to factors controlling metabolic clearance. Suggestive evidence that slowly adapting cellular mechanisms controlling secretion of ANP may play a role in the perioperative changes in plasma ANP observed in these patients includes: (1) the reduced atria1 content of ANP in native atria; and (2) the negative correlation between atria1 ANP content and postoperative plasma ANP. The reduced atria1 content of ANP is consistent with prior animal studies, suggesting that the decreased content reflects a state of accelerated synthesis nnd secretion of ANP. Reductions in tissue levels of ANP (concomitant with increased plasma ANP) have been reported in cardiomyopathic hamsters,t0-12 hypertensive rats,t3J4 and rats with extensive myocardial infarctions. 15~1h Structural changes in atria1 cardiocytes accompanying these decreased tissue levels include (1) a decrease in the number, size, and ANP content of secretory granules in atria1 cardiocytes; (2) an increase in the rough endoplasmic reticulum population; (3) an increase in the size and complexity of the Golgi complext2; and (4) an increase in ANP mRNA.t6 Cantin et al’* suggested that these changes are consistent with a state of maximal ANP synthesis and secretion, with synthesis not keeping pace with secretion. l2 These cellular adaptations may be responsible for the decreased influence of acute changes in atria1 stretch on ANP release in severe CHF.” Because these structural cellular changes would take time to resolve, such cellular changes may be part of the nonhemodynamic factors influencing plasma ANP in these patients. This hypothesis is supported by the finding of a significant negative correlation between native atria1 ANP concentration and posttransplant plasma ANP (r = -0.75, P < .Ol; Fig 5). A similar negative correlation between tissue and plasma concentrations of ANP has been re-
ported in a rat model of chronic CHF.15 The reason why a correlation is evident in the posttransplant period, but not in the pretransplant period, is unclear. One possible explanation is the removal of the failing ventricles as a variable source of additional ANP. Although ventricular tissue is thought to play a very minimal role in ANP secretion in normal hearts, prior animal studies have shown that ventricular tissue in failing hearts may be “recruited” for ANP production. 12,16Because ventricular ANP secretion may be constitutive rather than regulated,12 this may be another factor accounting for the lack of correlation between filling pressures and plasma ANP in these patients. Reduced atria1 content of ANP has not been previously demonstrated in humans with severe CHF. Sugawara et al’* found a significant increase in atria1 ANP content in patients with class 3 or 4 CHF compared to patients with class 1 or 2 CHF. The data did suggest a possible decrease in atria1 ANP in the class 4 patients (n = 3) compared to the class 3 patients (n = 5) but this decrease was not significant with this small number of patients. Naruse et alI9 also demonstrated an increase in atria1 ANP content in patients with mitral valve disease compared to patients without CHF. However, these authors also found that in these patients with mitral valve disease (1) left atria1 ANP content was lower than right atria1 ANP content; and (2) left atria1 ANP content did not correlate with hemodynamits, whereas right atria1 ANP content did. These authors speculated that these two findings might be explained by accelerated (and less regulated) ANP secretion in left atria1 tissue due to increased left atria1 stretch secondary to mitral valve disease. Both of these studies raise the possibility that human atria1 ANP content might increase with initial heart failure, and then decrease as failure progresses. The finding of a decrease in human atria1 ANP content, in contrast to the findings of Sugawara and Naruse, may therefore result from the more advanced state of CHF present in these patients. However, the authors believe another important difference between these two studies and the present one is the site from which atria1 tissue was sampled. This difference may also explain why the present results are more consistent with prior animal studies. Two animal studies have shown increased ANP concentrations in auricular atria1 tissue as compared to nonauricular tissue.9J3 Both Sugawarats and Naruse’” sampled atria1 tissue from the auricular appendages because a small amount of this tissue is often excised for CPB cannulation and/or for surgical exposure. Due to surgical considerations, the authors sampled tissue from the nonauricular free wall. Atria1 tissue in this free wall may be exposed to greater stress with volume overload than tissue in the auricular appendages (due to the small radius of curvature of the appendage) and, hence, respond differently than auricular tissue to progressive CHF. Because the auricular appendage represents only a small portion of the total atrium, the results using samples of the free wall may be more analogous to prior animal studies in which the atria were analyzed en bloc.*“-12J4-16These results (ie, a decrease in atria1 ANP content with severe CHF) are consistent with these animal studies. A significant negative correlation was also observed
690
SAKAI ET AL
between plasma ANP and creatinine clearance (r = -0.76, P < .Ol). This is the first documentation of a significant correlation between altered renal function and elevated plasma ANP in patients with CHF. As suggested by Raine et al,’ this correlation may relate to reduced ANP clearance secondary to reduced renal perfusion in these patients. The relative contribution of the kidneys to total ANP clearance is controversial.2n-2z However, this explanation is consistent with (1) the high concentrations of ANP clearance receptors (C-receptors) and atriopeptidase in renal tissue20-23; (2) the demonstration by radioautography in rats that the kidney is a predominant organ involved in ANP clearance**; (3) reported decreases in ANP clearance*’ and elevations of plasma ANP24 with decreased glomerular filtration rate; and (4) the relative importance of the kidneys as a target organ for ANP. Because ANP clearance was not measured, it is not clear whether this relationship is due to reduced metabolic clearance of ANP. Alternative hypotheses to explain the observed correlation include (1) the elevation in plasma ANP is a physiologic response to decreased renal function rather than a result of decreased renal clearance; and (2) the decrease in creatinine clearance simply reflects the physiologic severity of CHF, and some other undefined factors associated with worsening CHF caused the increased plasma ANP. The combined influence of preoperative creatinine clearance and native tissue ANP concentration on postoperative plasma ANP was assessed using multiple linear regression analysis. This analysis suggested that up to 85% of the variability in early postoperative plasma ANP could be accounted for on the basis of patient variability in these two parameters. Although this analysis does not identify the primary factors responsible for the 75% reduction in postoperative plasma ANP, it does suggest that factors unrelated to acute changes in hemodynamics and atrial stretch are important determinants of early postoperative variations in plasma ANP between patients. This analysis suggests that these factors may involve: (1) structural cellular adaptations influencing ANP synthesis and secretion in CHF; and (2) renal clearance of ANP. The pattern of changes in plasma ANP observed during CPB (an initial decrease, followed by an increase at the end of CPB) was qualitatively similar to that previously reported in nontransplant cardiac surgical patients. Proposed mechanisms for the initial decrease in plasma ANP include decreased atria1 perfusion, decreased heart rate, hemodilution, hypothermia, and atria1 decompression.25-*6 Secretion of ANP by other organs may contribute to the residual ANP detected after virtual exclusion of the heart from the
circulation at point P6.*’ Possible mechanisms responsible for the increase in ANP at the end of CPB (while the atria are still decompressed) include physiologic response to rewarming, tachycardia, elevated catecholamines, atrial “reperfusion,” and atria1 ischemia or trauma during CPB.25.2h,28J9In most other studies,2s,2b,29-33plasma ANP at the end of CPB (or shortly after CPB) either returned to or exceeded baseline plasma ANP. In this study, plasma ANP at the end of CPB remained significantly less than baseline plasma ANP. This observation may relate to (1) high basal rate of ANP secretion in these patients with severe CHF; (2) excision of a portion of the native atria (approximately 25%); and (3) removal of the failing ventricles as an additional source of ANP. The magnitude of perioperative reduction in plasma ANP (75% decrease) in these transplant patients was greater than that previously measured in other types of cardiac surgery. XX~‘) Based on prior studies in animals and humans, this large change in plasma ANP may have significant physiologic effects. Studies using infusions of ANP to produce acute elevations in plasma ANP have demonstrated reversible effects of ANP on renal function, baroreflexes, atrial pressures, systemic vascular resistance, and cardiac function.2.34-37 A more recent study has also demonstrated significant effects from acute reductions in ANP using ANP antibodies.?s Injection of ANP antibodies in rats with CHF caused significant increases in right atrial pressure, left ventricular end-diastolic pressure, and systemic vascular resistance. Whether this reduction in ANP following heart transplantation might also play a role in the rapid reversal of baroreflex dysfunction following transplantation’” has not been investigated. In summary, these results suggest that factors other than hemodynamics may be important determinants of plasma ANP concentration in the first 24 hours after heart transplantation. Regression analysis demonstrated that creatinine clearance and tissue concentration of ANP in the native atria were both negatively correlated with early postoperative plasma ANP. The reduced content of ANP in the native atria, as well as the negative correlation between native atrial ANP content and early postoperative plasma ANP, are both consistent with prior animal studies suggesting that severe CHF leads to cellular adaptations limiting ANP storage and promoting ANP secretion. ACKNOWLEDGMENT The authors wish to thank Ann Johnson for her excellent secretarial assistance in preparation of the manuscript, and Nancy Johnson for her vital help in collecting patient data.
REFERENCES 1. Needleman P, Greenwald JE: Atriopeptin: A cardiac hormone intimately involved in fluid, electrolyte, and blood pressure hemostasis. N Engl J Med 314:828-834, 1986 2. Cody RJ, Atlas SA, Laragh JH, et al: Atria1 natriuretic factor in normal subjects and heart failure patients. Plasma levels and renal, hormonal, and hemodynamic responses to peptide infusion. J Clin Invest 78:1362-1374,1986 3. Bates ER, Shenker
Y, Grekin
RJ: The relationship
between
plasma levels of immunoreactive atrial natriuretic hormone and hemodynamic function in man. Circulation 73:1155-1161, 1986 4. Burnett JC, Kao PC, Hu DC, et al: Atrial natriuretic peptide elevation in congestive heart failure in the human. Science 231:11451147,1986 5. Hare JM, Baughman KL, Kass DA, et al: Influence of dilated cardiomyopathy, myocarditis and cardiac transplantation on the relation between plasma atria1 natriuretic factor and atrial pressures. Am J Cardiol67:391-397, 1991
PLASMA
691
ANP AND HEART TRANSPLANTATION
6. Haass M, Deitz R, Fischer TA, et al: Role of right and left atrial dimensions for release of atrial natriuretic peptide in left-sided valvular heart disease and idiopathic dilated cardiomyopathy. Am J Cardiol62:764-770,1988 7. Raine AEG, Phil D, Erne P, et al: Atrial natriuretic peptide and atrial pressure in patients with congestive heart failure. N Engl J Med 315:533-537,1986 8. Arendt RM, Gerbes AL, Ritter D, et al: Atrial natriuretic factor in plasma of patients with arterial hypertension, heart failure or cirrhosis of the liver. J Hyperten 4:S131-S1351986 9. Cernacek P, Maher E, Crawhall JC, Levy M: Molecular forms of atrial natriuretic peptides in dog atrium and plasma. 42:25332539,1988 10. Chimosky JE, Spielman WS, Brandt MA, Heidemann SR: Cardiac atria of BIO 14.6 hamsters are deficient in natriuretic factor. Science 223:820-822,1984 11. Franch H, Callahan LT, Blaine EH: Plasma and atrial content of atrial natriuretic factor in cardiomyopathic hamsters. Life Sci 39:1151-1159, 1986 12. Cantin M, Thibault G, Ding J, Gutkowska J, et al: ANF in experimental congestive heart failure. Am J Pathol 130:552-568, 1988 13. Morii N, Nakao K, Kihara M, et al: Decreased content in left atrium and increased plasma concentration of atria1 natriuretic polypeptide in spontaneously hypertensive rats (SHR) and SHR stroke-prone. Biochem Biophys Res Commun 135:74-81,1986 14. Sonnenberg A, Milojevic S, Chong CK, Veress AT: Atrial natriuretic factor: Reduced cardiac content in spontaneously hypertensive rats. Hypertension 5:672-675,1983 15. Tsundoda K, Hodsman GP, Sumithran E, Johnston CI: Atrial natriuretic peptide in chronic heart failure in the rat: A correlation with ventricular dysfunction. Circ Res 59:256-261,1986 16. Drexler H, Hanze J, Finckh M, Lu W, et al: Atrial natriuretic peptide in a rat model of cardiac failure. Atria1 and ventricular mRNA, atria1 content, plasma levels, and effect of volume loading. Circulation 79:620-633,1989 17. Arendt RM, Gerbes AL, Stangl E, et al: Baseline and stimulated ANF plasma levels: Is an impaired stimulus-response coupling diagnostically meaningful? Klin Wochenschr 65:122-126, 1987 18. Sugawara A, Kakao K, Morii N, et al: Synthesis of atrial natriuretic polypeptide in human failing hearts. Evidence of altered processing of atria1 natriuretic polypeptide precursor and augmented synthesis of B-human ANP. J Clin Invest 81:1962-1970, 1988 19. Naruse M, Hiroe M, Naruse K, et al: Increased levels of B-human atria1 natriuretic peptide-like immunoreactivity in chronically overloaded atrial tissue. Am J Hypertens 3:105-110, 1990 20. Gerbes AL, Vollmar AM: Degradation and clearance of atrial natriuretic factors (ANF). Life Sci 47:1173-1180, 1990 21. Woods RL: Contribution of the kidney to metabolic clearance of atria1 natriuretic peptide. Am J Physiol 255:E934-E941, 1988 22. Chondra CL, Leidy EA, Bunting P, et al: Clearance and early hydrolysis of atrial natriuretic factor in vivo. Structural analysis of cleavage sites and design of an analogue that inhibits hormone cleavage. J Clin Invest 81:1348-1354,1988
23. Almeida FA, Suzuki M, Scharhrough M, et al: Clearance function of type C receptors of atria1 natriuretic factor in rats. Am J Physiol256:R469-R475, 1989 24. Czekalski S, Michel C, Dussaule JC, et al: Atrial natriuretic peptide and adaptation of sodium urinary excretion in patients with chronic renal failure. Clin Sci 75:243-249, 1988 25. Oyama T, Kudo T, Kudo M, et al: Plasma concentration of o-human atria1 natriuretic polypeptide (o-hANP) during cardiopulmonary bypass under fentanyl anaesthesia. Agressologie 27:687689,1986 26. Kharasch ED, Yeo KT, Kenny MA, Amory DW: Influence of hypothermic cardiopulmonary bypass on atrial natriuretic factor levels. Can J Anaesth 36:545-553,1989 27. Teran N, Rodriguez-Iturbe B, Parra G, Gutkowska J: Atrial natriuretic peptide levels in brain venous outflow during cardiopulmonary bypass in humans: Evidence for extracardiac hormonal production. J Cardiothorac Vast Anesth 5:343-347,199l 28. Bilder GE, Siegl PKS, Schofield TL, Friedman PA: Chronotropic stimulation: A primary effector for release of atrial natriuretic factor. Circ Res 64:799-805, 1989 29. Flezzani P, McIntyre RW, Xuan YT, et al: Atrial natriuretic peptide plasma levels during cardiac surgery. J Cardiothor Anesth 2:274-280, 1988 30. Dewar ML, Walsh G, Chiu CJ, et al: Atrial natriuretic factor: Response to cardiac operation. J Thorac Cardiovasc Surg 96:266-270, 1988 31. Lehot JJ, Piriz H, Carry PY, et al: Influence of cardiac surgery on atria1 natriuretic factor. J Cardiothor Anesth 3:S63, 1989 32. Greeley W, Leslie JB, Su M, et al: Plasma peptide release during pediatric cardiovascular surgery. Anesthesiology 65:A414,1986
atrial natriuretic anesthesia and
33. Hedner J, Towle A, Saltzman L, et al: Atria1 natriuretic factor (ANF) in plasma of patients undergoing cardiopulmonary bypass surgery. Anesthesiology 65:A509, 1986 34. Crozier IG, Ikram H, Gomez HJ, et al: Haemodynamic effects of atrial peptide infusion in heart failure. Lancet 1242-1245, 1986 35. Saito Y, Nakao K, Nishimura K, et al: Clinical application of atrial natriuretic polypeptide in patients with congestive heart failure: Beneficial effects on left ventricular function. Circulation 76:115-124,1987 36. Floras JS, natriuretic factor 1990
Phil D: Sympathoinhibitory effects of atria1 in normal humans. Circulation 81:1860-1873,
37. Kuchel 0, Debinski W, Racz K, et al: An emerging relationship between peripheral sympathetic nervous activity and atria1 natriuretic factor. Life Sci 40:1545-1551,1987 38. Drexler H, Hirth C, Stasch HP, et al: Vasodilatory action of endogenous atria1 natriuretic factor in a rat model of chronic heart failure as determined by monoclonal ANF antibody. Circ Res 66:1371-1380,199O 39. Ellenbogen KA, Mohanty Arterial baroreflex abnormalities orthotopic cardiac transplantation.
PK, Szentpetery S, Thames MD: in heart failure. Reversal after Circulation 79:51-58, 1989
in Plasma
Atria1 Natriuretic Peptide Heart Transplantation
Concentration
During
Tetsuhiro Sakai, MD, Terry W. Latson, MD, Charles W. Whitten, MD, David N. O’Flaherty, MB, Dac Vu, MS Ill, Satya Krishnan, MS, James M. Lipton, PhD, and W. Steves Ring, MD Examination of changes in plasma atrial natriuretic peptide (ANP) concentrations during heart transplantation may provide important information about factors influencing plasma ANP in patients with severe heart failure. Serial changes in plasma ANP during heart transplantation, and atrial content of ANP in native and donor atria, were measured in 12 patients. Preoperative plasma ANP was elevated in all patients (387 f 77 pg/mL), whereas atrial content of ANP in native atria was reduced (0.36 f 0.082 Fg/mg protein). Preoperative plasma ANP did not correlate with hemodynamics, but was negatively correlated with creatinine clearance (r = -0.76, P < .Ol). lntraoperative plasma ANP prior to transplantation was strongly correlated with intraoperative plasma ANP after transplantation (r = 0.84, P < .OOl). Although postoperative plasma ANP was reduced from preoperative plasma ANP by 75%. these two measurements were also significantly correlated (r = 0.70, P < .02). Postoperative plasma ANP was not correlated with hemodynamics, but was negatively correlated with both creatinine clearance (r = -0.65, P < .05) and content of ANP in the native atria (r = -0.75, P < .Ol). Multiple linear regression analysis sug-
A
TRIAL natriuretic peptide (ANP) is a potent hormone with natriuretic, vasodilatory, and neuroregulatory actions.‘,? This hormone may play an important role in the regulation of blood pressure, blood volume, and sodium excretion. Plasma concentrations of ANP (plasma ANP) are elevated in patients with congestive heart failure (CHF).2-X The primary etiology of this increase in plasma ANP is thought to be increased atria1 secretion of ANP in response to atria1 stretch. However, several investigators have suggested that other factors may also play a role in the elevations in plasma ANP observed in patients with severe CHF. These factors include: (1) alterations in metabolic clearance; (2) cellular adaptations in atria1 myocytes to maximize ANP synthesis and secretion; and (3) recruitment of ventricular myocytes to synthesize additional ANP. Heart transplantation represents a unique clinical event during which hemodynamics, atria1 chamber size, and myocardial function are abruptly changed. Examinations of changes in plasma ANP during this event might provide additional insight into factors influencing plasma ANP. Strong correlations between pretransplant and posttransplant plasma ANP would provide important evidence that factors other than acute changes in hemodynamics and atria1 stretch play a significant role in determining plasma ANP. The goal of this study was to investigate these perioperative changes in plasma ANP in heart transplant patients. Possible correlations between plasma ANP and
From the Departments of Anesthesiology and Surgev, University of Texas Southwestern Medical Center at Dallas. Address reprint requests to Terry W. Latson, MD, CJ.T. Southwestern Medical Center, Department of Anesthesiology, 5323 Hany Hines Blvd, Dallas, TX 75235-8894. Copyright 0 I992 by W.B. Saunders Company 1053-077019210606-0009$03.00/O 686
Journalof
Cardiothoraoc
gested that up to 85% of the variability of early postoperative plasma ANP could be accounted for by the variability in these latter two parameters. The decrease in native atrial ANP content, in the context of elevated plasma ANP concentration, is consistent with prior animal studies suggesting that severe heart failure induces cellular adaptations favoring accelerated ANP synthesis and secretion (with resultant reduction in tissue content). The significant correlations between pretransplant and posttransplant measurements of plasma ANP suggest that factors that influence variability in plasma ANP between patients prior to heart transplantation remain important determinants of variability in plasma ANP in the first 24 hours after transplantation. These factors appear to be independent of the acute changes in atrial size and myocardial function that accompany heart transplantation, and may involve more slowly adapting biologic mechanisms regulating ANP synthesis, secretion, and clearance in patients with severe heart failure. Copyright 0 1992 by W. B. Saunders Company KEY WORDS: heart surgery, anesthesia, atrium, hormone
tissue concentrations of ANP in both the native and donor atria were also examined. METHODS After institutional Review Board approval, 12 heart transplant patients ranging in age from 43 to 65 years were studied. Informed written consent was obtained from all patients. The patients were sedated with midazolam, 0.03 to 0.05 mgikg, IV, upon arrival in the operating room. Following insertion of a radial artery catheter. anesthesia was induced with sufentanil, 1.5 to 8.5 kg/kg, and/or etomidate, 0.1 to 0.25 mgikg, along with vecuronium, 0.14 to 0. IX mgikg, for muscle relaxation. A pulmonary artery catheter (PAC) was inserted in all patients after induction of anesthesia. Anesthesia was maintained with sufentanil, supplemental isoflurane (0.2% to I%), and muscle relaxants (vecuronium or pancuronium). Ventilation was mechanically controlled using a rate of X to 10 breathsimin and a tidal volume of approximately 10 ml/kg. adjusted to maintain PaC02 at 35 + 5 mmHg. Blood samples for the measurements of plasma ANP levels were drawn from the arterial catheter at the following sampling points: Pl) before induction; P2) after induction; P3) after incision; P4) immediately before cardiopulmonary bypass (CPB); P.5) on CPB, immediately before removal of the old heart: P6) on CPB. immediately before reperfusion of the new heart; P7) on CPB. immediately before weaning from CPB; P8) 10 minutes after discontinuing CPB; P9) 60 minutes after discontinuing CPB; PlO) 24 hours postoperatively. Measurements of central venous pressure (CVP) and pulmonary artery pressure (PAP) were recorded at sample points P3 (the first sample point after insertion of the PAC), P8 (first sample point after CPB), and PlO (first postoperative day). All 4 mL samples of arterial blood were collected in chilled tubes containing EDTA and 100 FL of aprotinin (0.67 trypsin inhibiting units per mL of blood). The samples were immediately placed in ice and centrifuged within 10 minutes (3,000 rpm, 4”C, 15 minutes). After transferral of the platelet-rich plasma to polypropylene Eppendorf tubes (Brinkman. Westbury, NY), the samples were centrifuged in a microfuge for 1 minute. The plasma was stored at and
Vascular Anesthesia,
Vol6, No 6 (December),
1992: pp 686-691
PLASMA ANP AND HEART TRANSPLANTATION
-70°C until the assay was performed. Immunoreactive ANP was measured by radioimmunoassay (Peninsula Laboratories, Belmont, CA). Cross-reactivities of this assay with a-hANP, P-hANP, and r-hANP are lOO%, 94%, and 40%, respectively. Interassay variation with this technique was between 5% and 10%. Atria1 tissue samples were obtained from the nonauricular free wall of the right atrium for both native (n = 11) and donor (n = 8) hearts. In donor hearts, this sample was taken when the excised heart was brought into the operating room (donor hearts were perfused with cardioplegia solution before excision, and then immediately placed in iced saline). For the native heart, this sample was taken within minutes after the native heart (ie, both ventricles and approximately 25% of the atria) was excised. Tissue samples were immediately placed in iced saline, and subsequently frozen in liquid nitrogen and stored at -70°C until assays were performed. For measurements of ANP tissue concentration, the tissue samples were placed in an ice-cold solution of 0.1 mol/L acetic acid with 0.02N HCI and homogenized with a Polytron (VirTishear, Virtis Co, Gardiner, NY) for six 30-second intervals.9 Tissue homogenates were then sonicated for three 20-second intervals using a Virsonic 50 (Virtis Co). ANP in tissue homogenates was then measured using the same radioimmunoassay used for plasma ANP. ANP concentrations are reported as microgram of ANP per milligram of protein. Total protein was assessed by comparing the optical density of dyed tissue homogenates with the density of known standards (Biorad Laboratories, Richmond, CA). Significance of changes in plasma ANP measured at different sample times was assessed with the Wilcoxon signed-ranks test. Correlations between plasma ANP at different sample times, and between tissue ANP and plasma ANP, were assessed using least squares linear regression. The significance of differences in tissue concentrations of ANP in native and donor atria was assessed using the Mann-Whitney II test. For all statistical comparisons, P values less than 0.05 were considered significant. All values are reported as mean 2 SEM. RESULTS
Preinduction plasma ANP (387 ? 77 pg/mL) was elevated in all patients relative to normal values for this assay technique (10 to 50 pg/mL). However, there was considerable variation among patients (range, 177 to 940 pg/mL). No significant correlation was found between plasma ANP and hemodynamics at any of the three sample times for which simultaneous hemodynamic measurements were recorded (P3, P8, PlO). Similarly, there was no correlation between preinduction plasma ANP and hemodynamic parameters (CVP, PAP, pulmonary capillary wedge pressure [PCWP], cardiac index) recorded during prior cardiac catheterization. However, there was a significant negative correlation between preinduction plasma ANP and creatinine clearance (r = -0.76, P < .Ol; Fig 1). This correlation between creatinine clearance and plasma ANP was also present in the postoperative period (r = -0.65, P < .05). There were clear changes in plasma ANP at certain points during transplant surgery (Fig 2). Although there was no significant change in plasma ANP immediately after induction (Pl v P2), there was an approximate 35% decrease during maintenance of anesthesia prior to CPB (P3 and P4). With onset of CPB (P5), plasma ANP decreased to 23% of control level. After excision of the native heart, plasma ANP decreased to only 4% of control
PreTransplant Plasma ANP
I-=-0.76 p<.Ol
(pg/mL) 500
0 20
30
40
50
Creatinine Clearance (mL/min/m*)
60 2
Fig 1. Preoperative plasma ANP concentration plotted against preoperative creatinine clearance. The body surface area used to normalize creatinine clearance was based on ideal body weight. A recent creatinine clearance measurement was not available in one patient who required preoperative placement of an intraaortic balloon pump.
level (P6). By the end of CPB (P7), mean plasma ANP returned to approximately 60% of the control value, and remained at this level 10 minutes after CPB (P8). However, by 60 minutes after CPB (P9), plasma ANP was further reduced to 44% of the control level, and then to 25% of the control level on the first postoperative day. Despite these large variations in intraoperative plasma ANP, and despite implantation of a new heart, significant correlations were still found between pretransplant and posttransplant measurements of plasma ANP. Plasma ANP measured at the end of CPB (P7) and following CPB (P8) were both strongly correlated with prebypass plasma ANP (P4) (r = 0.84, P <:.OOl for P7 and P4; r = 0.86, P < .OOl for P8 and P4; Fig 3). Furthermore, plasma ANP measured 24 hours after transplantation (PlO) was significantly corre-
500
375
Plasma ANP
250
(p&W 125
0
Fig 2. Perioperative changes in plasma ANP concentration. See text for explanation of sample points. Brackets depict standard error of the mean. ‘P c .05; # P c .02; * P e .Ol Sample v control (point 1) +P < .Ol Sample vs initial post-CPB measurement (point 8).
SAKAI ET AL
666
500
500
400
300
300 Plasma ANP IOmin After CPB 200
Plasma ANP
at End of CPB (pg/mL)
200
(pg/mL)
0
v
600 200 400 Plasma ANP Prior to CPB (pg/mL)
o-
800
0
200
400
600
800
Plasma ANP Prior to CPB (pg/mL)
Fig 3. (A) Correlation between plasma ANP concentration measured just before termination of cardiopulmonary bypass (CPB) with plasma ANP measured prior to CPB. (B) Correlation between plasma ANP concentration measured 10 minutes after CPB with plasma ANP measured prior to CPB.
lated with preinduction plasma ANP (Pl) (r = 0.70, P < .02; Fig 4). Tissue concentrations of ANP in the native atria (0.36 k 0.082 p,g/mg protein) were significantly lower than tissue concentrations in the donor atria (2.55 + 0.84 p,g/mg protein, P < .002). There was no significant correlation between native atria1 ANP concentration and preinduction plasma ANP. However, there was a significant negative correlation between native atria1 ANP concentration and plasma ANP measured 24 hours after transplantation (r = -0.75, P < .Ol; Fig 5). There was no significant correlation between tissue concentrations of ANP in the donor atria and any post-bypass measurements of plasma ANP. The combined influence of creatinine clearance and native atria1 ANP concentration on postoperative plasma ANP was assessed using multiple linear regression (SAS analysis system). This analysis revealed a highly significant correlation of both independent parameters with postoperative plasma ANP (P < .0024 for tissue ANP; P < .0019 for creatinine clearance), with an overall r2 value of 0.854.
DISCUSSION
The elevated baseline levels of plasma AN? were expected in these patients with significant CHF.2-8 The wide variation in these baseline levels (177 to 940 pg/mL) has also been observed by other investigators studying similar patient groups. No correlation was found between plasma ANP and ventricular filling pressures. This observation is consistent with the plotted data in two recent reports by Hare et al’ and Haass et alh examining plasma ANP in patients with severe cardiomyopathy. In both reports, there is no apparent correlation between hemodynamics and plasma ANP for patients with PCWP 2 15 mmHg or CVP 28 mmHg. In the present group of 12 patients, only 2 patients had a PCWP less than 15 mmHg and/or a CVP less than 8 mmHg. These observations suggest that correlations between plasma ANP and ventricular filling pressures (which have been reported in other patient groups)‘J7 may not be present in patients with severe cardiomyopathy. The strong correlations observed in these patients bctween pretransplant and posttransplant plasma ANP also suggest that factors other than hemodynamics (and/or 160
160 .
\
Plasma ANP r=0.70 60 40
p<.oz ??
t
20
I
0
200 Pre-Induction
400
. I_j__ . .
140
600
800
Plasma ANP (pg/mL)
I
1000
Fig 4. Correlation between plasma ANP concentration measured on the first postoperative day with plasma ANP measured prior to induction.
PostOperative Plasma ANP (pg/mL)
120
r=-0.75
p<.Ol
.
8CI
-
400
800
1200
Native Atria1 ANP (pg/mg protein) Fig 5. Correlation between plasma ANP concentration measured on the first postoperative day with tissue concentrations of ANP measured in the native atria.
PLASMA
ANP AND HEART TRANSPLANTATION
atria1 stretch) are important determinants of plasma ANP in patients with severe CHF. The r2 values for the correlations between prebypass plasma ANP and plasma ANP at both the end of CPB and 10 minutes after CPB (Fig 3) suggest that up to 70% of the variation in plasma ANP between patients at these latter two times may be accounted for by similar variations in plasma ANP prior to CPB. These strong correlations were present despite the dramatic changes in hemodynamics, atria1 stretch, and atria1 size (via addition of new atria1 tissue) that occurred among these three time points. These observations suggest that factors influencing patient variability in plasma ANP in the prebypass period remained important determinants of patient variability in plasma ANP at both the end of CPB and after separation from CPB. The statistically significant correlation between baseline and postoperative plasma ANP (r = 0.70,P < .02) suggests that these factors influencing interindividual variability in plasma ANP remained influential in the early postoperative period. Both the lack of correlation between plasma ANP and hemodynamics, as well as the persistence of similar interindividual variations in plasma ANP, suggest that factors other than hemodynamics (and/or atria1 stretch) may be important perioperative determinants of plasma ANP in these patients with severe cardiomyopathy undergoing transplantation. Such factors may relate to slowly adapting cellular mechanisms controlling ANP synthesis and secretion, or to factors controlling metabolic clearance. Suggestive evidence that slowly adapting cellular mechanisms controlling secretion of ANP may play a role in the perioperative changes in plasma ANP observed in these patients includes: (1) the reduced atria1 content of ANP in native atria; and (2) the negative correlation between atria1 ANP content and postoperative plasma ANP. The reduced atria1 content of ANP is consistent with prior animal studies, suggesting that the decreased content reflects a state of accelerated synthesis nnd secretion of ANP. Reductions in tissue levels of ANP (concomitant with increased plasma ANP) have been reported in cardiomyopathic hamsters,t0-12 hypertensive rats,t3J4 and rats with extensive myocardial infarctions. 15~1h Structural changes in atria1 cardiocytes accompanying these decreased tissue levels include (1) a decrease in the number, size, and ANP content of secretory granules in atria1 cardiocytes; (2) an increase in the rough endoplasmic reticulum population; (3) an increase in the size and complexity of the Golgi complext2; and (4) an increase in ANP mRNA.t6 Cantin et al’* suggested that these changes are consistent with a state of maximal ANP synthesis and secretion, with synthesis not keeping pace with secretion. l2 These cellular adaptations may be responsible for the decreased influence of acute changes in atria1 stretch on ANP release in severe CHF.” Because these structural cellular changes would take time to resolve, such cellular changes may be part of the nonhemodynamic factors influencing plasma ANP in these patients. This hypothesis is supported by the finding of a significant negative correlation between native atria1 ANP concentration and posttransplant plasma ANP (r = -0.75, P < .Ol; Fig 5). A similar negative correlation between tissue and plasma concentrations of ANP has been re-
ported in a rat model of chronic CHF.15 The reason why a correlation is evident in the posttransplant period, but not in the pretransplant period, is unclear. One possible explanation is the removal of the failing ventricles as a variable source of additional ANP. Although ventricular tissue is thought to play a very minimal role in ANP secretion in normal hearts, prior animal studies have shown that ventricular tissue in failing hearts may be “recruited” for ANP production. 12,16Because ventricular ANP secretion may be constitutive rather than regulated,12 this may be another factor accounting for the lack of correlation between filling pressures and plasma ANP in these patients. Reduced atria1 content of ANP has not been previously demonstrated in humans with severe CHF. Sugawara et al’* found a significant increase in atria1 ANP content in patients with class 3 or 4 CHF compared to patients with class 1 or 2 CHF. The data did suggest a possible decrease in atria1 ANP in the class 4 patients (n = 3) compared to the class 3 patients (n = 5) but this decrease was not significant with this small number of patients. Naruse et alI9 also demonstrated an increase in atria1 ANP content in patients with mitral valve disease compared to patients without CHF. However, these authors also found that in these patients with mitral valve disease (1) left atria1 ANP content was lower than right atria1 ANP content; and (2) left atria1 ANP content did not correlate with hemodynamits, whereas right atria1 ANP content did. These authors speculated that these two findings might be explained by accelerated (and less regulated) ANP secretion in left atria1 tissue due to increased left atria1 stretch secondary to mitral valve disease. Both of these studies raise the possibility that human atria1 ANP content might increase with initial heart failure, and then decrease as failure progresses. The finding of a decrease in human atria1 ANP content, in contrast to the findings of Sugawara and Naruse, may therefore result from the more advanced state of CHF present in these patients. However, the authors believe another important difference between these two studies and the present one is the site from which atria1 tissue was sampled. This difference may also explain why the present results are more consistent with prior animal studies. Two animal studies have shown increased ANP concentrations in auricular atria1 tissue as compared to nonauricular tissue.9J3 Both Sugawarats and Naruse’” sampled atria1 tissue from the auricular appendages because a small amount of this tissue is often excised for CPB cannulation and/or for surgical exposure. Due to surgical considerations, the authors sampled tissue from the nonauricular free wall. Atria1 tissue in this free wall may be exposed to greater stress with volume overload than tissue in the auricular appendages (due to the small radius of curvature of the appendage) and, hence, respond differently than auricular tissue to progressive CHF. Because the auricular appendage represents only a small portion of the total atrium, the results using samples of the free wall may be more analogous to prior animal studies in which the atria were analyzed en bloc.*“-12J4-16These results (ie, a decrease in atria1 ANP content with severe CHF) are consistent with these animal studies. A significant negative correlation was also observed
690
SAKAI ET AL
between plasma ANP and creatinine clearance (r = -0.76, P < .Ol). This is the first documentation of a significant correlation between altered renal function and elevated plasma ANP in patients with CHF. As suggested by Raine et al,’ this correlation may relate to reduced ANP clearance secondary to reduced renal perfusion in these patients. The relative contribution of the kidneys to total ANP clearance is controversial.2n-2z However, this explanation is consistent with (1) the high concentrations of ANP clearance receptors (C-receptors) and atriopeptidase in renal tissue20-23; (2) the demonstration by radioautography in rats that the kidney is a predominant organ involved in ANP clearance**; (3) reported decreases in ANP clearance*’ and elevations of plasma ANP24 with decreased glomerular filtration rate; and (4) the relative importance of the kidneys as a target organ for ANP. Because ANP clearance was not measured, it is not clear whether this relationship is due to reduced metabolic clearance of ANP. Alternative hypotheses to explain the observed correlation include (1) the elevation in plasma ANP is a physiologic response to decreased renal function rather than a result of decreased renal clearance; and (2) the decrease in creatinine clearance simply reflects the physiologic severity of CHF, and some other undefined factors associated with worsening CHF caused the increased plasma ANP. The combined influence of preoperative creatinine clearance and native tissue ANP concentration on postoperative plasma ANP was assessed using multiple linear regression analysis. This analysis suggested that up to 85% of the variability in early postoperative plasma ANP could be accounted for on the basis of patient variability in these two parameters. Although this analysis does not identify the primary factors responsible for the 75% reduction in postoperative plasma ANP, it does suggest that factors unrelated to acute changes in hemodynamics and atrial stretch are important determinants of early postoperative variations in plasma ANP between patients. This analysis suggests that these factors may involve: (1) structural cellular adaptations influencing ANP synthesis and secretion in CHF; and (2) renal clearance of ANP. The pattern of changes in plasma ANP observed during CPB (an initial decrease, followed by an increase at the end of CPB) was qualitatively similar to that previously reported in nontransplant cardiac surgical patients. Proposed mechanisms for the initial decrease in plasma ANP include decreased atria1 perfusion, decreased heart rate, hemodilution, hypothermia, and atria1 decompression.25-*6 Secretion of ANP by other organs may contribute to the residual ANP detected after virtual exclusion of the heart from the
circulation at point P6.*’ Possible mechanisms responsible for the increase in ANP at the end of CPB (while the atria are still decompressed) include physiologic response to rewarming, tachycardia, elevated catecholamines, atrial “reperfusion,” and atria1 ischemia or trauma during CPB.25.2h,28J9In most other studies,2s,2b,29-33plasma ANP at the end of CPB (or shortly after CPB) either returned to or exceeded baseline plasma ANP. In this study, plasma ANP at the end of CPB remained significantly less than baseline plasma ANP. This observation may relate to (1) high basal rate of ANP secretion in these patients with severe CHF; (2) excision of a portion of the native atria (approximately 25%); and (3) removal of the failing ventricles as an additional source of ANP. The magnitude of perioperative reduction in plasma ANP (75% decrease) in these transplant patients was greater than that previously measured in other types of cardiac surgery. XX~‘) Based on prior studies in animals and humans, this large change in plasma ANP may have significant physiologic effects. Studies using infusions of ANP to produce acute elevations in plasma ANP have demonstrated reversible effects of ANP on renal function, baroreflexes, atrial pressures, systemic vascular resistance, and cardiac function.2.34-37 A more recent study has also demonstrated significant effects from acute reductions in ANP using ANP antibodies.?s Injection of ANP antibodies in rats with CHF caused significant increases in right atrial pressure, left ventricular end-diastolic pressure, and systemic vascular resistance. Whether this reduction in ANP following heart transplantation might also play a role in the rapid reversal of baroreflex dysfunction following transplantation’” has not been investigated. In summary, these results suggest that factors other than hemodynamics may be important determinants of plasma ANP concentration in the first 24 hours after heart transplantation. Regression analysis demonstrated that creatinine clearance and tissue concentration of ANP in the native atria were both negatively correlated with early postoperative plasma ANP. The reduced content of ANP in the native atria, as well as the negative correlation between native atrial ANP content and early postoperative plasma ANP, are both consistent with prior animal studies suggesting that severe CHF leads to cellular adaptations limiting ANP storage and promoting ANP secretion. ACKNOWLEDGMENT The authors wish to thank Ann Johnson for her excellent secretarial assistance in preparation of the manuscript, and Nancy Johnson for her vital help in collecting patient data.
REFERENCES 1. Needleman P, Greenwald JE: Atriopeptin: A cardiac hormone intimately involved in fluid, electrolyte, and blood pressure hemostasis. N Engl J Med 314:828-834, 1986 2. Cody RJ, Atlas SA, Laragh JH, et al: Atria1 natriuretic factor in normal subjects and heart failure patients. Plasma levels and renal, hormonal, and hemodynamic responses to peptide infusion. J Clin Invest 78:1362-1374,1986 3. Bates ER, Shenker
Y, Grekin
RJ: The relationship
between
plasma levels of immunoreactive atrial natriuretic hormone and hemodynamic function in man. Circulation 73:1155-1161, 1986 4. Burnett JC, Kao PC, Hu DC, et al: Atrial natriuretic peptide elevation in congestive heart failure in the human. Science 231:11451147,1986 5. Hare JM, Baughman KL, Kass DA, et al: Influence of dilated cardiomyopathy, myocarditis and cardiac transplantation on the relation between plasma atria1 natriuretic factor and atrial pressures. Am J Cardiol67:391-397, 1991
PLASMA
691
ANP AND HEART TRANSPLANTATION
6. Haass M, Deitz R, Fischer TA, et al: Role of right and left atrial dimensions for release of atrial natriuretic peptide in left-sided valvular heart disease and idiopathic dilated cardiomyopathy. Am J Cardiol62:764-770,1988 7. Raine AEG, Phil D, Erne P, et al: Atrial natriuretic peptide and atrial pressure in patients with congestive heart failure. N Engl J Med 315:533-537,1986 8. Arendt RM, Gerbes AL, Ritter D, et al: Atrial natriuretic factor in plasma of patients with arterial hypertension, heart failure or cirrhosis of the liver. J Hyperten 4:S131-S1351986 9. Cernacek P, Maher E, Crawhall JC, Levy M: Molecular forms of atrial natriuretic peptides in dog atrium and plasma. 42:25332539,1988 10. Chimosky JE, Spielman WS, Brandt MA, Heidemann SR: Cardiac atria of BIO 14.6 hamsters are deficient in natriuretic factor. Science 223:820-822,1984 11. Franch H, Callahan LT, Blaine EH: Plasma and atrial content of atrial natriuretic factor in cardiomyopathic hamsters. Life Sci 39:1151-1159, 1986 12. Cantin M, Thibault G, Ding J, Gutkowska J, et al: ANF in experimental congestive heart failure. Am J Pathol 130:552-568, 1988 13. Morii N, Nakao K, Kihara M, et al: Decreased content in left atrium and increased plasma concentration of atria1 natriuretic polypeptide in spontaneously hypertensive rats (SHR) and SHR stroke-prone. Biochem Biophys Res Commun 135:74-81,1986 14. Sonnenberg A, Milojevic S, Chong CK, Veress AT: Atrial natriuretic factor: Reduced cardiac content in spontaneously hypertensive rats. Hypertension 5:672-675,1983 15. Tsundoda K, Hodsman GP, Sumithran E, Johnston CI: Atrial natriuretic peptide in chronic heart failure in the rat: A correlation with ventricular dysfunction. Circ Res 59:256-261,1986 16. Drexler H, Hanze J, Finckh M, Lu W, et al: Atrial natriuretic peptide in a rat model of cardiac failure. Atria1 and ventricular mRNA, atria1 content, plasma levels, and effect of volume loading. Circulation 79:620-633,1989 17. Arendt RM, Gerbes AL, Stangl E, et al: Baseline and stimulated ANF plasma levels: Is an impaired stimulus-response coupling diagnostically meaningful? Klin Wochenschr 65:122-126, 1987 18. Sugawara A, Kakao K, Morii N, et al: Synthesis of atrial natriuretic polypeptide in human failing hearts. Evidence of altered processing of atria1 natriuretic polypeptide precursor and augmented synthesis of B-human ANP. J Clin Invest 81:1962-1970, 1988 19. Naruse M, Hiroe M, Naruse K, et al: Increased levels of B-human atria1 natriuretic peptide-like immunoreactivity in chronically overloaded atrial tissue. Am J Hypertens 3:105-110, 1990 20. Gerbes AL, Vollmar AM: Degradation and clearance of atrial natriuretic factors (ANF). Life Sci 47:1173-1180, 1990 21. Woods RL: Contribution of the kidney to metabolic clearance of atria1 natriuretic peptide. Am J Physiol 255:E934-E941, 1988 22. Chondra CL, Leidy EA, Bunting P, et al: Clearance and early hydrolysis of atrial natriuretic factor in vivo. Structural analysis of cleavage sites and design of an analogue that inhibits hormone cleavage. J Clin Invest 81:1348-1354,1988
23. Almeida FA, Suzuki M, Scharhrough M, et al: Clearance function of type C receptors of atria1 natriuretic factor in rats. Am J Physiol256:R469-R475, 1989 24. Czekalski S, Michel C, Dussaule JC, et al: Atrial natriuretic peptide and adaptation of sodium urinary excretion in patients with chronic renal failure. Clin Sci 75:243-249, 1988 25. Oyama T, Kudo T, Kudo M, et al: Plasma concentration of o-human atria1 natriuretic polypeptide (o-hANP) during cardiopulmonary bypass under fentanyl anaesthesia. Agressologie 27:687689,1986 26. Kharasch ED, Yeo KT, Kenny MA, Amory DW: Influence of hypothermic cardiopulmonary bypass on atrial natriuretic factor levels. Can J Anaesth 36:545-553,1989 27. Teran N, Rodriguez-Iturbe B, Parra G, Gutkowska J: Atrial natriuretic peptide levels in brain venous outflow during cardiopulmonary bypass in humans: Evidence for extracardiac hormonal production. J Cardiothorac Vast Anesth 5:343-347,199l 28. Bilder GE, Siegl PKS, Schofield TL, Friedman PA: Chronotropic stimulation: A primary effector for release of atrial natriuretic factor. Circ Res 64:799-805, 1989 29. Flezzani P, McIntyre RW, Xuan YT, et al: Atrial natriuretic peptide plasma levels during cardiac surgery. J Cardiothor Anesth 2:274-280, 1988 30. Dewar ML, Walsh G, Chiu CJ, et al: Atrial natriuretic factor: Response to cardiac operation. J Thorac Cardiovasc Surg 96:266-270, 1988 31. Lehot JJ, Piriz H, Carry PY, et al: Influence of cardiac surgery on atria1 natriuretic factor. J Cardiothor Anesth 3:S63, 1989 32. Greeley W, Leslie JB, Su M, et al: Plasma peptide release during pediatric cardiovascular surgery. Anesthesiology 65:A414,1986
atrial natriuretic anesthesia and
33. Hedner J, Towle A, Saltzman L, et al: Atria1 natriuretic factor (ANF) in plasma of patients undergoing cardiopulmonary bypass surgery. Anesthesiology 65:A509, 1986 34. Crozier IG, Ikram H, Gomez HJ, et al: Haemodynamic effects of atrial peptide infusion in heart failure. Lancet 1242-1245, 1986 35. Saito Y, Nakao K, Nishimura K, et al: Clinical application of atrial natriuretic polypeptide in patients with congestive heart failure: Beneficial effects on left ventricular function. Circulation 76:115-124,1987 36. Floras JS, natriuretic factor 1990
Phil D: Sympathoinhibitory effects of atria1 in normal humans. Circulation 81:1860-1873,
37. Kuchel 0, Debinski W, Racz K, et al: An emerging relationship between peripheral sympathetic nervous activity and atria1 natriuretic factor. Life Sci 40:1545-1551,1987 38. Drexler H, Hirth C, Stasch HP, et al: Vasodilatory action of endogenous atria1 natriuretic factor in a rat model of chronic heart failure as determined by monoclonal ANF antibody. Circ Res 66:1371-1380,199O 39. Ellenbogen KA, Mohanty Arterial baroreflex abnormalities orthotopic cardiac transplantation.
PK, Szentpetery S, Thames MD: in heart failure. Reversal after Circulation 79:51-58, 1989