| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
J Thorac Cardiovasc Surg 1995;110:111-118
© 1995 Mosby, Inc.
CARDIOPULMONARY BYPASS, |
Boston, Mass.
Supported by National Institutes of Health Trauma Center Grant No. 5P50-36428.
Received for publication May 30, 1994. Accepted for publication Dec. 8, 1994. Address for reprints: Douglas W. Wilmore, MD, Brigham and Women's Hospital, Surgical Metabolism and Nutrition, Thorn Research Building, 14th Floor, 20 Shattuck St., Boston, MA 02115.
Abstract
Excessive fluid accumulation is associated with increased morbidity and prolonged convalescence after cardiopulmonary bypass. However, water fluctuations are difficult to assess solely on the basis of changes in body weight and fluid balance. Bioelectric impedance analysis is a simple, rapid, noninvasive bedside technique that measures the resistance of the body to a weak alternating current (50 kHz). The change in resistance is inversely proportional to the change in total body water. To investigate the relationship between body weight, fluid balance, and resistance, 18 patients who had cardiopulmonary bypass (9 men, 9 women, aged 61 ± 3 years, weighing 80 ± 4 kg, with ejection fraction 54% ± 3% and bypass time 113 ± 8 minutes [mean plus or minus standard error of the mean]) were followed up for 7 postoperative days. Body weight, fluid balance, and whole body and regional resistance were determined at 24-hour intervals. In the immediate postoperative period, fluid retention was accurately detected by simultaneous measurements of weight gain and decreased resistance (p < 0.001). Both measurements detected the initiation of diuresis by postoperative day 2 (p < 0.01). Whole body resistance returned to baseline values by day 7 (p > 0.05), and body weight returned to baseline on day 4 (p > 0.05). Change in weight and change in whole body resistance were highly correlated with cumulative fluid balance (r = 0.84, p < 0.001, andr = -0.81, p < 0.001, respectively), and these two measures were also related to each other throughout the study (r = -0.89, p < 0.001). The initial change in resistance was the best measurement associated with postoperative outcome (p 0.01). The data suggest that the measurement of electric resistance across the body can accurately detect acute changes in total body water and in fluid redistribution through the body. However, determining the relative day-to-day change in whole body resistance seems more appropriate than calculating absolute fluid changes over time. Bioelectric impedance offers a simple, rapid, noninvasive method to monitor serial changes in total body water. This technique can be useful in situations in which rapid alterations in water compartments occur, and it may be useful in predicting outcome after cardiopulmonary bypass. (J THORACCARDIOVASCSURG1995;110:111-8)
Water metabolism is markedly disturbed in patients undergoing heart operations who require cardiopulmonary bypass (CPB).
1 During the course of the operation impaired tissue perfusion, regional hypoxia, complement activation, and the release of vasoactive substances (such as lysosomal enzymes, endotoxins, and kinins) increase capillary permeability. 2-4 The leakage of plasma proteins from the vascular compartment and the loss of proteins on the CPB circuit further decrease plasma oncotic pressure and contribute to extravascular water accumulation. 5-7 In addition to preexisting disease, factors that alter normal water distribution include the enhanced release of hormones that augment salt and water retention, the fluid load associated with priming of CPB, and the type of anesthetic agent used. 8,9
Expansion of total body water (TBW) and extracellular fluid compartments has been documented by dilutional techniques after CPB, and these compartments are known to remain expanded until the eighth postoperative day. 10,11 The consequences of this acute fluid retention on a patient's clinical outcome are largely unknown, although a direct relationship between body weight gain and mortality has been described. 12
Acute changes in body weight have traditionally been used to assess alterations in body hydration. In patients undergoing CPB an initial weight gain of 1 to 7 kg 8 is followed by weight loss associated with spontaneous or pharmacologically stimulated diuresis. However, weight loss also reflects loss of body fat and protein-containing tissue associated with hypocaloric intake and the catabolic response to operation. 13 Thus when tissue loss and fluid gain occur simultaneously changes in body weight may no longer solely reflect changes in body water. This disparity is particularly common in patients with a prolonged and complicated postoperative course.
Although accurate monitoring of body hydration is highly desirable, precise measurements are often limited by practical constraints. Daily weight and fluid intake and output measurements may be imprecise, and other techniques such as isotope-dilutional studies and measurement of central pressures are time consuming, invasive, and often not suitable for routine bedside evaluation. Furthermore, central pressures reflect intravascular volume and not tissue edema.
The human body is composed of approximately 55% to 60% water and therefore offers little resistance to an electric current. 14 The ability to oppose the conduction of an alternating current is called impedance, which is measurable by bioelectric impedance analysis (BIA), a safe, noninvasive, rapid, inexpensive bedside technique that can be used to determine the resistance of the body. The measurement of resistance is inversely related to TBW. 15-17 The purpose of this study was to evaluate the potential use of BIA in adult cardiac surgical patients and to determine whether this approach could be applied to other groups of critically ill patients.
MATERIAL AND METHODS
Subjects
Eighteen patients undergoing either elective or urgent heart operations that necessitated CPB were studied over a 7-day period (Table I). The patients received a variety of medications, as determined by their individual physicians. The protocol was approved by the Hospital's Committee for the Protection of Human Subjects from Research Risks, and signed informed consent was obtained from all patients before operation.
|
During each measurement subjects remained in a semirecumbent position with the limbs in slight abduction. Body weight (Slingscale 2001-A, Scale-Tronix, Inc., White Plains, N.Y.), fluid balance, and standard clinical chemistry values (hematocrit, white blood cell count, and serum sodium) were also recorded during the hospital stay. The patients were treated with diuretics, vasoactive drugs, and fluid restriction as necessary.
Whole body and regional impedance measurements
Adhesive surface electrodes (RJL Systems, Mt. Clemens, Mich.) were positioned as previously described. 16,18 The impedance readings were obtained with the use of current-transmitting clips of an impedance plethysmograph (800 µA; model 101A, RLJ Systems, Detroit, Mich.), which were connected to the electrodes. The skin was cleaned with alcohol at the sites of the electrode placement. The limbs were slightly abducted and palms were placed flat on the bed. Measurements for "whole body" and body segments were obtained. Those measurements determined from electrodes placed on proximal portions of the forearm and lower leg are referred to as proximal. Electrodes placed on the shoulder and lower leg were used to obtain values referred to as shoulder. Measurements were also taken in the trunk, arm, and unoperated leg. The impedance to the conduction of current was displayed by the plethysmograph. The night before each measurement the plethysmograph was charged and calibrated with a standard resistor.
Definitions and calculations
Two components contribute to the measured electrical impedance, Z (ohms). The larger is resistance (R), which accounts for 98% of the signal, whereas only a small contribution (2%) arises from reactance (Xc). Reactance reflects the part of impedance that is caused by the storage of electric charge (capacitance). Impedance is calculated by the following formula:
TBW was calculated from whole body resistance with the use of group and sex-specific equations described by Kushner and Schoeller 19 :
men: TBW = 0.396 (height2/R) + 0.143 (weight) + 8.399
women: TBW = 0.382 (height2/R) + 0.105 (weight) + 8.315
Daily fluid balance was calculated from intake and output data in the patient's medical records, including intravenous and oral fluids, chest tube drainage, urine volume, and estimates of insensible losses 20
Changes in the impedance variables and body weight were compared with preoperative data taken at baseline. The daily fluid balance was expressed as cumulative values and compared with changes in resistance and body weight.
Statistical analysis
Data were analyzed with statistical software (STATISTICA 3.0b, StatSoft Inc., Tulsa, Okla.) on a personal computer (Macintosh LC III, Apple Computer, Inc., Cupertino, Calif.). One-way analysis of variance of repeated measures was used to determine whether measurements of impedance and body weight changed during the study period. Pairwise post-hoc comparisons of means were made by Fisher's probability least standard deviation test. Linear regression analysis that accounted for between-subject variability 21 was used to describe relationships between changes in impedance, body weight, and cumulative fluid balance. Multiple regression analysis was done by forward stepwise regression. A probability value less than 0.05 was considered statistically significant. Results are expressed as the mean plus or minus the SEM.
RESULTS
Indications for bypass were coronary artery disease (n = 16), mitral valve disease (n = 2), and combined coronary artery disease and aortic valve disease (n = 2) (Table II). The number of coronary arteries bypassed ranged from two to five.
|
All 18 patients survived and were discharged from the hospital after operation and recovery. The change in hematocrit value reflected the blood loss and hemodilution. Within the group neither serum sodium value nor white blood cell count changed significantly over time (Table III).
|
|
|
Table IV). These relationships were similar to each other. Change in body weight was also related to whole body resistance (r = -0.89, p < 0.001). Correction for patient's height did not significantly improve the relationships between resistance and both fluid balance and weight (r = 0.83 and 0.91, respectively).
|
|
The initial change in whole body resistance was the primary factor influencing length of hospital stay. Other factors such as prolonged operative periods and low preoperative albumin concentrations also influenced outcome (Table V). Patients whose resistance values returned earlier to preoperative values tended to have fewer postoperative complications (not significant).
|
In this study we performed serial measurements of the body's resistance to an electric current in a group of patients undergoing CPB. Previous studies have validated the ability of BIA to predict TBW. 17 However, only recently have the relationships between changes in the electric resistance of the body and changes in body fluid balance (and/or weight) been measured in critically ill patients. The present study confirms that BIA is a reliable method of monitoring fluid shifts after CPB. Infusion of large fluid volumes can be detected by measuring body resistance, 22 and this method may be used along with body weight and fluid balance to monitor changes in body water. The late return of whole body resistance to baseline values suggested that the patients may have remained overhydrated at the time of hospital discharge, despite the early return of body weight to preoperative values. The calculation of TBW was derived from equations that were determined from normal populations and thus they may not be the most precise method for determing changes in body water in postbypass patients. Body resistance may be more sensitive to water changes than body weight, which may be related to sequestration of fluid in the thorax or abdomen or changes in intracellular or extracellular solute concentration.
The resistance in the leg seemed to be more closely related to changes in weight (see Fig. 3) and this may be a more clinically applicable method of monitoring water content in critically ill patients.
In addition, correcting resistance for height (that is, height2/whole body resistance) explained 82% of body weight variability and had the lowest estimated standard error compared with single resistance measurements. However, the use of this formula throughout the study did not significantly improve the relationship between weight and whole body resistance, and from a practical standpoint we have favored monitoring percent change in resistance that occurs in the leg. However, one should be aware that resistance in the extremities, particularly in the arms of cardiac surgical patients, may be influenced by gravity, operation, and intravenous infusions. 23 Measurements in the nonoperated leg may be more appropriate because the conduction through this homogeneous cylinder may most accurately represent changes in body hydration.
Others have related postbypass fluid retention and weight to mortality. Analysis of our data in a small group of patients revealed that the greater initial changes in whole body resistance are related to a more prolonged hospital and ICU stay. Other variables associated with this effect included the length of the operation and the initial level of serum albumin. A greater change in resistance was also associated with a longer period of mechanical ventilation.
The results of this study confirm that BIA (50 kHz) is a satisfactory noninvasive method of following water balance in patients undergoing CPB. Continuous monitoring of fluid changes by BIA can be useful in accurately assessing changes in body hydration in critically ill patients. The relative changes measured over time seem more appropriate than the calculation of absolute volume of TBW on a daily basis. We believe that this technique will be most useful in ICU patients with morbid postoperative convalescence who simultaneously have tissue loss and fluid gain. Under such circumstances body weight may not accurately reflect changes only in body water. Although the role of BIA in predicting outcome is largely unknown, it may be a promising technique for monitoring fluid retention and its distribution in the critically ill.
References
This article has been cited by other articles:
|
|
 |
|
  Y. Hamada, K. Kawachi, N. Tsunooka, Y. Nakamura, S. Takano, and H. Imagawa Capillary Leakage in Cardiac Surgery with Cardiopulmonary Bypass Asian Cardiovasc Thorac Ann, September 1, 2004; 12(3): 193 - 197. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. L. Ackland, D. Singh-Ranger, S. Fox, B. McClaskey, J. F. Down, D. Farrar, M. Sivaloganathan, and M. G. Mythen Assessment of preoperative fluid depletion using bioimpedance analysis Br. J. Anaesth., January 1, 2004; 92(1): 134 - 136. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. Shum-Tim, C. I. Tchervenkov, E. Laliberte, A.-M. Jamal, T. Nimeh, C.-Y. Luo, B. Bittira, and A. Philip Timing of steroid treatment is important for cerebral protection during cardiopulmonary bypass and circulatory arrest: minimal protection of pump prime methylprednisolone Eur. J. Cardiothorac. Surg., July 1, 2003; 24(1): 125 - 132. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. A. DiMaria-Ghalili Changes in Nutritional Status and Postoperative Outcomes in Elderly CABG Patients Biol Res Nurs, October 1, 2002; 4(2): 73 - 84. [Abstract] [PDF]
|
|
|
|
 |
|
 N. Ad, J. P. Suyderhoud, Y. D. Kim, M. A. Makary, K. W. DeGroot, H. C. Lue, E. A. Pirovic, W. Z. Duvall, and J. L. Cox Benefits of prophylactic continuous infusion of furosemide after the maze procedure for atrial fibrillation J. Thorac. Cardiovasc. Surg., February 1, 2002; 123(2): 232 - 236. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. Simonardottir, B. Torfason, and J. Magnusson Is compartment pressure related to plasma colloid osmotic pressure, in patients during and after cardiac surgery? Perfusion, March 1, 2001; 16(2): 137 - 145. [Abstract] [PDF]
|
|
|
|
 |
|
 T. Shin'oka, D. Shum-Tim, P. C. Laussen, S. M. Zinkovsky, H. G. W. Lidov, A. d. Plessis, and R. A. Jonas Effects of Oncotic Pressure and Hematocrit on Outcome After Hypothermic Circulatory Arrest Ann. Thorac. Surg., January 1, 1998; 65(1): 155 - 164. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. M. FCPS, A. Samenesco, D. Royston, and A. Rees 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., January 1, 1998; 115(1): 178 - 189. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. O. Jacobs Bioelectrical Impedance Analysis: Implications for Clinical Practice Nutr Clin Pract, October 1, 1997; 12(5): 204 - 210. [Abstract] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |
p < 0.01 versus baseline by analysis of variance and Fisher's probability least standard deviation.
