The volume of fluid that leaked into the interstitium did correlate with the infused fluid volume (Pearson correlation analysis: r = 0.42, P = .01, 0.03 < ρ ≤ 0.70; Spearman correlation analysis: r =0.45, P = .003, 0.07 < ρ ≤ 0.72; significant correlation; Figure 3). After the reinfusion of autologous blood, the volume of fluid that leaked into the interstitium decreased (Figure 4). The urine output was 383 ± 356 mL after surgery and 567 ± 424 mL after reinfusion. The infused volume was not strongly correlated with the leakage into the interstitium after blood reinfusion (Pearson correlation analysis: r = 0.25, P = .12, −0.16 < ρ ≤ 0.59; Spearman correlation analysis: r = 0.25, P = .11, −0.16 < ρ ≤ 0.59; not significant correlation; Figure 4).
In this study, we analyzed the kinetics of infused fluid using the dilution rate of a biomarker in the bloodstream.18 This strategy has been described as a VK study and is used commonly to analyze and to simulate the distribution and elimination of infused fluids.19 The methodology has been applied widely to assess the distribution and elimination of fluids in various clinical settings.20–23 The present analysis used this model to quantify fluid movement in healthy patients undergoing jaw surgery. Because ANH was routine in our cases, we estimate the BV from the hemoglobin dilution rate.
Although VK studies are a well-established technique for exploring fluid shifts, we validated our estimate by comparing our results with the BV value estimated using a different formula. The value we estimated was similar to the value obtained in a former isotope study15 (4079 ± 1240 mL in our study versus 4135 ± 626 mL using the standard formula from the isotope study), which supports our method. Jacob et al24 previously estimated BV using the VK after ANH method and also found good agreement between the VK method and indocyanine green dilution and fluorescent red cells (−0.53% ± 7.84%). A shortcoming of the VK approach is that the BV or plasma value is estimated based on the hemoglobin level in the peripheral vessels, which may differ from the whole body hemoglobin level. The ratio between the whole body hematocrit level and the large vessel hematocrit level, called the F cell ratio, is assumed to be 0.8 to 1.0.25 Because our aim was to determine the relative change in intravascular fluid, we did not use the F-cell ratio to correct the value.
The fluid that “leaked” into the interstitium or “third” space after ANH (400 mL of blood withdrawn and 500 mL of HES infusion) was estimated based on the results of the VK study. This value varied substantially (232 ± 159 mL) but was consistent with our study protocol (at a hematocrit level of 40%, the volume of withdrawn plasma is approximately 240 mL, and if the volume of infused fluid [HES] is 500 mL, then 500 − 240 mL = 260 mL).
During the intraoperative period, we found that the infused fluid often moved from the intravascular to interstitial space. Our data showed that the retained intravascular volume was not strongly correlated with the infused volume. In several cases, only a few milligram per kilogram remained in the intravascular space (y = 0 corresponds to the absence of fluid; Figure 1). Our data are consistent with those of Rehm et al,26 who reported that the infused intraoperative crystalloid had little effect on the postoperative BV, and with other studies that found a minimal increase in BV despite a large positive balance of crystalloid in cardiovascular surgery patients.27,28
Previously, the Starling law, which describes an equilibrium between hydrostatic and oncotic forces, was considered responsible for governing the fractional distribution of infused fluid. Research on the glycocalyx as a protective barrier for fluid exchange, however, has suggested that this principle may not be uniformly correct.9 According to the revised Starling theory, outward movement of fluid is common, even in venules where hydrostatic pressure is low.9 In contrast, fluid reabsorption from the interstitium to the intravascular space, driven by the osmotic pressure difference between the plasma and the interstitium, occurs only transiently because the glycocalyx layer nullifies the difference between the extravascular and the intravascular osmotic pressures. Thus, return of interstitial fluid into the intravascular space occurs mostly because of lymph flow and not the osmotic pressure difference. Our clinical results agreed with this concept. Increases in infused fluid volume may have increased intravascular pressure, leading to more outward fluid movement from the intravascular to the interstitial compartment (“leakage into the interstitium”). Because the inward flow is driven by lymph accompanying protein movement, the volume of fluid retained intravascularly from the interstitial compartment back to the intravascular space would not necessarily be correlated with infused fluid volume (as we observed). The large variation in fluid absorption supports the hypothesis that fluid movement is driven not only by physiochemical forces but also by other forces such as the endocrine system.5 In other interpretations, absorption is dependent on intravascular hydrostatic pressure.
The distribution area of the infused crystalloid apparently consists of the extracellular space plus a small proportion of the intravascular space.29,30 Thus, the infused crystalloid may contribute to postoperative edema formation, particularly in the presence of anesthesia and/or surgical stress.31,32 Brandstrup1 has demonstrated that crystalloid accumulation is linked to postoperative complications, and several days are required for intraoperatively administered crystalloid to be completely excreted from the body. We demonstrated that large amounts of intraoperatively infused fluid may be distributed to the extracellular space, thus suggesting a cause of postoperative edema formation. However, the amount of fluid that leaked into the extracellular space decreased over time even after the reinfusion of homologous blood. Thus, our results should be applied to sustained postoperative edema formation with caution.
The concept of the “third space” provides a justification for the excess administration of fluids to support crystalloid-based BV maintenance. If the volume of the third space is determined by surgical stress and is self-limited, fluid administration would eventually expand the intravascular space. However, our observations that the fluid volume that leaked into the interstitium correlated with the infused volume suggest that the volume of the third space is unlimited and might increase with intraoperative infusion.3 In other words, the third space may be a product of fluid therapy, rather than an actual space determined by surgical stress.5,7,33
Current evidence does suggest that limiting perioperative fluid administration can improve outcomes. A 2014 trial administering crystalloid at the minimum maintenance fluid level found better outcomes among patients undergoing extensive urologic surgery.4 Clinical trials of a “zero-balance” fluid policy also have reported improved outcomes.3 Taken together, our data and those of previous trials suggest that minimizing crystalloid administration in surgical patients may not be detrimental.
Our study has limitations. We used HES to replace autologous blood removal, which may have influenced our results as HES may have a “plugging” effect on vascular permeability.34–36 Even so, we found that the amount of fluid that leaked into the interstitium was often greater than the amount of fluid that remained in the intravascular space. Our results also should be extrapolated to general surgical cases with caution. Intraoperative bleeding was limited in our patients, and cases with more blood loss may experience greater absorption to the intravascular space to maintain a minimum BV. In such cases, we predict that absorption would become predominant, rather than leakage into the interstitium. Further study of fluid replacement in bleeding patients is needed. Jacob et al30 showed that withdrawn BVs of >1000 mL cannot be restored using 3 times the amount of Ringer’s solution, and they estimated that only 17% ± 10% was retained intravascularly. This finding suggests that fluid replacement in bleeding patients may demonstrate the same “leakage” effect as in our study.
There were other limitations to this study. We had no formal control group within this case series of orthognathic surgery patients. Also, we could only estimate the blood loss by weighing sponges and measuring the amount of suction. Although the amount of bleeding could not be measured precisely, the impact of any error in the bleeding amount was thought to be very small in this study. The average bleeding amount corresponded to 17.5% ± 12.5% of the extravascular leakage fluid volume (leakage into the interstitium, urine output, and bleeding amount). Therefore, even if there was a 10% error in the estimated amount of bleeding, the error in the estimated amount of leakage fluid would be <2%.
In conclusion, we demonstrated that infused crystalloid was mostly leaked postoperatively in cases with minimal bleeding. This finding supports the concept behind the revised Starling law and a clinical practice of giving less intraoperative fluid. A greater understanding of the fundamental kinetics of infused fluids during surgery will help refine our understanding and optimal application of intraoperative fluid therapy.
ESTIMATION OF FLUID MOVEMENT
First formula, before the withdrawal of whole blood: The arterial hemoglobin concentration (Hb① [g/dL]) was expressed as [the total hemoglobin content (HB [g])]/[circulating blood volume (BV [mL])]. HB refers to the initial hemoglobin content in the whole blood.
Second formula, after the withdrawal of whole blood: The arterial hemoglobin concentration (Hb② [g/dL]) was expressed as [the total hemoglobin content (HB [g]) − the lost hemoglobin content because of withdrawal (−400 × Hb① [g])]/[circulating blood volume (BV [mL]) − 400 mL from withdrawal + α]. The parameter α was defined as the inflow to the intravascular space, including the infused HES.
Third formula, after HES infusion: The blood sample was collected immediately after the infusion of 500 mL of HES. The arterial hemoglobin concentration (Hb③ [g/dL]) was expressed as [the total hemoglobin content (HB [g]) − the lost hemoglobin content because of withdrawal (−400 × Hb① [g])]/[circulating blood volume (BV [mL]) − 400 mL from withdrawal + α + 500 mL of HES]. This procedure was considered to represent a 500-mL volume expansion.
Fourth formula, after surgery: The arterial hemoglobin concentration (Hb④ [g/dL]) was expressed as [the total hemoglobin content (HB [g]) − the lost hemoglobin content because of withdrawal (400×Hb① [g]) − the lost hemoglobin content during surgery (L × (Hb③ + Hb④)/2)]/[circulating blood volume (BV [mL]) − 400 mL from withdrawal + α + 500 mL of HES − L − U① + β]. The parameter L represents the amount of blood loss during surgery. The parameter U① represents the urine volume during surgery. The parameter β was defined as the inflow of fluid into the intravascular space (absorption) from the extravascular space and interstitial tissue. This flow included the infusion of fluid as well as fluid exchange between the intravascular and extravascular spaces. The β value corresponded to the retained intravascular fluid volume during surgery. The hemoglobin value of the hemorrhagic blood was estimated using the average of the values obtained after HES infusion and after surgery as follows: (Hb③+Hb④)/2.
Fifth formula, after reinfusion of autologous blood: The arterial hemoglobin concentration (Hb⑤ [g/dL]) was expressed as [the total hemoglobin content (HB [g]) − the lost hemoglobin content because of withdrawal (400 × Hb① [g]) − the lost hemoglobin content during surgery (L × (Hb③ + Hb④)/2) + the hemoglobin contained in the stored autologous blood (400 × Hb① [g])]/[circulating blood volume (BV [mL]) − 400 mL from withdrawal + α + 500 mL of HES − L − U② + γ + 400 mL of homologous blood]. The parameter U② was defined as the total urine volume after the reinfusion of autologous blood. The parameter γ was defined as the final volume of absorption after the reinfusion of the autologous blood.
STATISTICAL POWER ANALYSIS
We performed a power analysis using G*Power software, version 3.1.4 (free software written by Franz Faul, Kiel University, Germany). Our results showed that r = 0.416 (between the infused volume and the leakage into the interstitium) and r = 0.016 (between the retained intravascular volume and the infused volume). From these data, the estimated total sample size was 31 patients to achieve a power of 0.8 for a correlation analysis between the infused volume and the leakage into the interstitium, and the total sample size was 24,144 patients to achieve a power of 0.8 for a correlation analysis between the retained intravascular volume and the infused volume. Therefore, a positive correlation observed for the 41 enrolled patients can be considered valid for the former correlation, whereas a negative correlation cannot be denied for the latter correlation until 24,144 patients have been enrolled. Because we used all enrolled 41 cases for statistical analysis instead of 31, we set significance level to <0.01 to avoid spurious judgment of significance.
Name: Akiko Nishimura, DDS, PhD.
Contribution: This author helped design the study, conduct the study, analyze the data, and write the manuscript.
Name: Yoko Tabuchi, DDS.
Contribution: This author helped conduct the study and analyze the data.
Name: Mutsumi Kikuchi, DDS, PhD.
Contribution: This author helped conduct the study and analyze the data.
Name: Rikuo Masuda, DDS, PhD.
Contribution: This author helped analyze the data.
Name: Kinuko Goto, DDS, PhD.
Contribution: This author helped analyze the data.
Name: Takehiko Iijima, DDS, DMSc, PhD.
Contribution: This author helped design the study, conduct the study, and write the manuscript.
This manuscript was handled by: Avery Tung, MD, FCCM.
1. Brandstrup B. Fluid therapy for the surgical patient. Best Pract Res Clin Anaesthesiol. 2006;20:265–283.
2. Lowell JA, Schifferdecker C, Driscoll DF, Benotti PN, Bistrian BR. Postoperative fluid overload: not a benign problem. Crit Care Med. 1990;18:728–733.
3. Brandstrup B, Svendsen PE, Rasmussen M, et al. Which goal for fluid therapy during colorectal surgery is followed by the best outcome: near-maximal stroke volume or zero fluid balance? Br J Anaesth. 2012;109:191–199.
4. Wuethrich PY, Burkhard FC, Thalmann GN, Stueber F, Studer UE. Restrictive deferred hydration combined with preemptive norepinephrine infusion during radical cystectomy reduces postoperative complications and hospitalization time: a randomized clinical trial. Anesthesiology. 2014;120:365–377.
5. Iijima T,, Brandstrup B, Rodhe P, Andrijauskas A, Svensen CH. The maintenance and monitoring of perioperative blood volume. Perioper Med. 2013;2:9–21.
6. Miller TE, Roche AM, Mythen M. Fluid management and goal-directed therapy as an adjunct to enhanced recovery after surgery (ERAS). Can J Anaesth. 2015;62:158–168.
7. Chappell D, Jacob M, Hofmann-Kiefer K, Conzen P, Rehm M. A rational approach to perioperative fluid management. Anesthesiology. 2008;109:723–740.
8. Woodcock TE, Woodcock TM. Revised Starling equation and the glycocalyx model of transvascular fluid exchange: an improved paradigm for prescribing intravenous fluid therapy. Br J Anaesth. 2012;108:384–394.
9. Levick JR, Michel CC. Microvascular fluid exchange and the revised Starling principle. Cardiovasc Res. 2010;87:198–210.
10. Jacob M, Bruegger D, Rehm M, et al. The endothelial glycocalyx affords compatibility of Starling’s principle and high cardiac interstitial albumin levels. Cardiovasc Res. 2007;73:575–586.
11. Michel CC. Starling: the formulation of his hypothesis of microvascular fluid exchange and its significance after 100 years. Exp Physiol. 1997;82:1–30.
12. Svensén C, Hahn RG. Volume kinetics of Ringer solution, dextran 70, and hypertonic saline in male volunteers. Anesthesiology. 1997;87:204–212.
13. Svensén CH, Brauer KP, Hahn RG, et al. Elimination rate constant describing clearance of infused fluid from plasma is independent of large infusion volumes of 0.9% saline in sheep. Anesthesiology. 2004;101:666–674.
14. Ueyama H, He YL, Tanigami H, Mashimo T, Yoshiya I. Effects of crystalloid and colloid preload on blood volume in the parturient undergoing spinal anesthesia for elective Cesarean section. Anesthesiology. 1999;91:1571–1576.
15. Ogawa R, Fujita T, Fukuda G. Blood volume studies in healthy Japanese adults. Kokyuu to Junkan. 1970;18:833–838.
16. Iijima T, Ueyama H, Oi Y, et al. Determination of the standard value of circulating blood volume during anesthesia using pulse dye-densitometry: a multicenter study in Japan. J Anesth. 2005;19:193–198.
17. Iijima T, Aoyagi T, Iwao Y, et al. Cardiac output and circulating blood volume analysis by pulse dye-densitometry. J Clin Monit. 1997;13:81–89.
18. Hahn RG, Svensén C. Plasma dilution and the rate of infusion of Ringer’s solution. Br J Anaesth. 1997;79:64–67.
19. Hahn RG. Volume kinetics for infusion fluids. Anesthesiology. 2010;113:470–481.
20. Norberg A, Brauer KI, Prough DS, et al. Volume turnover kinetics of fluid shifts after hemorrhage, fluid infusion, and the combination of hemorrhage and fluid infusion in sheep. Anesthesiology. 2005;102:985–994.
21. Svensén CH, Olsson J, Hahn RG. Intravascular fluid administration and hemodynamic performance during open abdominal surgery. Anesth Analg. 2006;103:671–676.
22. Svensén CH, Waldrop KS, Edsberg L, Hahn RG. Natriuresis and the extracellular volume expansion by hypertonic saline. J Surg Res. 2003;113:6–12.
23. Olsson J, Svensén CH, Hahn RG. The volume kinetics of acetated Ringer’s solution during laparoscopic cholecystectomy. Anesth Analg. 2004;99:1854–1860.
24. Jacob M, Bruegger D, Conzen P, Becker BF, Finsterer U, Rehm M. Development and validation of a mathematical algorithm for quantifying preoperative blood volume by means of the decrease in hematocrit resulting from acute normovolemic hemodilution. Transfusion. 2005;45:562–571.
25. Davies JWL, Topley E. A critical evaluation of red cell and plasma volume techniques patients with civilian injuries. J Clin Path. 1959;12:289–302.
26. Rehm M, Haller M, Brechtelsbauer H, Akbulut C, Finsterer U. Extra protein loss not caused by surgical bleeding in patients with ovarian cancer. Acta Anaesthesiol Scand. 1998;42:39–46.
27. Bremer F, Schiele A, Sagkob J, Palmaers T, Tschaikowsky K. Perioperative monitoring of circulating and central blood volume in cardiac surgery by pulse dye densitometry. Intensive Care Med. 2004;30:2053–2059.
28. Barta E, Kuzela L, Tordová E, Horecký J, Babusíková F. The blood volume and the renin-angiotensin-aldosterone system following open-heart surgery. Resuscitation. 1980;8:137–146.
29. Tatara T, Tsunetoh T, Tashiro C. Crystalloid infusion rate during fluid resuscitation from acute haemorrhage. Br J Anaesth. 2007;99:212–217.
30. Jacob M, Chappell D, Hofmann-Kiefer K, et al. The intravascular volume effect of Ringer’s lactate is below 20%: a prospective study in humans. Crit Care. 2012;16:R86.
31. Norberg A, Hahn RG, Li H, et al. Population volume kinetics predicts retention of 0.9% saline infused in awake and isoflurane-anesthetized volunteers. Anesthesiology. 2007;107:24–32.
32. Watanabe T, Ogawa R. Suppression of surgical hyperaldosteronism by potassium canrenoate during gynecologic surgery under sevoflurane anesthesia. Acta Anaesthesiol Scand. 2000;44:758–762.
33. Jacob M, Chappell D, Rehm M. The ‘third space’—fact or fiction? Best Pract Res Clin Anaesthesiol. 2009;23:145–157.
34. Tatara T, Itani M, Sugi T, Fujita K. Physical plugging does not account for attenuation of capillary leakage by hydroxyethyl starch 130/0.4: a synthetic gel layer model. J Biomed Mater Res B Appl Biomater. 2013;101:85–90.
35. Strunden MS, Bornscheuer A, Schuster A, Kiefmann R, Goetz AE, Heckel K. Glycocalyx degradation causes microvascular perfusion failure in the ex vivo perfused mouse lung: hydroxyethyl starch 130/0.4 pretreatment attenuates this response. Shock. 2012;38:559–566.
© 2016 International Anesthesia Research Society
36. Marx G, Pedder S, Smith L, et al. Attenuation of capillary leakage by hydroxyethyl starch (130/0.42) in a porcine model of septic shock. Crit Care Med. 2006;34:3005–3010.