Anesth Analg 2003;96:1504-1509
© 2003 International Anesthesia Research Society
GENERAL ARTICLES
Physiologic Effects of Intravenous Fluid Administration in Healthy Volunteers
Kathrine Holte, MD,
Peter Jensen, MS, and
Henrik Kehlet, MD PhD
Department of Surgical Gastroenterology, Hvidovre University Hospital, Denmark
Address correspondence and reprint requests to Kathrine Holte, Department of Surgical Gastroenterology, Hvidovre University Hospital, DK-2650 Hvidovre, Denmark. Address e-mail to kathrine.holte{at}dadlnet.dk
 |
Abstract
|
|---|
Dose regimens in perioperative fluid management are rarely evidence based. Therefore, we investigated responses to an IV fluid infusion in healthy volunteers to assess basic physiologic effects of a fluid infusion per se. In a prospective, double-blinded, cross-randomized study, 12 healthy volunteers with a median age of 63 yr (range, 59–67 yr) received an infusion of lactated Ringer’s solution 40 mL/kg (median, 2820 mL) or 5 mL/kg (median, 353 mL; background infusion) in random order on two separate occasions. The study was designed to mimic the perioperative course with preoperative fasting, infusion of the fluid over 3 h in the morning, and additionally 24-h hospitalization under standardized conditions. Primary outcome assessments were pulmonary function (spirometry), exercise capacity (submaximal treadmill test), balance function (BalanceMaster®), and weight. Infusion of 40 mL/kg of lactated Ringer’s solution compared with the background infusion (5 mL/kg) resulted in a significant decrease in pulmonary function and a significant weight gain of median 0.85 kg (range, -0.2–1.6 kg; P = 0.003) persisting 24 h after the infusion. Exercise capacity and balance function were not influenced by fluid administration. These findings may serve as a basis for clinical studies applying the same type of fluid in different amounts to determine the optimal amount of perioperative fluid in various surgical procedures.
IMPLICATIONS: Infusion of 40 mL/kg of lactated Ringer’s solution in volunteers led to a significant decrease in pulmonary function and a significant weight gain for 24 h but without effects on exercise capacity. These findings may serve as basis information for clinical studies of perioperative fluid management.
|
Introduction
|
|---|
There is a lack of scientific evidence on the optimal dose regimens (i.e., the fluid regimen resulting in optimal perioperative organ function) for perioperative IV fluid administration. Subsequently, there are large variations in perioperative fluid management, with amounts from 1 to 4 L of IV crystalloid used for intraoperative fluid management in minor to moderately intensive surgery (1). Research into perioperative fluid management has primarily focused on the various types of fluid substitutes available for fluid replacement and not on the actual amount of fluid administered. Thus, very limited information is available from randomized clinical studies on the effects of administering various amounts of fluids in either major or minor surgical procedures on postoperative organ functions/morbidity as primary outcome variables (1). Therefore, in an effort to understand the physiologic effects of a fluid infusion per se and to provide a rational background for future clinical studies, studies of fluid infusions in healthy volunteers are required. The effects of a fluid infusion on organ functions have only been investigated in very few studies in young healthy volunteers (1). A reduction in pulmonary function has been demonstrated within 1 h of the infusion of 22 mL/kg or 1–2 L of saline (2,3) but without later assessments. No studies are available on fluid infusion and exercise capacity or balance function. Thus, the prolonged and perioperatively relevant (up to 24 h) effects of a fluid infusion have not been investigated. Recent studies in major surgery have suggested that fluid restriction may improve recovery (4) but also that perioperative dehydration may delay recovery (5,6) after minor surgical procedures.
To investigate the effects of a fluid infusion per se, this study was designed to mimic the typical perioperative course after minor to moderately sized surgery but without surgery being performed, including preoperative fasting, an intraoperatively relevant amount of fluid infused over 3 h in the morning (40 mL/kg of lactated Ringer’s solution), and hospitalization for 24 h. We tested the hypothesis that a fluid infusion may adversely affect organ functions with pulmo-nary function, exercise capacity, 24-h weight gain, and balance function as the primary outcome variables. Because the majority of surgical patients are 60 yr or older, we investigated the 24-h physiologic responses of this fluid administration in subjects between 59 and 67 yr.
|
Methods
|
|---|
We studied 12 healthy volunteers, nine women and three men, median age 63 yr (range, 59–67 yr) and median height 169 cm (range, 158–180 cm). Before inclusion, a medical history was taken, and an objective examination including arterial blood pressure and electrocardiography was performed. All subjects were nonmedicated and were excluded upon any abnormal findings in the medical history (in particular hypertension, previous history of cardiopulmonary or renal disease, or history of voiding problems), clinical examination, or electrocardiography. The Regional Ethics Committee approved the study, and subjects gave written informed consent before inclusion.
The study design was crossover, randomized, and double-blinded and consisted of 2 visits of approximately 30 h each, with a minimum of 2 wk between visits. Subjects were randomized to a 40 mL/kg infusion of lactated Ringer’s solution or a background infusion (5 mL/kg of lactated Ringer’s solution) (composition of the lactated Ringer’s solution: Na+ 130 mmol/L, K+ 4 mmol/L, chloride 109 mmol/L, lactate 28 mmol/L, and calcium 1.4 mmol/L) on the first visit and the opposite fluid infusion at the subsequent visit in random order. Randomization was performed by the closed envelope method. The fluid infusion bags were hidden in black sacks (thus ensuring subject blinding). One of the investigators administered the fluid and thus was not blinded. The other blinded investigator made all the postinfusion assessments at both visits unaware of the administered amount of fluid. An independent ward nurse made the postinfusion weight measurements, not allowing either the subject or the blinded investigator to observe the subjects weight.
The study was designed to mimic the typical perioperative care regimens with emphasis on fluid management (but without surgery). Subjects were fasted from the midnight before the investigation. Additionally, subjects registered food intake 3 days before the first visit and subsequently duplicated this food intake before the second visit to ensure similar nutritional status before each visit. On the morning of the study, breakfast was standardized to one piece of bread, one cup of coffee, and 250 mL of water. The allocated fluid amount was given through a peripheral vein between 9 and 12 AM. Subjects were continuously monitored with electrocardiography during the infusion in the presence of one of the investigators and continued to fast until 2 h after the infusion. After completion of the 2-h assessments, subjects drank 300 mL of water; after the 4-h assessments, they drank 500 mL of water and had a standardized meal. After completion of the 8-h assessments, subjects drank 650 mL of water and had a standardized meal. The food intake was weighed, and calorie intake was standardized to two thirds of the average daily need (obtained from estimates by hospital dietitians as 4666 kJ [weight less than 65 kg] and 6000 kJ [weight more than 65 kg]). In the morning on the second day, breakfast was standardized to 250 mL of water, one cup of coffee or tea, and one piece of bread. Subjects were allowed to walk in the ward, but they were told to be indoors for the duration of the study and were in bed from 11:00 PM to 7:30 AM. Study assessments were made before the infusion and 0, 2, 4, 8, and 24 h after the completion of the infusion (see below).
Spirometry was performed with the subject in the standing position by measuring forced expiratory volume in the first second (L) (FEV1), forced vital capacity (L) (FVC), and peak flow rate (L/min). At each session, the best of three attempts was chosen for further analysis. Spirometry was performed before and 0, 2, 4, 8, and 24 h after completion of the fluid infusion. Oxygen saturation (SpO2) was measured at the same time points with a portable pulse oximeter using a finger probe.
A submaximal treadmill exercise test was performed on a Quinton ClubTrack 612 treadmill (Lifetech, Inc, Wheeling, IL) with adjustable speed and elevation. The speed of the treadmill was adjusted individually according to body weight to achieve an initial workload of 15 W in all subjects. Workload was then increased every 2 min by elevating the treadmill by 2% until a 14% elevation was reached (increase in workload by approximately 15 W each time) and subsequently to 15% (maximal possible increase on the treadmill) and further by a 3-times increase in speed (0.5 km/h each time). The test was continued until subjects reached a heart rate of 120 bpm or speed was increased the maximum of 3 times at maximal elevation. The testing paradigm has previously been validated (7). Heart rate was monitored continuously by a chest belt transmitting heart beat signals to a controller within the treadmill. The subjects fasted 2 h before each test. Blood pressure was measured before and after the exercise test. Exercise capacity was assessed before and 2, 4, 8, and 24 h after completion of the fluid infusion. Exercise data were only available from 10 subjects.
Subjects were weighed in standardized hospital clothing before and 0, 2, 4, 8, and 24 h after completion of the infusion.
Urine output was measured before and 0, 2, 4, 8, and 24 h after completion of the infusion immediately after the subjects had been asked to void.
General well being, thirst, headache, dizziness, and drowsiness were evaluated using a 100-mm visual analog scale (VAS; 0 = no symptom and 100 = worst symptom possible). Fatigue was evaluated on a previously validated 10-point fatigue scale (1 = no fatigue and 10 = worst fatigue imaginable) (8). Subjects were asked to fill out the VAS scales before and 0, 2, 4, 8, and 24 h after completion of the infusion.
Balance function was assessed with a Basic Balance Master® system (NeuroCom International Inc, Clackamas, OR). The system consists of two force plates resting on force transducers recording vertical ground reaction forces and subsequently calculating center of pressure, sway angles, and movement directions. The balance assessments consisted of five tests: three static and two dynamic tests. In the static tests, subjects were asked to stand still with open eyes for 20 s (Test 1), stand still with closed eyes for 20 s (Test 2), and to keep the center of gravity (displayed on the monitor) within a certain area on the screen for 20 s (Test 3). The dynamic test consisted of rhythmic weight shifts from left to right at 3-, 2-, and 1-s pacing (Test 4) and rhythmic weight shifts from back to front at 3-, 2-, and 1-s pacing (Test 5). Measurements were made before and 0, 2, 4, 8, and 24 h after completion of the fluid infusion. Measurements with the Basic Balance Master® system have previously been validated (9,10).
Data were analyzed using nonparametric statistical methods. Wilcoxon’s signed rank test for paired observations was used to describe differences at certain time points between the two visits (between groups) and differences between actual and preinfusion values within one visit (within one group). Repeated measurements (weight, pulmonary function, exercise capacity, and balance function) were analyzed by nonparametric repeated measures analysis of variance (ANOVA) (Friedman’s ANOVA). Where a statistically significant difference was detected with the ANOVA, further paired comparisons were made between the individual time points and preinfusion values using the Wilcoxon’s signed rank test. Urine output at each time point as well as the cumulated urine output were analyzed with Wilcoxon’s signed rank test comparing data between visits. Summary measures (area under the curve [AUC]) were used to analyze data from VAS scales. P < 0.05 was considered significant.
|
Results
|
|---|
There were no differences in preinfusion measurements in any variable between the visits except a slightly higher VAS score for thirst before administration of the small-fluid compared with the large-fluid infusion (P < 0.05). At the first visit, five subjects had 40 mL/kg of lactated Ringer’s solution, and seven subjects had the background infusion administered. There were no differences between calorie and food intake between study sessions. Peroral fluid intake was exactly 1700 mL in all subjects during both visits. The infused amounts of fluid were median 352.5 mL (range, 270–410 mL) with the background infusion (5 mL/kg) and 2820 mL (range, 2160–3280 mL) with the 40-mL/kg infusion.
Compared with preinfusion values, FVC decreased significantly 0–8 h after receiving the 40-mL/kg infusion, whereas no difference in FVC was seen after the infusion of the background volume (Fig. 1). FEV1 was also decreased 0, 2, and 8 h after receiving the 40-mL/kg infusion compared with preinfusion values, whereas there was no difference after the background infusion (Fig. 1). Peak flow decreased 0 and 2 h after the 40-mL/kg infusion (Fig. 1) and was significantly decreased after both fluid administrations 8 and 24 h after the infusion. Compared with preinfusion values, SpO2 increased significantly 0 h after the administration of the 40-mL/kg infusion (from median 98% to median 100%), whereas no significant differences in SpO2 were found at any other time point compared with preinfusion values (Fig. 1).
View larger version (16K):
[in this window]
[in a new window]
|
Figure 1. Effect of 40 mL/kg of lactated Ringer’s solution versus a background infusion (5 mL/kg of lactated Ringer’s solution) on pulmonary function. FVC = forced vital capacity; FEV1 = forced expiratory volume in 1 s. Data were analyzed with Friedman’s analysis of variance (ANOVA) and subsequently with Wilcoxon’s signed rank test, as described. *P < 0.05 compared with preinfusion values.
|
|
Exercise capacity did not differ between subjects when receiving either fluid volume at any time point. The results are summarized in Figure 2.
View larger version (17K):
[in this window]
[in a new window]
|
Figure 2. Effect of 40 mL/kg of lactated Ringer’s solution versus a background infusion (5 mL/kg of lactated Ringer’s solution) on exercise capacity (n = 10). Data were analyzed with Friedman’s analysis of variance (ANOVA) and subsequently with Wilcoxon’s signed rank test, as described. No significant differences were found between visits or between pre- and postinfusion values.
|
|
Twenty-four-hour weight gain had a median of 0.85 kg (range, -0.2–1.6 kg) when receiving the 40-mL/kg infusion and median 0 kg (range, -0.8–2.5 kg) when receiving the background infusion (P < 0.05; Fig. 3).
View larger version (11K):
[in this window]
[in a new window]
|
Figure 3. Effect of 40 mL/kg of lactated Ringer’s solution versus a background infusion (5 mL/kg of lactated Ringer’s solution) on weight. Median values shown. Data were analyzed with Friedman’s analysis of variance (ANOVA) and subsequently with Wilcoxon’s signed rank test, as described. *P < 0.05 compared with preinfusion values.
|
|
Urine output was significantly larger at any time point between 0–24 h comparing 40 mL/kg with the background infusion, and the cumulated urine output was a median of 3413 mL (range, 2125–4080 mL) when receiving the 40-mL/kg infusion compared with a median of 1888 mL (range, 1320–2350 mL) when receiving the background infusion (P < 0.05) (data not shown).
VAS scale data were analyzed as AUCs for 0–8 h and 0–24 h. No differences were found between visits in dizziness, headache, drowsiness, or general well being. A slight reduction in thirst (AUC, 0–8 h; P < 0.05) and fatigue (AUC, 0–24 h; P < 0.05) was found after the 40-mL/kg infusion compared with background fluid administration (data not shown).
No difference in balance function in static mea-surements was found after either fluid administration (P > 0.05; data not shown). Only a slight change in balance function in the dynamic balance tests was seen (an improvement in directional control after the background fluid administration in back to front 3-s pacing [Test 5] at the 24-h test and an improvement in directional control after the background fluid administration in back to front 1-s pacing [Test 5] at 2, 4, and 8 h after the fluid administration; P < 0.05; data not shown). Otherwise, no significant changes in balance assessments were seen.
|
Discussion
|
|---|
In summary, infusion of 40 mL/kg of lactated Ringer’s solution during 3 hours led to a decrease in pulmonary function (FVC, FEV1, and peak flow) lasting eight hours and a weight gain of 0.85 kg after 24 hours compared with a background infusion of 5 mL/kg of lactated Ringer’s solution. In contrast, the infusion of 40 mL/kg of lactated Ringer’s solution did not affect exercise capacity or balance function.
The study was designed to imitate aspects of perioperative fluid management but without surgery for 24 hours, including preinfusion fasting from midnight and semistarvation during the study period. The effects of an infusion of a fluid amount equivalent to daily practice in minor to moderate surgical procedures (1,5), i.e., surgical procedures that often may be performed in an ambulatory setting (median, 2820 mL infused over three hours) was assessed within a double-blinded, crossover design under standardized conditions. This is the first study to examine various physiological effects of a fluid infusion beyond one to two hours in subjects with an age equivalent to the majority of the surgical patient population. We designed the study to ensure that subjects were not dehydrated with an oral fluid intake of 1700 mL during the 24-hour study duration at both visits. Thus, we intended to investigate the effects of fluid administration in excess of normohydration and chose 40 mL/kg of lactated Ringer’s solution because this amount is frequently administered during surgery in minor to moderately sized surgical procedures (1).
Based on findings from studies in healthy volunteers to find a decrease in pulmonary function after an infusion of 1–2 L of saline after one hour (2,3), and reports suggesting that postoperative pulmonary edema (11) and postoperative ileus (4) may be related to the administered amount of fluid, we hypothesized that infusion of fluid in excess of normohydration may have physiologically significant adverse effects, and these adverse effects may be attributable to the accumulation of fluid in the interstitial tissue.
A decrease in pulmonary function is obligatory after surgery and may predispose to development of atelectasis and pneumonia. The pathogenesis for the postoperative impairment in pulmonary function is multifactorial, including inhibitory reflexes and pain (12). We found that pulmonary function decreased significantly for eight hours after the administration of 40 mL/kg of lactated Ringer’s solution. It may be argued that the magnitude of this decrease (approximately 5%–7%) is relatively small compared with the decrease in pulmonary function seen after surgery (30%–50%). However, our results suggest that excess fluid may contribute to the impairment in pulmonary function seen after surgery. Obviously, the clinical relevance of these findings needs to be determined in randomized clinical studies applying large versus small fluid regimens. The significantly improved SpO2 saturation immediately after the infusion of 40 mL/kg of lactated Ringer’s solution is difficult to explain because pulmonary function was significantly decreased at the same time, but peripheral vasodilatation may be of importance because the peripheral perfusion is known to influence measurements by pulse oximetry. Care was taken to ensure that other potentially confounding factors, such as subject movement and ambient light, did not differ between visits.
Despite a larger urine output after the administration of the large-fluid volume, a weight gain of median 0.85 kg was persistent 24 hours after the infusion of 40 mL/kg of lactated Ringer’s solution. Whereas it may not be surprising that weight increases initially after an administration of 3 L of fluid, the inability to excrete the fluid even after 24 hours may be of clinical importance in the perioperative period. Previously, other studies in healthy volunteers have shown that an acute saline infusion of 22 mL/kg takes approximately two days to be excreted in young healthy volunteers (1,13). In a large detailed study with 374 normotensive subjects given 2 L of saline and a standardized oral sodium intake of 150 mEq (total sodium intake, 458 mEq), urinary collections found only between 150–166 mEq of sodium to be excreted within 24 hours after the infusion (14). However, despite a weight gain of 0.85 kg after receiving 40 mL/kg of lactated Ringer’s solution, we found no changes in exercise capacity at any time point after the fluid infusion.
Because dizziness has been determined an independent predictor of prolonged hospital stay after minor surgical procedures (6), objective determination of balance function may be of potential value as an indicator of recovery. However, we found that fluid infusion per se did not affect balance function as determined with either the Basic Balance System® or by assessment of dizziness on a VAS-scale.
In summary, infusion of 40 mL/kg of lactated Ringer’s solution led to a decrease in pulmonary function for eight hours and a significant weight gain for 24 hours. These findings may serve as a basis for randomized clinical studies where large (i.e., fluid administration in excess of normohydration) versus small (i.e., fluid administration aiming for normohydration) are applied. For example, in a relatively minor procedure such as laparoscopic cholecystectomy where fluid administration may vary from 1–4 L (1), randomized clinical studies are required to determine the optimal fluid regimen based on measurements of perioperative organ functions and outcome.
|
Acknowledgments
|
|---|
Supported, in part, by a grant from the University of Copenhagen and the Danish Research Council (no. 22–01–0160)
|
References
|
|---|
- Holte K, Sharrock NE, Kehlet H. Pathophysiology and clinical implications of perioperative fluid excess. Br J Anaesth 2002; 89: 622–32.[Abstract/Free Full Text]
- Hillebrecht A, Schulz H, Meyer M, et al. Pulmonary responses to lower body negative pressure and fluid loading during head-down tilt bedrest. Acta Physiol Scand Suppl 1992; 604: 35–42.
- Collins JV, Cochrane GM, Davis J, et al. Some aspects of pulmonary function after rapid saline infusion in healthy subjects. Clin Sci Mol Med 1973; 45: 407–10.[ISI][Medline]
- Lobo DN, Bostock KA, Neal KR, et al. Effect of salt and water balance on recovery of gastrointestinal function after elective colonic resection: a randomised controlled trial. Lancet 2002; 359: 1812–8.[ISI][Medline]
- Holte K, Kehlet H. Compensatory fluid administration for preoperative dehydration: does it improve outcome? Acta Anaesthesiol Scand 2002; 46: 1089–93.[ISI][Medline]
- Chung F, Mezei G. Factors contributing to a prolonged stay after ambulatory surgery. Anesth Analg 1999; 89: 1352–9.[Abstract/Free Full Text]
- Kjaersgaard M, Lie C, Bisgaard T, et al. Effect of oxygen on postoperative cardiovascular response to exercise. Eur J Surg 2000; 166: 915–9.[ISI][Medline]
- Christensen T, Bendix T, Kehlet H. Fatigue and cardiorespiratory function following abdominal surgery. Br J Surg 1982; 69: 417–9.[ISI][Medline]
- Persson F, Kristensen BB, Lund C, et al. Postural stability after inguinal herniorrhaphy under local infiltration anaesthesia. Eur J Surg 2001; 167: 449–52.[ISI][Medline]
- Song D, Chung F, Wong J, et al. The assessment of postural stability after ambulatory anesthesia: a comparison of desflurane with propofol. Anesth Analg 2002; 94: 60–4.[Abstract/Free Full Text]
- Arieff AI. Fatal postoperative pulmonary edema: pathogenesis and literature review. Chest 1999; 115: 1371–7.[Abstract/Free Full Text]
- Warner DO. Preventing postoperative pulmonary complica-tions: the role of the anesthesiologist. Anesthesiology 2000; 92: 1467–72.[ISI][Medline]
- Drummer C, Heer M, Baisch F, et al. Diuresis and natriuresis following isotonic saline infusion in healthy young volunteers before, during, and after HDT. Acta Physiol Scand Suppl 1992; 604: 101–11.[Medline]
- Luft FC, Grim CE, Fineberg N, et al. Effects of volume expansion and contraction in normotensive whites, blacks, and subjects of different ages. Circulation 1979; 59: 643–50.[Abstract/Free Full Text]
Accepted for publication December 23, 2002.
Top
Abstract
Introduction
