Effects of hypertonic saline-dextran solution in cardiac valve surgery with cardiopulmonary bypass

CARDIOVASCULAR

Effects of Hypertonic Saline-Dextran Solution in Cardiac Valve Surgery With Cardiopulmonary Bypass Ronaldo Bueno, MD, PhD, Adailton Carvalho Resende, MD, Ricardo Melo, MD, Vicente Avila Neto, MD, and Noedir A. G. Stolf, MD, PhD Department of Cardiovascular Surgery, Beneficeˆncia Portuguesa Hospital, and Department of Cardiovascular Surgery, Instituto do Corac¸a˜o, University of Sa˜o Paulo, Sa˜o Paulo, Brazil

Background. Hypertonic saline-dextran (HSD) solution may be beneficial in patients undergoing coronary artery surgery with cardiopulmonary bypass. Valvular dysfunction is associated with high pulmonary wedge pressure, pulmonary hypertension, and ventricular dysfunction. Fluid overload or transient left ventricular failure may occur with HSD infusion in such patients. This study evaluates the cardiorespiratory effects and tolerance of HSD solution infusion in patients undergoing cardiac valve surgery. Methods. This prospective, randomized, double-blind study compared clinical, laboratory, hemodynamic, and respiratory measurements, and fluid balance in 50 patients over a 48-hour period after cardiopulmonary bypass for cardiac valve surgery. Twenty-five patients received 4 mL/kg of HSD during 20 minutes before cardiopulmonary bypass (HSD group). The control group received the same volume of Ringer’s solution (Ringer group). Results. Hospital mortality was zero. The HSD patients had a near zero fluid balance (6.5 ⴞ 13.5 mL/Kg/48 hours), and the control patients had a positive balance (91.0 ⴞ 33.7 mL/Kg/48 hours). Hemoglobin was similar in both groups, but more blood transfusions were necessary in the Ringer group (1.21 ⴞ 1.28 vs 0.48 ⴞ 0.59 units per

patients). The HSD solution induced a higher cardiac index and left ventricular systolic work index postoperatively, and a lower systemic vascular resistance index until 6, 24, and 48 hours. Right ventricular systolic work index increased and pulmonary vascular resistance index decreased after HSD infusion. A better PaO2/FiO2 relation was observed at 1 and 6 hours postoperatively in the HSD group and was associated with a shorter extubation time (432.0 ⴞ 123.6 vs 520.8 ⴞ 130.2 minutes). Increased oxygen delivery index occurred in the HSD group. The HSD infusion was well tolerated as none of the patients experienced fluid overload or had left ventricular failure develop. No other complication attributable to the use of HSD solution was observed. Conclusions. The HSD solution infusion in patients during cardiac valve surgery with cardiopulmonary bypass was well tolerated. Hemodynamic and respiratory functions improved and fluid balance was near zero during the first 48 hours as compared with a large positive balance in the control group. We conclude that HSD infusion is advantageous for patients undergoing cardiac valve surgery.

T

CPB in patients with cardiac valve disease undergoing heart surgery.

he benefits of hypertonic saline solution (7.5%) in the treatment of hemorrhagic shock were first described by Velasco and colleagues [1] in 1980. Its association with a hyperoncotic macromolecular agent like hypertonic saline-dextran (HSD) solution or hydroxiethyl starch prolongs its hemodynamic effects [2]. Several studies [3–11] describe hypertonic saline solution use in coronary artery surgery with cardiopulmonary bypass (CPB). However, valvular dysfunction is generally associated with high pulmonary capillary wedge pressure, pulmonary hypertension, and ventricular dysfunction. Fluid overload or transient left ventricular failure may occur with HSD infusion in these patients. The aim of the present study is to evaluate the cardiorespiratory effects and tolerability of HSD solution infusion during the first 48 hours after Accepted for publication July 30, 2003. Address reprint requests to Dr Bueno, R. Loureiro da Cruz 121-22, Aclimac¸ a˜ o, CEP-01529-020, Sa˜ o Paulo, Brazil; e-mail: rmbueno@ matrix.com.br.

© 2004 by The Society of Thoracic Surgeons Published by Elsevier Inc

(Ann Thorac Surg 2004;77:604 –11) © 2004 by The Society of Thoracic Surgeons

Patients and Methods This study was appproved by the Ethics Committee of Beneficeˆncia Portuguesa Hospital and Instituto do Corac¸a˜o, University of Sa˜o Paulo (Sa˜o Paulo, Brazil). Informed consent was obtained from each patient. A prospective, randomized, double-blinded study compared the infusion of 7.5% NaCl in 6% dextran 70 (HSD group) with the infusion of Ringer’s solution ([RS] group) in 50 patients undergoing elective surgery for cardiac valve disease, operated on from January 2000 to March 2001. Exclusion criteria were age greater than 70 years old or less than 18 years old, renal failure, lung disease, New York Heart Association functional class IV, and emergency surgery. No patient was excluded postoperatively. The HSD or RS solution was administered in a single 0003-4975/04/$30.00 doi:10.1016/S0003-4975(03)01486-3

Ann Thorac Surg 2004;77:604 –11

Clinical: neurologic alterations (confusion, seizures, coma), vasoactive drug use, thoracic drain losses, reoperation, bleeding, and extubation time. Laboratory: hemoglobin, plasmatic sodium, chloride and potassium, lactate, and arterial and venous blood gases. Hemodynamic: heart rate, mean arterial pressure, central venous pressure, mean pulmonary pressure, pulmonary capillary wedge pressure, cardiac index, systemic and pulmonary vascular resistance index, and right and left ventricular systolic work index. Respiratory: oxygen consumption index, oxygen delivery index, and Pao2/Fio2 relationship. Fluid balance: calculated as the difference between the input (randomized solution, crystalloid solutions, medicine and vasoactive drugs, blood transfusion), and output (chest tube drainage, bleeding, and urine output). Patients were pre-medicated with midazolam (0.1 mg/ kg). Anesthesia was induced with midazolam (0.2 mg/ kg), fentanyl (10 to 20 ␮g/kg), and pancuronium (0.15 mg/kg), and were maintained with continuous infusion of fentanyl (0.1 to 0.3 ␮g/kg/min) and propofol (300 ␮g/kg/min). During CPB, midazolam and pancuronium were used to prevent awareness and aid muscle relaxation. All patients had continuous electrocardiographic monitoring. A radial artery was cannulated for blood pressure recording and arterial blood sampling. After induction of anesthesia, a pulmonary artery catheter (Swan-Ganz Catheter, Irvine, CA) was introduced through the internal jugular or subclavian vein for mixed venous blood samples and hemodynamic measurements. Cardiac output was determined with the thermodilution technique and a cardiac output computer (Dixtal DX 2010 [Manaus, AM, Brazil]). The mean of three measurements was used. Hemodynamic and respiratory derived variables were calculated with a standard formula. Cardiopulmonary bypass was performed with a membrane oxygenator, and priming of the extracorporeal system consisted of crystalloid solution. Ringer’s solution was added to perfusate to maintain flow, and packed red cells were transfused if the hemoglobin concentration dropped below 7 g/dL. Perfusion arterial flow was 2 to 2.4 L/min/m2. Moderate hypothermia was used in all patients, and myocardial protection was performed with Buckberg cardioplegia solution at 20-minute intervals. After successful weaning from CPB, residual blood from the circuit was reinfused to avoid losses. After the operation, patients were transferred to the intensive care unit

605

Table 1. Surgical Procedures of 50 Patients Undergoing Cardiac Valve Operation Hypertonic Saline Dextran Group Surgery Mitral Aortic Mitral-aortic Mitral-tricuspid Total a

Ringer Solution Group

No.

%

No.

16 05 03 01 25

64 20 12 4.0 100

18 05 01 01 25

% 72 20 4.0 4.0 100

pa

0.856

Fisher’s exact test.

for at least 48 hours. The management of patients was performed by an intensive care physician not involved in the study, including removal of mechanical ventilation, prescription of vasoactive drugs, and volume infusion (crystalloids or packed red cells when hemoglobin value was less than 9 g/dL). Statistical analysis was performed with the Statistical Analysis System (SAS Institute Inc, 1989). All variables are expressed as mean values and standard deviations. Differences between the HSD and RS groups were evaluated through mean comparisons (t tests). For percentage comparisons, the ␹2 or Fischer’s exact test were used. For datum with multiple measurements, repeated measures of analysis of variance were used to simultaneously evaluate the effects of time and group. Significance was established as p less than 0.05.

Results Table 1 shows the surgical procedures performed in all patients included in this study. Demographic data and data from preoperative and perioperative times were not different between the 2 groups (Table 2). Hospital mortality was zero. Morbidity and complications occurred in both groups. In the HSD group, atrial fibrillation (cardioversion) and incisional hematoma (drainage) occurred in 1 patient and pulmonary infection occurred in another. In the RS group, atrial fibrillation also occurred in 1 patient who experienced confusion, in another patient who had pneumothorax and pleural fistulas develop, and in a third patient who was reoperated on because of bleeding that resulted from coagulopathy. This patient was excluded for analysis of thoracic losses and need of hemotransfusion. No neurologic alterations were observed, except the patient in the Ringer group who had atrial fibrillation and confusion develop. The total volume of chest tube drainage was 7.09 ⫾ 4.12 mL/kg in the HSD group and 9.27 ⫾ 6.20 mL/kg in the RS group, which was a difference that was not statistically significant (p ⫽ 0.15). Hemoglobin values (Table 3) were similar between groups throughout the entire observation period, except at the end of surgery. Nevertheless, the needed amount of packed red cells for transfusion was greater in the RS group (1.21 ⫾ 1.28 vs 0.48 ⫾ 0.59 U/patients; p ⫽ 0.01). During weaning from

CARDIOVASCULAR

dose of 4 mL/kg through a central venous infusion line for 20 minutes after induction of anesthesia and before the beginning of CPB. Clinical, laboratory, hemodynamic, and pulmonary measurements as well as fluid balance were monitored before the operation, 5 minutes after HSD or RS solution infusion, at the end of the operation, and then again at 1, 6, 12, 24, and 48 hours postoperatively. The following factors were evaluated:

BUENO ET AL HYPERTONIC SOLUTION AND VALVE SURGERY

606

BUENO ET AL HYPERTONIC SOLUTION AND VALVE SURGERY

Ann Thorac Surg 2004;77:604 –11

Table 2. Demographic, Preoperative, and Cardiopulmonary Data for the Two Groups HSD CARDIOVASCULAR

Variable

Category

Sex Age (years)b Preoperative AF Postoperative AF Redo-operation LVEF ⬍ 0.5 Pulmonary hypertension Valve repair Valve replacement CPB time (min)b Cross-clamp time (min)b

Male Female 䡠䡠䡠 Yes No Yes No Yes No Yes No Yes No Yes No Yes No 䡠䡠䡠 䡠䡠䡠

No.

Ringer %

No.

%

7 28 10 40 18 72 15 60 45.4 ⫾ 12.8 40.6 ⫾ 12.8 11 44 09 36 14 56 16 64 01 4.0 02 08 24 96 23 92 08 32 08 32 17 68 17 68 06 24 03 12 19 76 22 88 14 56 11 44 11 44 14 56 10 40 09 36 15 60 16 64 15 60 16 64 10 40 09 36 58.2 ⫾ 24.2 57.2 ⫾ 27.7 44.1 ⫾ 23.2 43.3 ⫾ 25.6

p 0.370a 0.191c 0.564a 1.000a 1.000a 0.463a 0.396a 0.771a

Fig 1. Full bars show the number of patients using one or more vasoactive drugs (noradrenaline, dopamine, dobutamine, or sodium nitroprusside) in each timeframe. Interrupted bars show the number of patients using sodium nitroprusside (alone or associated with other vasoactive drugs) in each timeframe. (A ⫽ after infusion; B ⫽ before infusion; EOS ⫽ end of surgery; h ⫽ hour; HSD ⫽ hypertonic saline-dextran; PO ⫽ postoperative; RS ⫽ Ringer’s solution.)

0.771a 0.948c 0.908c

a b Chi-square test or Fisher’s exact test; Values are expressed as c mean ⫾ standard deviation; Student’s t test.

AF ⫽ Atrial Fibrillation; CPB ⫽ cardiopulmonary bypass; HSD ⫽ Hypertonic Saline Dextran Group; LVEF ⫽ left ventricular ejection fraction.

CPB and even in the intensive care unit, vasoactive drugs were more often necessary and were given at a higher dosage in the RS group as shown in Figure 1. The use of furosemide was similar in both groups. In regard to laboratory measurements, plasma sodium values peaked immediately after HSD infusion at a maximum value of 161 mEq/L, and they remained higher than that of the RS group throughout the study period.

Table 3. Laboratory Parameters in the Two Groups Variable Before infusion After infusion End of surgery 1 h postoperative 6 h postoperative 12 h postoperative 24 h postoperative 48 h postoperative Goup/time interaction Group Time a

Group

Hemoglobin (g/dL)

Sodium (mEq/L)

Chloride (mEq/L)

HSD RS HSD RS HSD RS HSD RS HSD RS HSD RS HSD RS HSD RS 䡠䡠䡠 䡠䡠䡠 䡠䡠䡠

13.4 ⫾ 1.8 13.4 ⫾ 2.0 12.1 ⫾ 1.5 13.0 ⫾ 2.0 10.5 ⫾ 1.5a 11.5 ⫾ 1.8 11.2 ⫾ 1.5 12.0 ⫾ 1.8 11.3 ⫾ 1.5 11.3 ⫾ 2.0 10.8 ⫾ 1.1 10.8 ⫾ 1.8 10.5 ⫾ 1.1 10.7 ⫾ 1.6 10.6 ⫾ 0.7 10.5 ⫾ 1.7 0.1102 0.0640 ⬍ 0.001

138.4 ⫾ 3.1 137.0 ⫾ 2.1 150.2 ⫾ 4.6a 136.9 ⫾ 2.3 141.0 ⫾ 3.3a 132.6 ⫾ 2.6 140.6 ⫾ 3.3a 133.4 ⫾ 1.9 140.8 ⫾ 3.5a 134.3 ⫾ 2.9 139.9 ⫾ 2.8a 133.9 ⫾ 3.2 138.6 ⫾ 3.6a 133.3 ⫾ 3.0 136.1 ⫾ 3.6a 133.5 ⫾ 2.8 ⬍ 0.001 䡠䡠䡠

102.1 ⫾ 3.7 101.2 ⫾ 2.8 116.4 ⫾ 4.6a 102.5 ⫾ 3.0 109.7 ⫾ 4.9a 100.0 ⫾ 4.0 112.9 ⫾ 4.8a 105.4 ⫾ 4.3 114.7 ⫾ 2.9a 107.3 ⫾ 3.8 112.8 ⫾ 3.4a 107.4 ⫾ 4.3 107.2 ⫾ 3.8a 103.9 ⫾ 4.4 105.2 ⫾ 4.3a 103.1 ⫾ 2.9 ⬍ 0.001 䡠䡠䡠 䡠䡠䡠

p ⬍ 0.05 versus RS group.

Values are expressed as mean ⫾ standard deviation. h ⫽ hour;

HSD ⫽ hypertonic saline dextran group;

RS ⫽ Ringer’s solution group.

Lactate (mg/dL) 17.7 ⫾ 6.1 15.4 ⫾ 5.2 17.3 ⫾ 7.3 19.2 ⫾ 8.2 31.7 ⫾ 9.6a 41.6 ⫾ 17.8 31.1 ⫾ 12.1 37.0 ⫾ 14.5 37.3 ⫾ 10.8 45.6 ⫾ 20.0 27.2 ⫾ 10.3a 40.3 ⫾ 23.6 19.0 ⫾ 7.0a 25.7 ⫾ 9.8 13.5 ⫾ 4.6 17.8 ⫾ 10.1 0.0332 䡠䡠䡠 䡠䡠䡠

Potassium (mEq/L) 3.6 ⫾ 0.3 3.9 ⫾ 0.5 3.3 ⫾ 0.5 3.8 ⫾ 0.4 3.9 ⫾ 0.7 3.6 ⫾ 0.6 3.8 ⫾ 0.4 3.6 ⫾ 0.7 3.9 ⫾ 0.5 3.8 ⫾ 0.4 4.2 ⫾ 0.6 4.5 ⫾ 0.5 3.8 ⫾ 0.3a 4.2 ⫾ 0.4 3.7 ⫾ 0.3 3.9 ⫾ 0.2 ⬍ 0.001 䡠䡠䡠 䡠䡠䡠

Ann Thorac Surg 2004;77:604 –11

BUENO ET AL HYPERTONIC SOLUTION AND VALVE SURGERY

607

Variable Before infusion After infusion End of Surgery 1 h postoperative 6 h postoperative 12 h postoperative 24 h postoperative 48 h postoperative Group/time interaction Group Time a

Group

Heart Rate (beats/min)

Mean Arterial Pressure (mm Hg)

HSD RS HSD RS HSD RS HSD RS HSD RS HSD RS HSD RS HSD RS 䡠䡠䡠 䡠䡠䡠 䡠䡠䡠

100.9 ⫾ 27.5 91.4 ⫾ 27.1 102.7 ⫾ 21.8 94.7 ⫾ 27.1 113.3 ⫾ 17.2 120.3 ⫾ 14.5 101.4 ⫾ 24.7 104.0 ⫾ 20.7 95.2 ⫾ 14.0 94.8 ⫾ 21.3 89.4 ⫾ 14.4 88.3 ⫾ 16.1 91.3 ⫾ 12.7 88.7 ⫾ 15.1 92.1 ⫾ 9.7 92.6 ⫾ 16.4 0.3007 0.6409 ⬍ 0.001

79.4 ⫾ 15.0 70.0 ⫾ 11.8 78.8 ⫾ 10.5 75.2 ⫾ 12.8 70.2 ⫾ 8.7 69.0 ⫾ 9.5 77.0 ⫾ 10.7 74.8 ⫾ 14.3 81.6 ⫾ 8.9 81.4 ⫾ 13.2 81.0 ⫾ 9.8 83.4 ⫾ 11.0 82.2 ⫾ 9.5 81.6 ⫾ 11.7 84.2 ⫾ 8.5 80.4 ⫾ 11.1 0.1831 0.0568 ⬍ 0.001

Central Venous Pressure (mm Hg)

Pulmonary Mean Arterial Pressure (mm Hg)

Pulmonary Capillary Wedge Pressure (mm Hg)

5.8 ⫾ 3.2 7.1 ⫾ 4.3 10.5 ⫾ 3.4 9.0 ⫾ 4.6 8.4 ⫾ 3.3 7.4 ⫾ 4.0 7.7 ⫾ 2.8 6.7 ⫾ 2.9 8.1 ⫾ 3.2a 6.3 ⫾ 3.0 8.2 ⫾ 3.3 7.2 ⫾ 3.0 9.3 ⫾ 3.3 8.2 ⫾ 3.5 9.4 ⫾ 3.2 8.0 ⫾ 3.2 0.1007 0.2239 ⬍ 0.001

24.3 ⫾ 12.4 23.5 ⫾ 14.6 37.1 ⫾ 13.1 27.3 ⫾ 13.9 23.2 ⫾ 7.1 21.9 ⫾ 9.5 20.9 ⫾ 5.5 19.9 ⫾ 9.0 22.8 ⫾ 7.3 21.3 ⫾ 8.6 21.4 ⫾ 6.1 23.4 ⫾ 9.6 22.7 ⫾ 5.0 25.6 ⫾ 11.3 23.9 ⫾ 6.7 25.7 ⫾ 11.7 0.0013 䡠䡠䡠 䡠䡠䡠

16.8 ⫾ 9.0 17.8 ⫾ 11.6 27.4 ⫾ 9.8a 21.2 ⫾ 11.6 16.5 ⫾ 5.0 16.4 ⫾ 6.4 14.2 ⫾ 4.5 13.4 ⫾ 4.8 14.4 ⫾ 4.7 14.0 ⫾ 5.6 14.4 ⫾ 4.4 14.8 ⫾ 6.2 14.8 ⫾ 4.0 15.6 ⫾ 8.0 15.2 ⫾ 3.9 15.7 ⫾ 7.5 0.0301 䡠䡠䡠 䡠䡠䡠

p ⬍ 0.05 versus RS group.

Values are expressed as mean ⫾ standard deviation. h ⫽ hour;

HSD ⫽ hypertonic saline dextran group;

RS ⫽ Ringer’s solution.

However, these values returned to less than 145 mEq/L at the end of the operation. Accordingly, serum chloride was higher with HSD infusion. Serum lactate increased in both groups after CPB, but this increase was higher in the RS group, beeing difference statistically significant at the end of surgery and at 12 and 24 hours after the operation. The HSD infusion decreased plasma potassium just after its infusion (Table 3). In regard to hemodynamic measurements, no difference in mean arterial pressure and heart rate could be observed during the investigation period (Table 4). Mean pulmonary pressure, central venous pressure, and pulmonary capillary wedge pressure increased after volume infusion, with a greater increase in the HSD group (Table 4). The systemic vascular resistance index decreased in the HSD group and remained unaltered after RS solution. This effect lasted until 24 hours after surgery (Fig 2). The cardiac index increased in both groups after the infusion of HSD or RS solution, but this increase was significantly greater in the HSD group and lasted for 48 hours, although the increase was not significant at 24 hours (Fig 3). A similar behavior showed the left ventricular systolic work index with a greater increase after HSD infusion, which was an effect that lasted 48 hours, although at 12 and 24 hours it was not significant (Fig 3). The pulmonary vascular resistance index decreased after HSD infusion, a slight increase that occurred in the RS

Fig 2. (A) The evolution of pulmonary vascular resistance index (PVRI) during all study periods in patients undergoing cardiac valve operations with cardiopulmonary bypass who received hypertonic saline-dextran (HSD) solution or Ringer’s solution. (B) The evolution of systemic vascular resistance index (SVRI) in the same patients. *p less than 0.05 versus Ringer’s solution group (RS). Values are expressed as mean ⫾ standard deviation. (A ⫽ after infusion; B ⫽ before infusion; EOS ⫽ end of surgery; h ⫽ hour; HSD ⫽ hypertonic saline dextran solution group; PO ⫽ postoperative.)

CARDIOVASCULAR

Table 4. Hemodynamic Parameters in the Two Groups

608

BUENO ET AL HYPERTONIC SOLUTION AND VALVE SURGERY

Ann Thorac Surg 2004;77:604 –11

CARDIOVASCULAR

more than 48 hours was near zero in the HSD group (6.54 ⫾ 13.57 mL/kg/48 hours), but large and positive in the RS group (91.03 ⫾ 33.70 mL/kg/48 hours). The difference was highly significant (p ⬍ 0.001). Urine output, fluid intake, and fluid balance at each time frame are presented in Table 6.

Comment

Fig 3. (A) Evolution of cardiac index, (B) right ventricular systolic work index (RVSWI), and (C) left ventricular systolic work index (LVSWI) during all study periods in patients undergoing cardiac valve surgery with cardiopulmonary bypass who received hypertonic saline-dextran (HSD) solution or Ringer’s solution (RS). *p less than 0.05 versus RS. Values are expressed as mean ⫾ standard deviation. (A ⫽ after infusion; B ⫽ before infusion; CI ⫽ cardiac index; EOS ⫽ end of surgery; h ⫽ hour; HSD ⫽ hypertonic saline-dextran solution group; PO ⫽ postoperative; RS ⫽ Ringer’s solution.)

group. It remained lower in the HSD group at 12 and 24 hours after surgery (Fig 2). Soon after the infusion, HSD increased the right ventricular systolic work index with statistical significance, but at the end of surgery until the end of the study it remained at normal values (Fig 3). Cardiopulmonary bypass decreased the Pao2/Fio2 relation in both groups, but it remained statistically, significantly higher in the HSD group at 1 and 6 hours after surgery (Table 5). The HSD patients had extubation criteria at an earlier period; this resulted in a significantly shorter period of ventilatory support. In regard to the oxygen consumption index, there was no difference between the groups (with it higher in the HSD only at 1 hour). However, the oxygen delivery index increased significantly in the HSD patients until 48 hours, although at 24 it was not significant (Table 5). The HSD group significantly increased urine output shortly after solution infusion and at the end of the operation (Table 6). However, no significant difference was observed in total urine output in the HSD group (124.00 ⫾ 33.71 mL/kg/48 hours) in comparison with that in the RS group (105.10 ⫾ 37.09 mL/kg/48 hours) (p ⫽ 0.06). Fluid intake was significantly higher in the RS group (200.07 ⫾ 55.82 mL/kg/48 hours vs 135.08 ⫾ 33.17 mL/kg/48 hours; p ⬍ 0.001). The total fluid balance at

Valvular disease is a frequent indication for cardiac surgery. Hypertonic saline solutions have been studied in patients undergoing coronary artery surgery with CPB. Oliveira and colleagues [6] included 7 patients with valvular disease in their cohort. Recently, Sirieix and colleagues [12] demonstrated that hypertonic solution infused in the postoperative period of mitral valve repair increased left ventricular pre-load and left ventricular ejection fraction. They concluded that the use of these solutions may be of interest for use in these patients. The suggested explanations for the beneficial effects of hyperosmolarity are (1) compartmental redistribution with a fluid shift to the vascular bed and consequent plasma expansion [1, 13]; (2) a widespread pre-capillary dilatation and direct vasodilation effect acting on vascular smooth muscle in systemic and pulmonary circulation [14]; (3) a positive inotropic effect on myocardial contractility as well as myocardial stiffness due to direct action [15]; (4) a reduction in blood viscosity through hemodilution and reduced endothelial and red blood cell swelling, improved microcirculation [16]; and (5) hormonal and immunologic effects [17] that are possible mechanisms of the long-lasting impact of hypertonic saline solutions. In our study, a greater number of patients (especially in the RS group) received vasoactive drugs compared with patients in other studies [6, 10, 11]. The probable cause of this is the greater number of patients with at least moderate ventricular dysfunction (left ventricular ejection fraction ⬍ 0.50; 20% of patients), atrial fibrillation (40% of patients), and pulmonary hypertension (PASP ⬎ 50 mm Hg; 50% of patients). Hypertonic saline-dextran infusion improved hemodynamic measurements. Cardiac index and left ventricular systolic work index increased after HSD infusion and remained higher than that in the RS group. Systemic vascular resistance index decreased significantly after HSD infusion, and this effect persisted for 24 hours after surgery. Plasma lactate increased after CPB in both groups, but the increase was less pronounced after HSD infusion, the difference being statistically significant at the end of the operation and at 12 and 24 hours after the operation. Perhaps the lower values of plasma lactate reflect a better tissue perfusion and oxygenation in the HSD group. During CPB, microcirculatory dysfunction is caused by reduced arteriolar pressure and endothelial swelling. Blood contact with foreign surfaces of the extracorporeal circuit and damage to the cells, especially red and white blood cells, activated several mediator cascades that affect microcirculation [18]. Hypertonic

Ann Thorac Surg 2004;77:604 –11

BUENO ET AL HYPERTONIC SOLUTION AND VALVE SURGERY

609

Table 5. Respiratory Parameters in the Two Groups

Before infusion After infusion End of surgery 1 h postoperative 6 h postoperative 12 h postoperative 24 h postoperative 48 h postoperative Group/time interaction Group Time a

Group

Vo2I (mL/min/m2)

Do2I (mL/min/m2)

Pao2/Fio2 (Units)

HSD RS HSD RS HSD RS HSD RS HSD RS HSD RS HSD RS HSD RS 䡠䡠䡠 䡠䡠䡠 䡠䡠䡠

133.9 ⫾ 46.4 175.1 ⫾ 74.3 158.3 ⫾ 54.9 166.3 ⫾ 64.6 150.7 ⫾ 47.7 127.9 ⫾ 53.5 202.1 ⫾ 61.3 147.9 ⫾ 50.9 187.6 ⫾ 83.3 152.4 ⫾ 38.1 159.8 ⫾ 52.9 139.4 ⫾ 38.7 147.4 ⫾ 41.2 159.4 ⫾ 53.5 162.4 ⫾ 51.3 158.0 ⫾ 47.6 ⬍ 0.001 䡠䡠䡠 䡠䡠䡠

584.8 ⫾ 155.3 633.1 ⫾ 133.7 905.9 ⫾ 264.5 660.5 ⫾ 148.2 737.5 ⫾ 229.0 629.6 ⫾ 120.6 669.5 ⫾ 193.3 570.7 ⫾ 148.3 661.5 ⫾ 195.2 518.5 ⫾ 129.7 602.5 ⫾ 102.2a 507.9 ⫾ 102.0 549.2 ⫾ 96.1 503.3 ⫾ 119.7 559.2 ⫾ 112.8 482.6 ⫾ 121.1 ⬍ 0.001 䡠䡠䡠 䡠䡠䡠

395.4 ⫾ 107.3 381.6 ⫾ 117.0 396.6 ⫾ 120.8 374.8 ⫾ 129.4 332.5 ⫾ 134.5 266.0 ⫾ 124.4 352.4 ⫾ 207.1a 242.2 ⫾ 125.3 409.8 ⫾ 165.2a 282.6 ⫾ 106.7 396.9 ⫾ 175.7 390.8 ⫾ 147.9 318.2 ⫾ 107.6 312.2 ⫾ 154.1 324.7 ⫾ 114.2 262.7 ⫾ 106.2 0.1096 0.0351 ⬍ 0.001

Extubation Time (min) CARDIOVASCULAR

Variable

432.0 ⫾ 123.6a 520.8 ⫾ 130.2

p ⬍ 0.05 versus RS group.

Values are expressed as mean ⫾ standard deviation. Do2I ⫽ oxygen delivery index; h ⫽ hour; fraction relation; RS ⫽ Ringer’s solution;

HSD ⫽ hypertonic saline dextran group; Vo2I ⫽ oxygen consumption index.

saline solutions seem to improve microcirculation in patients undergoing CPB [19]. These hemodynamic effects of HSD infusion were

Pao2/Fio2 ⫽ arterial oxygen tension/inspired oxygen

more prolonged in our patients with valvular diseases in comparison with those reported in coronary artery disease surgery patients. No definitive explanation is readily

Table 6. Urine Output, Fluid Intake and Fluid Balance in the Two Groups Variable Before infusion After infusion End of surgery 1 h postoperative 6 h postoperative 12 h postoperative 24 h postoperative 48 h postoperative Group/time interaction Group Time a

Group

Urine Ouput (mL/kg)

Fluid Intake (mL/kg)

Fluid Balance (mL/kg)

HSD RS HSD RS HSD RS HSD RS HSD RS HSD RS HSD RS HSD RS 䡠䡠䡠 䡠䡠䡠 䡠䡠䡠

0.99 ⫾ 1.73 0.94 ⫾ 0.80 0.94 ⫾ 0.38a 0.52 ⫾ 0.58 16.40 ⫾ 5.84a 9.90 ⫾ 5.97 20.04 ⫾ 5.77 17.33 ⫾ 8.59 22.98 ⫾ 7.60 18.10 ⫾ 11.17 14.19 ⫾ 8.14 13.21 ⫾ 8.96 18.15 ⫾ 7.57 17.65 ⫾ 12.05 30.31 ⫾ 11.55 27.44 ⫾ 14.14 0.2599 0.0654 ⬍ 0.001

4.06 ⫾ 1.99 5.43 ⫾ 3.04 4.74 ⫾ 1.05 5.60 ⫾ 1.18 27.74 ⫾ 6.12a 40.12 ⫾ 10.36 7.41 ⫾ 5.84 10.54 ⫾ 5.93 25.45 ⫾ 8.42a 32.24 ⫾ 13.25 19.66 ⫾ 8.95a 31.61 ⫾ 13.23 20.00 ⫾ 7.95a 35.46 ⫾ 16.72 26.01 ⫾ 10.79a 39.06 ⫾ 10.61 ⬍ 0.001 䡠䡠䡠 䡠䡠䡠

3.07 ⫾ 1.82 4.49 ⫾ 2.75 3.80 ⫾ 1.14 5.08 ⫾ 1.18 11.78 ⫾ 7.20a 31.18 ⫾ 12.00 ⫺13.47 ⫾ 6.41a ⫺7.47 ⫾ 10.93 1.80 ⫾ 5.34a 13.15 ⫾ 10.22 4.39 ⫾ 8.41a 18.52 ⫾ 9.97 0.41 ⫾ 7.93a 16.25 ⫾ 14.98 ⫺5.25 ⫾ 8.71a 9.83 ⫾ 11.28 ⬍ 0.001 䡠䡠䡠 䡠䡠䡠

p ⬍ 0.05 versus RS group.

Values are expressed as mean ⫾ standard deviation. h ⫽ hour;

HSD ⫽ hypertonic saline dextran group;

RS ⫽ Ringer’s solution group.

610

BUENO ET AL HYPERTONIC SOLUTION AND VALVE SURGERY

CARDIOVASCULAR

apparent. Perhaps direct effects of HSD solution on myocardium and smooth muscle cells make these patients more sensitive to vasoactive drugs. Another hypothesis could be that HSD solution affects the neurohumoral system. Cardiopulmonary bypass increases plasma values of epinephrine, norepinephrine, renin, angiotensin, vasopressin, aldosterone, and atrial natriuretic peptide [20]. Furthermore, heart failure (all patients were New York Heart Association functional class II or III) presents with high plasma concentrations of norepinephrine, renin-angiotensin-aldosterone system, vasopressin, and atrial natriuretic factor. Cross and colleauges [21] showed that hypertonic solution infused in patients undergoing CPB suppressed the increase in the plasma concentration of adrenocorticotropic hormone, cortisol, angiotensin II, and aldosterone, but atrial natriuretic factor remained unaltered. These effects of hypertonic saline solutions on the neurohumoral system, which reduce plasma levels of vasoconstrictor mediators, may explain the long-lasting effects of HSD in patients with cardiac valve diseases undergoing CPB. Increased cardiac index and right ventricular systolic work index associated with a decreased pulmonary vascular resistance index may suggest that using HSD solution would enhance pulmonary vascular compliance avoiding fluid overload and right ventricular dysfunction in such patients. Although the oxygen consumption index was not significantly altered, the increase in the oxygen delivery index until 12 hours after surgery and a better Pao2/Fio2 relation at 1 and 6 hours after surgery explains the shorter extubation time in the HSD group and suggests better pulmonary gas exchange. Hypertonic saline solution, when administered during hemorrhagic shock, increases renal blood flow and glomerular filtration rate, induces renal vasodilation [22], and reduces plasma aldosterone values increasing urine output [23]. In our study, HSD increased urine output shortly after infusion and at the end of surgery, although no significant difference in total urine output existed between the groups. On the other hand, an important increase in fluid requirement was necessary in the RS group to maintain adequate hemodynamic values. Consequently, as also demonstrated by Oliveira and colleagues [6], a large positive fluid balance was observed in the RS group and a near zero balance was seen in the HSD group 48 hours after surgery. When each time period is analyzed, it can also be seen that the HSD group had a statistically significant less positive fluid balance in all study periods, especially during CPB. Extracorporeal circulation is associated with the release of a wide range of vasoactive mediators, as well as hormones, autacoids, and cytokines that can contribute to the development of endothelial damage, increased vascular permeability, tissue edema, and vasoconstriction. Fluid balance during CPB and in the postoperative period is of considerable importance to organ function, particularly pulmonary function [24]. During CPB, fluid accumulation occurs with a 33% increase in measured extracellular fluid space [25] postoperatively, and with intracellular edema, too

Ann Thorac Surg 2004;77:604 –11

(especially endothelial edema). With HSD infusion, the fluid is shifted from the intracellular space (first from the erythrocytes and endothelial cells and then from other tissue cells) and interstitial space (by osmotic gradient) into intravascular space. The HSD solution seems to allow utilization of intracellular and interstitial water to increase the intravascular compartment, reducing fluid requirements during CPB and in the postoperative period, while improving macro and microcirculation and reducing interstitial edema, and perhaps even reducing the degree of pulmonary and other organ dysfunctions that occur after CPB. Hypernatremia occurred after HSD infusion with a maximum value of 161 mEq/L, but at the end of surgery normal plasma sodium levels were present (although higher than those in the RS group). No neurologic symptoms were observed in the HSD group, except 1 patient who had confusion associated with atrial fibrillation develop, but in the RS group. In conclusion, HSD infusion in patients undergoing CPB for cardiac valve disease treatment improved hemodynamic and respiratory function as well as a near zero fluid balance compared with a large positive balance in the control group. The infusion was well tolerated and no complication or side effect attributable to its use was observed. More studies with larger series of patients are necessary, and long-term results need further evaluation.

References 1. Velasco IT, Pontieri V, Rocha e Silva M, Lopes OU. Hyperosmotic NaCl and severe hemorrhagic shock. Amer J Physiol 1980;239:H664 –73. 2. Smith GJ, Kramer GC, Perron P, Nakayama S, Gunther RA, Holcroft JW. A comparison of several hypertonic solutions for resuscitation of bled sheep. J Surg Res 1985;39:517–28. 3. Cross JS, Gruber DP, Burchard KW, Singh AK, Moran JM, Gann DS. Hypertonic saline fluid therapy following surgery: a prospective study. J Trauma 1989;29:817–25. 4. Boldt J, Kling D, Herold C, Dapper F, Hempelmann G. Volume therapy with hypertonic hydroxyethyl starch solution in cardiac surgery. Anaesthesia 1990;45:928 –34. 5. Boldt J, Zickmann B, Ballesteros M, Herold C, Dapper F, Hempelmann G. Cardiorespiratory response to hypertonic saline solution in cardiac operations. Ann Thorac Surg 1991;51:610 –5. 6. Oliveira SA, Bueno RM, Souza JM, Senra DF, Rocha e Silva M. Effects of hypertonic saline dextran on the postoperative evolution of Jehovah’s witness patients submitted to cardiac surgery with cardiopulmonary bypass. Shock 1995;6:391–4. 7. Mazhar R, Samenesco A, Royston D, Rees A. Cardiopulmonary effects of 7.2% saline solution compared with gelatin infusion in the early postoperative period after coronary artery bypass grafting. J Thorac Cardiovasc Surg 1998;115: 178 –89. 8. Prien T, Thulig B, Wusten R, Shoofs J, Weyand M, Lawin P. Hypertonic hyperoncotic volume replacement (7.5% NaCl/ 10% hydroxyethyl starch 200 000/0.5) in patients with coronary artery stenoses. Zentralbl Chir 1993;118:257–63. 9. Brock H, Rapf B, Necek S, et al. Vergleichende Untersuchungen zur postoperativen Volumentherapie bei Kardiochirurgischen Patienten. Anaesthesist 1995;44:486 –92. 10. Tollofsrud S, Noddeland H. Hypertonic saline and dextran after coronary artery surgery mobilizes fluid excess and improves cardiorespiratory functions. Acta Anaesthesiol Scand 1998;42:154 –61.

11. Jarvela K, Kaukinem S. Hypertonic saline (7.5%) after coronary artery bypass grafting. Eur J Anaesthesiol 2001;18: 100 –7. 12. Sirieix D, Hongnat JM, Delayance S, et al. Comparison of the acute hemodynamic effects of hypertonic or colloid infusions immediately after mitral valve repair. Crit Care Med 1999; 27:2159 –65. 13. Baue AE, Tragus ET, Parkins WM. A comparison of isotonic and hypertonic solutions on blood flow and oxygen consumption in the initial treatment of hemorrhagic shock. J Trauma 1967;7:743–56. 14. Marshall RJ, Shepherd JT. Effect of injections of hypertonic solutions on blood flow through the femoral artery of the dog. Am J Physiol 1959;197:951–4. 15. Kien ND, Kramer GC. Cardiac performance following hypertonic saline. Braz J Med Biol Res 1989;22:245–8. 16. Mazzoni M, Borgstrom P, Intaglietta M, Arfors K. Capillary narrowing in hemorrhagic shock is rectified by hyperosmotic saline-dextran reinfusion. Circ Shock 1990;31:407–18. 17. Rocha e Silva M. Hypertonic saline resuscitation. Medicina (Buenos Aires) 1998;58:393–402. 18. Chenoweth DE, Cooper SW, Hugli TE, Stewart RW, Blackstone EH, Kirklin JW. Complement activation during cardiopulmonary bypass. Evidence for generation of C3a and C5a anaphylatoxins. New Engl J Med 1981;304:497–503.

BUENO ET AL HYPERTONIC SOLUTION AND VALVE SURGERY

611

19. Boldt J, Zickmann B, Herold C, Ballesteros M, Dapper F, Hempelmann G. Influence of hypertonic volume replacement on the microcirculation in cardiac surgery. Brit J Anaesth 1991;67:595–602. 20. Downing SW, Edmunds LH Jr. Release of vasoactive substances during cardiopulmonary bypass. Ann Thorac Surg 1992;54:1236 –43. 21. Cross JS, Gruber DP, Gann DS, Singh AK, Moran JM, Burchard KW. Hypertonic saline attenuates the hormonal responses to injury. Ann Surg 1989;209:684 –91. 22. Fujita T, Matsuda Y, Shibamoto T, Uematsu H, Sawano F, Koyama S. Effect of hypertonic saline infusion on renal vascular resistance in anesthetized dogs. Jpn J Physiol 1991; 41:653–63. 23. Drummer C, Gerzer R, Heer M, et al. Effects of an acute saline infusion on fluid and electrolyte metabolism in humans. Am J Physiol 1992;262(5 Pt 2):F744 –54. 24. Boldt J, Von Bormann B, Kling D, Scheld HH, Hempelmann G. The influence of extracorporeal circulation on extravascular lung water in coronary surgery patients. Thorac Cardiovasc Surg 1986;34:110 –5. 25. Breckenridge IM, Digerness SB, Kirklin JW. Increased extracellular fluid after open intracardiac operation. Surg Gynecol Obstet 1970;131:53–6.

INVITED COMMENTARY Valvular heart disease remains a disease of an aged population of patients with significant elements of decompensated congestive heart failure, pulmonary hypertension, and chronic artrial fibrillation. Despite improvements in preoperative medical care, fluid management for these patients in the postoperative period remains a challenge and an enigma. Fluid overload (extracellular and interstitial) coupled with intravascular hypovolemia represents a typical scenario in this group of patients, resulting in administration of fluids and diuretics at the same time. This study represents an acute method of preoperative medical management in the operating suite, which positively altered the course of postoperative management. The beneficial effects of hyperosmolarity are compartmental fluid redistribution with plasma expansion; precapillary/direct vasodilatory effects on vascular smooth muscle in both the systemic and the pulmonary circulations; a positive inotropic effect due to decreased myocardial stiffness; a reduction in blood viscosity with reduced endothelial and red blood cell swelling and, therefore, improved microcirculation; and multiple hormonal effects. The hypertonic saline-dextran solution (HSD) patients experience a near zero fluid balance, less transfusions, increased cardiac index, and lowered sys-

© 2004 by The Society of Thoracic Surgeons Published by Elsevier Inc

temic vascular and pulmonary vascular resistances; this resulted in shorter extubation times and increased oxygen delivery index. This study’s findings are consistent with the beneficial effects of a hyperosmotic solution. It is remarkable that a one-time infusion of a hyperosmolar fluid results in such dramatic improvements in the overall surgical management of valvular heart disease. Many cardiac patients have some degree of decompensated congestive heart failure and left ventricular dysfunction prior to openheart surgery. Perhaps a larger randomized trial would further validate this study and extend its usage to all patients prior to open-heart surgery. The authors ought to be commended on the design of the study and its usefulness for valvular heart patients. Nicholas C. Cavarocchi, MD Department of Cardiothoracic Surgery Mercy Hospital Wilkes-Barre 166 Hanover St Suite 303 Wilkes-Barre, PA 18702-3545 e-mail: [email protected]

0003-4975/04/$30.00 doi:10.1016/j.athoracsur.2003.10.109

CARDIOVASCULAR

Ann Thorac Surg 2004;77:604 –11