Anesth Analg 1999;88:181-187
© 1999 International Anesthesia Research Society
GENERAL ARTICLES
The Effects of Intrinsic Vasopressin on Urinary Aquaporin-2 Excretion and Urine Osmolality During Surgery Under General Anesthesia
Fumio Otsuka, MD*,
Kiyoshi Morita, MD
,
Mamoru Takeuchi, MD,
Takayoshi Yamauchi, MD*,
Toshio Ogura, MD
,
Kyouichi Sekine§,
Masakazu Miura, PhD§,
Masahisa Hirakawa, MD, and
Hirofumi Makino, MD*
Departments of
*Medicine III and
Anesthesiology, Okayama University Medical School;
Health and Medical Center, Okayama University, Okayama; and
§Mitsubishi Kagaku Bio-Clinical Laboratories Inc., Tokyo, Japan
Address correspondence to Fumio Otsuka, MD, Department of Medicine III, Okayama University Medical School, 2-5-1 Shikata-cho, Okayama City, Okayama 700-8558, Japan.
 |
Abstract
|
|---|
A radioimmunoassay has been established to measure urinary aquaporin-2 excretion (u-AQP2). To elucidate how u-AQP2 changes when endogenous vasopressin is increased independently of plasma osmolality, we estimated u-AQP2 during general anesthesia for surgery. We collected urine and blood samples from 50 patients before and 90 and 180 min after anesthetic induction. Plasma (29.1 ± 12.6 pg/mL) and urinary (565.1 ± 207.0 ng/gCr) vasopressin levels were markedly increased after anesthetic induction. Although no significant alteration of plasma osmolality or serum sodium concentration was observed during 180 min, u-AQP2 was significantly increased (preinduction 224.5 ± 24.2 fmol/mgCr; 90 min 243.3 ± 31.8; 180 min 331.4 ± 45.9), paralleling an increase of plasma and urinary vasopressin. The plasma vasopressin concentration after anesthetic induction was far in excess of that expected based on plasma osmolality. Individual plasma and urinary vasopressin concentrations correlated significantly with u-AQP2. At 180 min after anesthesia, plasma osmolality did not change, but urine osmolality decreased despite increased u-AQP2, and a preanesthetic positive correlation between urine osmolality and u-AQP2 disappeared. Thus, although u-AQP2 correlates with increased intrinsic vasopressin levels, the increase in u-AQP2 did not directly contribute to urine concentration. Apparently, an escape from the physiologic effects of high vasopressin level occurs during anesthesia via a mechanism independent of aquaporin-2. We conclude that the anesthetic would interfere with the urinary concentrating capacity at the level of AQP2-action.
Implications: The excessive increase of intrinsic vasopressin exactly augmented urinary aquaporin-2 excretion, resulting in urine concentration; however, anesthesia seemed to modify this process possibly by interfering with the aquaporin-2 action.
|
Introduction
|
|---|
Aquaporin-2 (AQP2) is an arginine vasopressin (AVP)-regulated water channel discovered in rat kidney collecting tubules in 1993 by Fushimi et al. (1), which has subsequently been cloned from the human genome (2). AQP2 is specifically localized to the apical region of collecting duct cells and determines water permeability in the renal collecting duct via shuttle trafficking (3–5). Kanno et al. (6) found that, in human subjects, the AQP2 protein was detectable in the urine, with its urinary excretion increasing after a period of dehydration and decreasing after hydration; this urinary excretion was responsive to exogenous desmopressin (6). Other researchers also confirmed the presence of urinary AQP2 (7,8). Rai et al. (9) further examined the characteristics of urinary AQP2 and refined the radioimmunoassay (RIA) for this protein, using relatively high concentrations of detergent and bovine serum albumin in the RIA buffer. Although urinary excretion of AQP2 responds to exogenous AVP (6) and shows a positive correlation with urine osmolality (9), a well designed protocol for clinical investigation has not been established. Additionally, how urinary AQP2 excretion responds to changes of endogenous AVP has not been clarified. To estimate the influence of intrinsic AVP on urinary AQP2 excretion, we studied it under a condition in which endogenous AVP secretion is markedly increased independent of plasma osmolality changes.
A number of factors stimulate secretion of AVP in addition to osmotic regulation. Decreased circulating blood volume, reduction in mean arterial or left atrial pressure, catecholamines, angiotensin II, atrial natriuretic peptide, prostaglandins, drugs with cholinergic or ß-adrenergic properties, opiates such as morphine, and anesthesia all stimulate AVP secretion (10). Although anesthesia often causes antidiuresis, plasma AVP levels are increased in patients undergoing anesthesia for surgery only after initiation of the surgical procedure (11). Operations under general anesthesia represent an endocrinologically specific condition that is characterized by inappropriate secretion of intrinsic AVP attributable to both surgical stress (12–17) and anesthetics (18–20). Moreover, this phenomenon during anesthesia does not seem to involve significant blood osmolar change (15). We therefore chose this clinical situation to elucidate the relationship between endogenous AVP levels, urinary AQP2 excretion, and changes in urine osmolality.
|
Methods
|
|---|
We studied 50 patients undergoing various operations under general anesthesia at our hospital. Patients who required extracorporeal circulation or disturbed the blood perfusion of aorta, vena cava, and kidneys for their surgeries were excluded in this study. All patients were premedicated IM with a minor tranquilizer approximately 30 min before the scheduled operation. Before anesthesia, an IV line was established in an upper extremity. Appropriate fluid replacement was performed during the operation, and a urinary catheter and a radial artery catheter also were inserted. All patients were anesthetized with inhaled or IV anesthetics and mechanically ventilated during surgery with use of the usual monitors. The surgical procedure was begun approximately 20–30 min after induction of anesthesia in all patients. Blood and urine samples were collected before induction and 90 min and 180 min after initiation of anesthesia. Collected blood samples were centrifuged immediately. Cells were separated from plasma or serum, which were stored at -30°C until assay. Urine samples also were stored at -30°C until assay. Osmolality was measured by the method of freezing point depression (21). Plasma and urinary AVP concentration was assayed by using RIA, and urinary AVP concentrations were corrected for creatinine to the excretion rate of AVP. Informed consent was obtained from all participants before the initiation of the study.
RIA for urinary AQP2 was performed according to the method previously described by Rai et al. (9). Briefly, a synthetic peptide (human AQP2 [V257-A271] with Tyr at the N-terminus) radioiodinated with [125I] Na (New England Nuclear, Boston, MA) by a chloramine-T method was mixed with carrier-free [125I] Na and chloramine-T. The reaction was terminated by the addition of ascorbic acid, followed by 10% KI in distilled water. The mixture was applied to a Sephadex G10 column and eluted with phosphate-buffered saline. For the assay, 0.1 mL of urine sample (diluted one to eight times) or standard, 0.1 mL of assay buffer, and 0.1 mL of anti-AQP2 antibody (a polyclonal antibody raised in rabbits against the synthetic portion of the C-terminal end of human AQP2; final dilution, x16000) were incubated at 4°C for 48 h, followed by the addition of 0.1 mL of [125I] peptide (the 15 C-terminal amino acids) (10,000 cpm) and further incubation at 4°C for 48 h. Bound ligands were separated from free ligands by a double-antibody method. The intra- and interassay coefficients of variation were 4.7%–12.9% and the lower detection limit was 60 fmol/mL (9). The values obtained were corrected for creatinine to estimate the excretion rate of AQP2.
Values in this study are presented as mean ± SEM. Differences were tested statistically using analysis of variance, and the correlation of each variable was assessed by using simple regression analysis (StatView version 4.5; Abacus Concepts, Berkeley, CA). P values <0.05 were accepted as statistically significant.
|
Results
|
|---|
Of the 50 patients, 29 were men and 21 were women; the ages were 51.7 ± 2.4 (range 15–78) yr. Operative regions included four groups: thoracic (n = 12), abdominal (n = 24), orthopedic (n = 8), and head and neck (n = 6). The anesthetic was sevoflurane (n = 27) or isoflurane (n = 10) mixed with oxygen and nitrous oxide and propofol (n = 13). The operative period and anesthetic period were 294.3 ± 23.6 min and 386.1 ± 24.2 min, respectively. None of the 50 patients showed remarkable changes in mean arterial blood pressure (declines exceeding 5%) or cardiac dysfunction during the observation period of 180 min. The volume of blood loss during the total duration of the surgical period was 804.7 ± 101.5 mL; fluid replacement, predominantly consisting of lactated Ringer's solution and acetate solution, saline, and blood transfusions when appropriate, was 4033.9 ± 289.0 mL during the total duration of the anesthetic period. During the study period of 180 min, only two and three patients were given infusions of suspended red cells and plasma component, respectively. Urinary volume and urine excretion rate during the total duration of the anesthetic period were 1024.4 ± 102.3 mL and 2.55 ± 0.14 mL/min, respectively.
The preanesthetic plasma AVP concentration relative to plasma osmolality was within the normal range in 96% of the patients (Figure 1). Ninety minutes after induction of anesthesia, the plasma AVP concentration was markedly increased. By 180 min after induction of anesthesia, this increase was more exaggerated, completely overriding the normal relationship between plasma AVP and osmolality (22).
Plasma AVP concentrations were significantly increased 90 min (11.9 ± 3.6 pg/mL) and 180 min (29.1 ± 12.6 pg/mL) (Figure 2) after induction of anesthesia compared with the preanesthetic value (1.50 ± 0.19 pg/mL). The urinary AVP excretion corrected for urinary creatinine excretion also was increased during anesthesia, especially 180 min after induction (565.1 ± 207.0 ng/gCr), significantly higher than that before anesthesia (35.4 ± 5.5 ng/gCr) and at 90 min (63.3 ± 22.2 ng/gCr). Urinary AQP2 excretion gradually increased during the anesthetic period; at 180 min after induction, it was significantly higher (331.4 ± 45.9 fmol/mgCr) than excretion before anesthesia (224.5 ± 24.2 fmol/mgCr) and at 90 min (243.3 ± 31.8 fmol/mgCr).
View larger version (19K):
[in this window]
[in a new window]
|
Figure 2. Plasma arginine vasopressin (AVP) and urinary aquaporin-2 (AQP2) changes during surgery under general anesthesia. The AVP concentration was significantly increased 90 and 180 min after the induction of anesthesia compared with the preanesthetic period. Urinary AVP also was significantly increased 180 min after anesthetic induction compared with both preanesthetic levels and those at 90 min. AQP2 excretion gradually increased, and the value at 180 min was significantly increased compared with both earlier values. Data are shown as mean ± SEM. #P < 0.05, ##P < 0.01 versus preanesthetic values. *P < 0.05 versus the values after 90 min.
|
|
Plasma osmolality and serum sodium concentration showed no significant changes during anesthesia (Figure 3). In contrast, urine osmolality and urinary sodium excretion were significantly decreased from preanesthetic values 90 and 180 min after the induction of anesthesia.
View larger version (25K):
[in this window]
[in a new window]
|
Figure 3. Change of osmolality and sodium during the operation under general anesthesia. Plasma osmolality and serum sodium concentration did not show a significant change. In contrast, urine osmolality and urinary sodium excretion decreased significantly 90 and 180 min after the induction of anesthesia compared with preanesthetic values. Data are shown as mean ± SEM. #P < 0.05, ##P < 0.01 versus preanesthetic values.
|
|
Preanesthetic correlations among urinary AQP2 excretion, plasma or urinary AVP level, and plasma or urine osmolality are shown in Figure 4. Urinary AQP2 excretion correlated positively with plasma AVP levels (Y = 100.942 + 82.167 X; r = 0.633; P < 0.01) (Figure 4A) and urinary AVP levels (Y = 115.264 + 3.09 X; r = 0.698; P < 0.01) (Figure 4B). Although urinary AQP2 excretion was not correlated with plasma osmolality (R = -0.004; P = 0.9763) (Figure 4C), there was a significant positive correlation between urinary AQP2 excretion and urine osmolality (Y = 60.343 + 0.319 X; r = 0.418; P < 0.01) (Figure 4D). Figure 5 shows correlations among urinary AQP2 excretion, plasma or urinary AVP level, and plasma or urine osmolality 180 min after induction of anesthesia. Urinary AQP2 excretion correlated positively with plasma AVP (Y = 247.212 + 2.891 X; r = 0.793; P < 0.01) (Figure 5A) and urinary AVP (Y = 238.065 + 0.165 X; r = 0.744; P < 0.01) (Figure 5B). Urinary AQP2 excretion 180 min after induction was not correlated with either plasma osmolality (R =-0.062; P = 0.6696) (Figure 5C) or urine osmolality (R = 0.240; P = 0.0929) (Figure 5D).
View larger version (28K):
[in this window]
[in a new window]
|
Figure 4. Correlation among urinary aquaporin-2 (AQP2) excretion, arginine vasopressin (AVP), and osmolality before anesthesia. AQP2 excretion correlated positively with plasma (A) and urinary (B) AVP levels before anesthetic induction. In this period, urinary AQP2 excretion also correlated significantly with urine osmolality (D), but not with plasma osmolality (C).
|
|
View larger version (27K):
[in this window]
[in a new window]
|
Figure 5. Correlation among urinary aquaporin-2 (AQP2) excretion, arginine vasopressin (AVP), and osmolality during anesthesia. AQP2 excretion correlated positively with plasma (A) and urinary (B) AVP levels 180 min after the induction of anesthesia. In this period, urinary AQP2 excretion was not correlated with either plasma (C) or urine osmolality (D).
|
|
|
Discussion
|
|---|
The AQP2 protein is an AVP-regulated water channel (1,2) specifically localized to the apical region of collecting duct cells, which determines water permeability in the renal collecting duct by the following mechanism. After the release of AVP from the posterior pituitary, binding of AVP to V2-receptors in the basolateral membrane of principal cells of the collecting duct and inner medullary collecting duct cells initiates an increase in cyclic adenosine monophosphate (cAMP) via a stimulatory G protein and adenylate cyclase. The increased cAMP levels activate protein kinase A and induce the fusion of AQP2-containing vesicles with the apical membrane, possibly through phosphorylation of phosphoproteins, which renders these cells water-permeable (5). In contrast, withdrawal of AVP triggers the endocytosis of AQP2 and restores the water-impermeable state of the apical side of the cell. This process has been called shuttle trafficking (3,4). Furthermore, because the urinary excretion of AQP2 was responsive to both exogenous AVP and osmolar change (6), the urinary AQP2 has been suggested to be closely related to the trafficking of this protein. However, the clinical significance of the excreted AQP2 has not been completely clarified.
In this study, we investigated how high endogenous AVP concentrations modulate urinary AQP2 excretion and urine osmolality in humans. Under conditions of normal homeostasis in humans, AVP secretion from the posterior pituitary plays a key role in responding to small changes in plasma osmolality to maintain plasma osmolality at a constant level. However, AVP is also released when plasma osmolality is stable if effective circulating volume is decreased or if other nonosmotic AVP stimuli are present (10). Reductions in urine flow and electrolyte excretion, particularly those affecting sodium, have been recognized in patients undergoing anesthesia and surgery (20). Although the precise mechanism responsible for AVP increases during anesthesia and surgery in humans is not known, anesthetics such as halothane (18–20), as well as surgical manipulations (12–17), are believed to contribute to the stimulation of AVP release from the posterior pituitary. Because endogenous AVP levels increase independently of plasma osmotic changes, we studied the operative period in patients under general anesthesia.
In the present study, plasma and urinary AVP levels were significantly increased during surgery under general anesthesia, whereas plasma osmolality and serum sodium level did not change. The mean arterial blood pressure did not change in excess of the 5% decrement that causes a plasma AVP increase (23). With progressive increases of plasma and urinary AVP levels, urinary AQP2 excretion shows a significant increase. In a previous study by Kanno et al. (6), plasma osmolar change caused increased urinary AQP2 excretion. However, the AVP release caused by osmotic regulation is so slight that the change of plasma AVP concentration is reported to be only 1 pg/mL in response to a 1% increase in plasma osmolality (24); therefore, the plasma osmotic change alone cannot increase endogenous AVP release to attain plasma concentrations >20 pg/mL in humans. The plasma AVP concentration in the present study exceeded 20 pg/mL without a significant change in plasma osmolality. Robertson (23) reported that the maximal effect of plasma AVP on urine concentration occurs at 20 pg/mL and that further increases do not affect urine osmolality (23). Our present finding also indicates that the supraphysiologic increase of plasma AVP level during surgery and anesthesia is not likely to affect urine concentration. We established that this excessive AVP release significantly affects the number of water channels (urinary AQP2 excretion).
Preanesthetic values for urinary AQP2 and plasma or urinary AVP concentration showed significant positive correlations that persisted 180 min after anesthetic induction. Urinary AQP2 excretion and urine osmolality showed significant positive correlation in the preanesthetic period, consistent with the reported results of Rai et al. (9). Large variations in preanesthetic urine osmolality (160–968 mOsm/kg) presumably reflect differing basal hydration status among subjects. However, 180 min after induction of anesthesia, the correlation between urinary AQP2 excretion and urine osmolality had become insignificant. Moreover, increased urinary AQP2 excretion was not accompanied by an increase in urine osmolality. These findings suggest that an excessive increase of intrinsic AVP release (plasma AVP concentration > 20 pg/mL) can affect the urinary excretion of AQP2, but not the concentration of the urine. Rai et al. (9) estimated that approximately 3% of the AQP2 present in rat kidney is excreted into the urine each day and stated that this fraction does not change significantly when the rats are dehydrated. The value of urinary AQP2 after anesthetic induction is probably related to AQP2 action with respect to the water permeability of renal collecting ducts. The apparent discrepancy between urinary AQP2 and urine osmolality indicates that urinary concentrating capacity is not determined simply by circulation AVP or the response of AQP2.
The high-AVP state resembles the syndrome of inappropriate secretion of antidiuretic hormone (SIADH). Ecelbarger et al. (25) confirmed an initial 1-day increase in AQP2 mRNA in a SIADH rat model created by the continuous administration of desmopressin with water loading, but they found that the AQP2 protein and mRNA were downregulated after 3 days in the kidney, which indicates that after an initial increase in AQP2 levels, an unknown factor interfered with AQP2 expression. This phenomenon in the SIADH rat model explains how patients with SIADH can excrete nonconcentrated urine in the presence of excessive AVP, escaping its influence. In contrast, downregulation of urinary AQP2 excretion was not observed during our 3-hour observation period. We, however, observed a lack of a urine-concentrating response to such high endogenous AVP. This result indicates the presence of escape phenomenon even in one early phase of exaggerated excretion of AVP.
In conclusion, we demonstrated that urinary AQP2 excretion is highly responsive to excessive intrinsic AVP levels that occur during surgery under general anesthesia. The AQP2 increase, however, was not correlated with urine concentration, which indicates that escape under conditions similar to SIADH may occur via a mechanism other than AQP2. Because the present study involves many variables, such as anesthetics, anesthetic or surgical procedure, volume of fluid replacement, opium and premedication, a precise study with careful consideration of every factor that is influential in the body fluid balance is required to reinforce these results.
|
References
|
|---|
-
Fushimi K, Uchida S, Hara Y, et al. Cloning and expression of apical membrane water channel of rat kidney collecting tubule. Nature 1993;361:549–52.[Medline]
-
Sasaki S, Fushimi K, Saito H, et al. Cloning, characterization, and chromosomal mapping of human aquaporin of collecting duct. J Clin Invest 1994;93:1250–6.
-
Nielsen S, Chou C, Marples D, et al. Vasopressin increases water permeability of kidney collecting duct by inducing translation of aquaporin-CD water channels to plasma membrane. USA 1995;92:1013–7.[Abstract/Free Full Text]
-
Yamamoto T, Sasaki S, Fushimi K, et al. Vasopressin increase AQP-CD water channel in apical membrane of collecting duct cells in Brattleboro rats. Am J Physiol 1995;268:C1546–51.[Abstract/Free Full Text]
-
Deen PM, Knoers NV. Physiology and pathophysiology of the aquaporin-2 water channel. Curr Opin Nephrol Hypertens 1998;7:37–42.[Web of Science][Medline]
-
Kanno K, Sasaki S, Hirata Y, et al. Urinary excretion of aquaporin-2 in patients with diabetes insipidus. N Engl J Med 1995;332:1540–5.[Abstract/Free Full Text]
-
Deen PM, van Aubel RA, van Lieburg AF, van Os CH. Urinary content of aquaporin 1 and 2 in nephrogenic diabetes insipidus. J Am Soc Nephrol 1996;7:836–41.[Abstract]
-
Elliot S, Goldsmith P, Knepper M, et al. Urinary excretion of aquaporin-2 in humans a potential marker of collecting duct responsiveness to vasopressin. Nephrol 1996;7:403–9.
-
Rai T, Sekine K, Kanno K, et al. Urinary excretion of aquaporin-2 water channel protein in human and rat. Nephrol 1997;8:1357–62.
-
Reeves WB, Andreoli TE. The posterior pituitary and water metabolism. In: Wilson JD, Foster DW, eds. Williams textbook of endocrinology. 8th ed. Philadelphia:WB Saunders, 1992:311–56.
-
Forsling ML, Ullmann EA. Non-osmotic stimulation of vasopressin release. In: Moses AM, Share L, eds. Neurohypophysis. Basel:S Karger, 1977:128–35.
-
Amano J, Suzuki A, Sunamori M. Antidiuretic hormone and cardiovascular responses during and after coronary artery bypass surgery. Thorac Cardiovasc Surg 1993;41:297–300.[Web of Science][Medline]
-
Haas M, Glich SM. Radioimmunoassayable plasma vasopressin associated with surgery. Arch Surg 1978;113:597–600.[Abstract/Free Full Text]
-
Ishihara H, Ishida K, Oyama T, et al. Effects of general anaesthesia and surgery on renal function and plasma ADH levels. Anaesth Soc J 1978;25:312–8.
-
Jensen NF, Block RI. Vasopressin levels in major head and neck surgery. Ear Nose Throat J 1997;76:87–94.[Medline]
-
Lehtinen AM, Fyhrquist F, Kivalo I. The effect of fentanyl on arginine vasopressin and cortisol secretion during anesthesia. Anesth Analg 1984;63:25–30.[Abstract/Free Full Text]
-
Stanley TH, Philbin DM, Coggins CH. Fentanyl oxygen anesthesia for coronary artery surgery cardiovascular and antidiuretic hormone responses. Can Anaesth Soc J 1979;26:168–72.[Web of Science][Medline]
-
Luna SP, Taylor PM, Bloomfield M. Endocrine changes in cerebrospinal fluid, pituitary effluent, and peripheral plasma of anesthetized ponies. Am J Vet Res 1997;58:765–70.[Web of Science][Medline]
-
Luna SP, Taylor PM, Wheeler MJ. Cardiorespiratory, endocrine and metabolic changes in ponies undergoing intravenous or inhalation anaesthesia. J Vet Pharmacol Ther 1996;19:251–8.[Web of Science][Medline]
-
Phibin DM, Coggins CH. The effect of anesthesia on antidiuretic hormone. Contemp Anesth Pract 1990;3:29–38.
-
Sweeney TE, Beuchat CA. Limitations of methods of osmometry; measuring the osmolality of biological fluids. Am J Physiol 1993;264:R469–80.[Abstract/Free Full Text]
-
Robertson GL. Vasopressin in osmotic regulation in man. Annu Rev Med 1974;25:315–22.[Web of Science][Medline]
-
Robertson GL. The regulation of vasopressin function in health and disease. Recent Prog Horm Res 1976;33:333–85.
-
Robertson GL, Mahr EA, Athar S, Sinha T. Development and clinical application of a new method for the radioimmunoassay of arginine vasopressin in human. J Clin Invest 1973;52:2340–52.
-
Ecelbarger CA, Nielsen S, Olson BR, et al. Role of renal aquaporins in escape from vasopressin-induced antidiuresis in rat. Invest 1997;99:1852–63.
Accepted for publication October 2, 1998.
Top
Abstract
Introduction
), the AVP concentration was markedly increased. Furthermore, 180 min after induction (
), a further increase that completely obscured relationships between plasma AVP and osmolality was seen.
