Sodium is the primary extracellular cation. With the introduction of point-of-care electrolyte analyzers, serum sodium concentration can be measured rapidly and inexpensively. The sodium dilution principle is based on a mass balance concept that assumes that sodium ions and water remain constant over time in physiologic fluids, except for additions and losses, such as urination or IV fluid administration.1 The serum sodium concentration at any given time point reflects the amount of sodium relative to water in the ECFV. If water is added to or removed from the ECFV, then this change should be reflected in the serum sodium concentration. On the basis of this premise, the serum concentration of sodium, in conjunction with urine volume, urine sodium concentration, and body weight, has been used to estimate changes in ECFV and ICFV.2 Similar approaches by use of serum osmolality have been described.3 The sodium dilution principle has been compared with other techniques to determine the change in the size of the ECFV in humans and has been considered to be a clinically acceptable approach of determining ECFV changes.4 This has allowed for a rapid and uncomplicated means of estimating the change in fluid compartment volumes.
Sodium-containing fluids that are isotonic to plasma initially expand the intravascular volume.5,6 There is subsequent distribution among the entire ECFV, including the interstitium; isotonic fluids reportedly have minimal effects on the ICFV.5 The distribution of fluids following IV administration between the ECFV and ICFV has been described in a mathematic model as being dependent on its tonicity relative to the rest of the ECFV.2 Saline (0.9% NaCl) solution is relatively hypertonic, compared with plasma in horses.7 Although often referred to as isotonic, the higher sodium concentration (154 mEq/L) of this fluid may therefore affect the fluid balance between the ECFV and ICFV associated with this slight hypertonicity.
Calculation of the exact volume of administered fluid that distributes to each of these fluid spaces is challenging. Although a number of models for quantitative estimation have been described, none have been proven to be applicable in the clinical setting. The sodium dilution principle has been used to evaluate fluid distribution in other species1,2,4 but not in horses. The purpose of the study reported here was to apply the sodium dilution principle to describe changes in ECFV and ICFV during dehydration and rehydation in horses. Electrolyte changes in both serum and urine associated with such fluid shifts were also evaluated.
Materials and Methods
Animals—Eight clinically normal mares with a median weight of 512.4 kg (range, 465.4 to 565.8 kg) were studied. Breeds included Thoroughbreds, Quarter Horses, Arabians, and grade horses. The study was approved by the University Animal Use and Care Administrative Advisory Committee of the University of California, Davis. The study was performed concurrent to another study8 evaluating bioelectrical impedance analysis in 6 of these horses. All horses were held off feed for 12 hours prior to the start of the study and both feed and water during the measurement period to minimize variability in body water associated with water in the gastrointestinal tract.
Measurements—Insensible fluid losses from the respiratory tract or skin were not evaluated in this study. These losses were believed to be minimal, especially during the short duration of fluid administration, and no clinical evidence of sweating or hyperventilation was observed. The intracellular space contains a small amount of sodium, but it was considered unlikely that there would have been substantial exchange between extracellular and intracellular sodium during this short period of fluid administration. Fecal production was noted during the experiment, but fecal sodium and water content were not measured or used in the calculation of ECFV changes. On the basis of a prior study9 in horses, the contribution of this loss is thought to be minimal.
Study design—The experiment consisted of 3 study periods that included dehydration (period 1), fluid administration (period 2), and a monitoring period (period 3). Prior to period 1, a 10-gauge cathetera was placed in the jugular vein and used for blood sample collection and fluid administration. Additionally, a 30-F Foley urinary catheterb was placed in the bladder of all mares and attached to a closed urinary collection system. One dose of furosemide (1 mg/kg, IV) was administered at the start of period 1, and urine output was measured hourly for the next 4 hours. At this point, saline solution was administered at 20 mL/kg/h (total, 40 mL/kg) over 2 hours (period 2). Urine output was monitored hourly during this 2-hour period of IV fluid administration. Monitoring continued for 1 hour following IV fluid administration (period 3). Electrical bioimpedance measurements, blood sample collection, and urine sample collection were performed (description to follow).
Bioimpedance—Bioimpedance measurements were made to determine the baseline ECFV prior to the start of period 1. Measurements were made by use of the previously described head-tail configuration because of the ease of electrode placement.10 Briefly, the hair was shaved in a 4 × 4-cm patch over the right cranial border of the first cervical vertebrae and over the caudal aspect of the region of the right tuber ischii. Skin surfaces were cleaned with alcohol and allowed to dry. Subdermal platinum electrodesc were then placed 2.5 cm apart in a configuration parallel to the ground surface within the shaved areas. An adhesive glued was used to hold the electrodes in place throughout the study.
A bioimpedance analyzere was attached to the subdermal electrodes and used to predict the ECFV according to a previously described model.10 Briefly, both resistance and reactance were obtained at 50 frequencies ranging from 5 to 1,000 kHz. The impedance and phase angle were then computed from these measured values and used to determine the resistance of extracellular and intracellular water. Equations used to estimate ECFVs in horses have been described.10
After placement of the electrodes, 3 baseline bioimpedance measurements were obtained in succession and used to predict the ECFV prior to any alterations in fluid balance. These baseline measurements were taken as part of another study.8
Urine—Urine was collected and the total volume recorded hourly through the study periods. At the same time points (hourly throughout periods 1 to 3), a urine sample was collected and frozen at 80°C. Within 8 weeks, the urine electrolytes (Na, K, and Cl) were determined by use of a commercial chemistry analyzer.f Specific gravity of the urine was also determined at these same time points by use of refractometry. The urine SSID was calculated as follows:
Blood sample collection—Blood samples were collected at baseline (beginning of period 1), after 4 hours of dehydration (end of period 1), at 1 and 2 hours of rehydration (during period 2), and 1 hour following rehydration (end of period 3). Blood samples were collected into evacuated serum tubes, allowed to clot, and centrifuged at 3,000 rpm in a commercial centrifuge for 10 minutes.g The serum was then harvested and frozen at 20°C. Serum electrolytes (Na, K, and Cl) and total carbon dioxide concentrations were determined by use of a commercial chemistry analyzer.f One serum sample during period 1 produced values that were significantly outside 2 SDs of the other 7 horses. This sample was reanalyzed with conflicting results, and the sample was therefore removed from the data set.
Estimation of changes in ECFV and ICFV—Calculations were used to estimate changes in ECFV and ICFV. The total amount of sodium within the ECFV prior to the fluid alteration, Na(t)ECF-0, was calculated as follows:
The change in ECFV, ΔECFV, was calculated as follows:
Statistical analysis—All data points are reported as median (range). Comparisons among fluid volumes, changes in fluid volumes, urine electrolyte concentrations, and serum electrolyte concentrations were made by use of a nonparametric (Friedman) test with a post hoc Dunn analysis. Values of P < 0.05 were considered significant.
Results
The sodium dilution method detected a significant increase in the ECFV from the end of period 1 to the end of period 2 (maximum dehydration to rehydration with fluids). There was a significant (P = 0.02) difference between the administered fluid volume and the calculated change in the ECFV of 2.7 L (range, 0 to 5.6 L) during fluid administration, with the latter being larger. This indicates an expansion of the ECFV beyond the amount of the administered fluid. The difference in these 2 fluid volumes represented a similar but opposite change in the ICFV; this was consistent with the significant (P = 0.02) change in ICFV of 2.7 L (range, 0.8 to −4.6 L) over the total fluid administration period (Table 1).
Median (range) absolute ECFV, change in ECFV, change in ICFV, and change in volume input during furosemide-induced dehydration and IV administration of saline (0.9% NaCI) solution in 8 horses.
| Time point | ECFV (L) | ΔECFV (L) | ΔICFV (L) | ΔVol |
|---|---|---|---|---|
| Baseline | 113.9 (100.8 to 126.9) | NA | NA | NA |
| End of period 1 (dehydration) | 104.1 (89.1 to 117.2) | −11.6 (− 6.0 to −17.5) | −0.6 (− 4.5 to 2.7) | −11.5 (− 9.9 to −15.3) |
| Midpoint of period 2 (total, 20 mL of fluid/kg) | 113.5 (101.5 to 121.2) | 11.9 (2.3 to 18.9)* | −2.6 (− 9.5 to 6.7) | 9.3 (8.8 to 10.4)* |
| End of period 2 (total, 40 mL of fluid/kg) | 123.8 (106.1 to 129.1)* | 9.8 (4.2 to 16.9)* | −0.2 (− 7.5 to 4.5) | 9.7 (6.6 to 10.6)* |
| End of period 3 (study completion) | 127.6 (110.1 to 137.7)* | 2.6 (− 9.9 to 4.0) | −3 3 (− 5.3 to 6.9) | −0.7 (− 3.0 to 0) |
Values significantly (P < 0.05) different from the end of period 1 within a column.
NA = Not applicable. ΔECFV = Change in ECFV from the previous time point. ΔICFV= Change in ICFV from the previous time point. ΔVol = Difference between the amount of fluid administered IV and the amount of fluid lost in the urine.
Following identification of the calculated changes in ECFV during fluid administration, a post hoc analysis was performed to identify factors that may have contributed to the degree of expansion of ECFV with saline solution administration. Specifically, serum sodium concentration measured immediately prior to fluid administration (end of period 1), urine fluid volume loss during dehydration, and initial ECFV (baseline) were evaluated as potential contributors to this expansion. Serum sodium concentration immediately before fluid administration (at maximal dehydration) was significantly associated with the magnitude of the change in ECFV in response to fluids. Horses with a high serum sodium concentration before fluid administration generally had a larger increase in the ECFV with the administration of fluids (Figure 1). Fluid volume loss during dehydration and initial changes in ECFV were not associated with the magnitude of ECFV change with fluid administration.
Relationship (P = 0.01) between serum sodium concentration and magnitude of change in ECFV following IV administration of an isotonic crystalloid (ie, total of 20 mL of fluid/kg provided by the midpoint of period 2) in 8 dehydrated horses.
Citation: American Journal of Veterinary Research 69, 11; 10.2460/ajvr.69.11.1506
Relationship (P = 0.01) between serum sodium concentration and magnitude of change in ECFV following IV administration of an isotonic crystalloid (ie, total of 20 mL of fluid/kg provided by the midpoint of period 2) in 8 dehydrated horses.
Citation: American Journal of Veterinary Research 69, 11; 10.2460/ajvr.69.11.1506
Relationship (P = 0.01) between serum sodium concentration and magnitude of change in ECFV following IV administration of an isotonic crystalloid (ie, total of 20 mL of fluid/kg provided by the midpoint of period 2) in 8 dehydrated horses.
Citation: American Journal of Veterinary Research 69, 11; 10.2460/ajvr.69.11.1506
Changes in the serum sodium concentration and the estimated total amount of sodium within the ECFV were calculated (Table 2). Although serum sodium concentration did not change significantly with either furosemide or fluid administration, total ECF sodium content increased during fluid administration, compared with the baseline ECF sodium concentration and the ECF sodium concentration following dehydration as a result of furosemide administration (end of period 1).
Median (range) serum sodium concentration and total ECF sodium concentration during furosemide-induced dehydration and IV administration of saline solution in 8 horses.
| Time point | Serum Na (mmol/L) | Total ECF Na (mEq) |
|---|---|---|
| Baseline | 137 (131 – 140) | 15,765.3 (13,322.7 – 17,263.8) |
| End of period 1 (dehydration) | 137 (134 – 144) | 14,392.5 (12,361.6 – 16,050.6) |
| Midpoint of period 2 (total, 20 mL of fluid/kg) | 136 (134 – 145) | 15,839.1 (13,860.2 – 16,246.3) |
| End of period 2 (total, 40 mL of fluid/kg) | 136 (133 – 141) | 17,077.1 (14,958.6 – 17,560.6)*† |
| End of period 3 (study completion) | 135 (130 – 141) | 16,786.8 (14,609 – 17,519.4)† |
Value significantly (P < 0.05) different from baseline within the column.
†Values significantly different from the end of period 1.
Total ECF Na = Serum sodium concentration multiplied by the ECFV (L) at the given time point.
A larger urine volume was produced during the furosemide-induced dehydration period, compared with production during the other study periods, including fluid administration (Table 3). Urine sodium losses followed a similar pattern (Table 4). Urine sodium concentration was lowest at the end of period 1 and highest following IV fluid administration (end of period 3). Urine chloride concentration followed a pattern that was similar to that of sodium. Urine potassium concentration was highest at baseline and lowest at the end of period 1 and at the end of period 3. The urine SSID was lowest during period 1 and highest during period 3.
Median (range) urine volume and urine sodium losses during furosemide-induced dehydration and IV administration of saline solution in 8 horses.
| Time point | Urine volume Δ from previous time point (L) | Urine Na loss Δ from previous time point (mEq) |
|---|---|---|
| Baseline | NA | NA |
| End of period 1 (dehydration) | 11.5 (9.9 – 15.3) | 1,306.7 (943.7 – 2,032.7) |
| Midpoint of period 2 (total, 20 mL of fluid/kg) | 0.5 (0 – 0.7)* | 11.8 (0 – 60.9)* |
| End of period 2 (total, 40 mL of fluid/kg) | 0.5 (0 – 3.6)* | 83.0 (0 – 709.2) |
| End of period 3 (study completion) | 1.4 (0 – 3.5)* | 137.2 (0 – 609.0) |
Δ = Change. See Table 1 for remainder of key.
Median (range) urine electrolyte concentrations during furosemide-induced dehydration and IV administration of saline solution in 8 horses.
| Time point | Na (mmol/L) | Cl (mmol/L) | K (mmol/L) | SSID (mmol/L) |
|---|---|---|---|---|
| Baseline | 104 (57 to 219) | 68 (39 to 206) | 187.1 (96.2 to 301.2) | 29 (− 102 to 98) |
| End of period 1 (dehydration) | 17 (0 to 80) | 16 (0 to 66) | 83 (51.3 to 150.8) | 4 (− 15 to 14) |
| Midpoint of period 2 (total, 20 mL of fluid/kg) | 39 (0 to 87) | 26 (0 to 50)* | 93.7 (67.2 to 157.6) | 15 (− 12 to 66) |
| End of period 2 (total, 40 mL of fluid/kg) | 176 (107 to 197)*† | 142 (50 to 163) | 41.1 (9 to 142.8) | 38 (12 to 57) |
| End of period 3 (study completion) | 200 (172 to 223)*† | 149 (97 to 176)*† | 19.3 (7.1 to 31.2)†‡ | 40 (31 to 99)* |
Values significantly different from that of the midpoint of period 2.
Value significantly different from that at baseline.
See Table 1 for remainder of key.
Serum chloride concentration increased during IV fluid administration (Table 5). There was a significant decrease in potassium concentration during this same study period. Total carbon dioxide concentration decreased at the end of IV fluid administration and following IV fluid administration, compared with baseline values. The serum SSID was lowest at the end of IV fluid administration.
Median (range) serum electrolyte concentrations during furosemide-induced dehydration and IV administration of saline solution in 8 horses.
| Time point | Na (mmol/L) | Cl (mmol/L) | K (mmol/L) | Total carbon dioxide (mmol/L) | SSID (mmol/L) |
|---|---|---|---|---|---|
| Baseline | 137 (131 – 140) | 95 (89 – 98) | 3.7 (3.3 – 4.1) | 25 (24 – 28) | 42 (39 – 44) |
| End of period 1 (dehydration) | 137 (134 – 144) | 93 (88 – 98) | 3.4 (3.1 – 3.5) | 27 (24 – 30) | 45 (41 – 47) |
| Midpoint of period 2 (total, 20 mL of fluid/kg) | 136 (134 – 145) | 97 (93 – 108) | 3.0 (2.9 – 3.4)* | 25 (22 – 31) | 39 (37 – 41) |
| End of period 2 (total, 40 mL of fluid/kg) | 136 (133 – 141) | 101 (94 – 104)*† | 3.2 (2.8 – 3.5)* | 22 (22 – 26)† | 37 (36 – 39)† |
| End of period 3 (study completion) | 135 (130 – 141) | 98 (94 – 102) | 3.3 (2.9 – 3.7) | 22 (18 – 24)*† | 37 (34 – 40)† |
Values significantly (P < 0.05) differentfrom baseline within a column.
Values significantly different from the end of period 1.
Discussion
Results of this study are similar to those of studies11–13 in other species, which indicated that the effect of administering near-isotonic, sodium-containing fluids IV is primarily limited to the ECFV. However, the results also indicated that the expansion of the ECFV was greater than the volume of fluid administered. On the basis of the study model used to calculate changes in fluid volume, this fluid appears to come from the ICFV. A possible explanation for this shift from the ICFV is osmotic draw produced by the administered crystalloid; saline solution may be hypertonic relative to the ECFV of horses and therefore osmotically draws fluid from the ICFV.
Unfortunately, plasma or ECF osmolarity was not measured in this study, and we cannot definitively know whether saline solution osmolarity was indeed higher than plasma and ECF. However, on the basis of published values14 of plasma osmolarity of clinically normal horses, this appears to be the case. The osmolarity of saline solution is 308 mOsm/L, compared with the reported mean ± SD plasma osmolarity of 288 ± 4.5 mOsm/kg (range, 279 to 296 mOsm/kg) for clinically normal Thoroughbred mares and 286 ± 2.8 mOsm/ kg (range, 281 to 292 mOsm/kg) in clinically normal adult Standardbreds.14 However, the hypothesis of an osmolarity-induced shift is questionable in light of the results of another study5 that had comparable findings after administration of a different isotonic crystalloid solution (an acetated Ringer's solution) with a slightly lower osmolality (270 mOsm/L). Measurement of plasma osmolarity before and after administration of saline solution in horses is therefore warranted to support or refute these hypotheses.
To further investigate this change in the ECFV, post hoc analysis was used to evaluate the role of additional factors, including serum sodium concentration prior to fluid administration, volume of urine produced, and estimated baseline ECFV, that could have influenced the degree of fluid expansion in response to fluid administration. Serum sodium concentration prior to fluid administration (at the end of period 1) appeared to be related to the degree of expansion of ECFV achieved with the fluids administered. To the authors' knowledge, this finding has not been previously described and warrants further study. These results may prove useful as an aid in identifying horses that will respond relatively more or less in terms of ECFV expansion to rapid IV fluid administrations over a short period (ie, IV bolus). The volume of fluid loss during dehydration (period 1) was not significantly related to the volume of ECF expansion with fluid administration. It is possible that this lack of association was the result of the small numbers of horses in the study; it is possible that larger numbers of horses may have identified a relationship between these 2 variables.
The amount of urine produced during bolus administration of fluids following dehydration was small. Accordingly, the amount of sodium lost during this administration was also small. A practical application for the technique described in this study is to measure serum sodium concentration prior to an IV bolus of fluids and then immediately after the fluid administration. These values, combined with the total amount of sodium administered in the fluids, can be used to give a rough estimate of the change in ECFV. This technique would be less accurate than measuring urinary sodium concentration but could potentially be used in an emergency clinical setting rapidly and without a urinary collection system.
The changes in urine and serum electrolyte concentrations following furosemide administration are similar to those observed in other studies.15,16 Sodium excretion increased relative to chloride elimination during and following fluid administration. This resulted in a decreased serum SSID and is most likely responsible for the observed change in total carbon dioxide concentration (Table 5).
In conclusion, the sodium dilution method is a practical means for describing changes to the ECFV and ICFV in horses, as has been shown in other species. It requires further validation in horses. Results of this study suggested that the administration of saline solution initially caused expansion of the ECFV to a greater extent than expected from the volume of fluid administered. This expansion appeared to occur at the expense of the ICFV.
ABBREVIATIONS
| ECF | Extracellular fluid |
| ECFV | Extracellular fluid volume |
| ICFV | Intracellular fluid volume |
| SSID | Simplified strong ion difference |
Angiocath, Becton, Dickinson & Co, Franklin Lakes, NJ.
Silicone-coated Foley catheter, Kendall, Mansfield, Mass.
Grass platinum tetrapolar subdermal 27-gauge needle electrodes, 122 cm, Astro-Med Inc, West Warwick, RI.
Super Flue, Super Glue Corp, Rancho Cucamonga, Calif.
Hydra ECF/ICF bioimpedance analyzer, Model 4200, Xitron Technologies, San Diego, Calif.
717 chemistry analyzer, Boehringer Mannheim Hitachi, Indianapolis, Ind.
Sorvall 7, Dupont Instruments, Newton, Conn.
References
- 1.↑
Hahn RG, Drobin D. Rapid water and slow sodium excretion of acetated Ringer's solution dehydrates cells. Anesth Analg 2003;97:1590–1594.
- 2.↑
Svensén CH, Waldrop KS, Edsberg L, et al. Natriuresis and the extracellular volume expansion by hypertonic saline. J Surg Res 2003;113:6–12.
- 3.↑
Busse MW, Maassen N, Boning D. The calculation of the osmotic volumes of distribution of hypertonic sodium bicarbonate and other hypertonic solutions: a theorectical approach. J Med 1989;20:143–161.
- 4.↑
Zdolsek JH, Lisander B, Hahn RG. Measuring the size of the extracellular flud space using bromide, iohexol, and sodium dilution. Anesth Analg 2005;101:1770–1777.
- 5.↑
Tølløfsrud S, Elgjo GI, Rough DS, et al. The dynamics of vascular volume and fluid shifts of lactated Ringer's solution and hypertonic-saline-dextran solution infused in normolvolemic sheep. Anesth Anal 2001;93:823–831.
- 6.
Brauer LP, Svensén CH, Hahn RG, et al. Influence of rate and volume of infusion on the kinetics of 0.9% saline and 7.5% saline/6.0% dextran in sheep. Anesth Analg 2002;95:1547–1556.
- 7.↑
Freeman DA, Cymbaluk NF, Schott HC, et al. Clinical, biochemical, and hygiene assessment of stabled horses provided continuous or intermittent access to drinking water. Am J Vet Res 1999;60:1445–1450.
- 8.↑
Fielding CL, Magdesian KG, Carlson GP, et al. Estimation of acute fluid shifts using bioelectrical impedance analysis in horses. J Vet Intern Med 2007;21:176–183.
- 9.↑
Sosa León LA, Hodgson DR, Carlson GP, et al. Effects of concentrated electrolytes administered via a paste on fluid, electrolyte, and acid base balance in horses. Am J Vet Res 1998;59:898–903.
- 10.↑
Fielding CL, Magdesian KG, Elliott DA, et al. Use of multifrequency bioelectrical impedance analysis for estimation of total body water and extracellular and intracellular fluid volumes in horses. Am J Vet Res 2004;65:320–326.
- 11.
Rose RJ. Some physiological and biochemical effects of the intravenous administration of five different electrolyte solution in the dog. J Vet Pharmacol Ther 1980;2:279–289.
- 12.
Rose RJ, Carter RJ, Ilkiw JE. Some physiological and biochemical effects of intravenous fluid administration during halothane anesthesia in the dog. J Vet Pharmacol Ther 1980;3:111–119.
- 13.
Hardy JD, Hardy KP, Turner MD. Massive Ringer's lactate infusion: comparison with dextrose 5% and whole blood. Ann Surg 1969;182:644–649.
- 14.↑
Lumsden JH, Rowe R, Mullen K. Haematology and biochemistry reference values for the light horse. Can J Comp Med 1980;44:32–42.
- 15.
Freestone JF, Carlson GP, Harrold DR, et al. Furosemide and sodium bicarbonate-induced alkalosis in the horse and response to oral KCl or NaCl therapy. Am J Vet Res 1989;50:1334–1339.
- 16.
Carlson GP, Jones JH. Effects of frusemide on electrolyte and acid-base balance during exercise. Equine Vet J Suppl 1999;30:370–374.
