Type: Evaluation Essays
Sample donated: Boyd Warren
Last updated: September 29, 2019
Abstract:Introduction. Perioperativefluid administration varies greatly in clinical practice and the use of point-of-careultrasonography is being discussed as a potential tool to guide management. Theaim of our pilot study is to evaluate whether preoperative inferior vena cavacollapsibility index (IVCCI) may be used to identify hypovolemia prior tosurgery and determine if there is an association between high IVCCI andintraoperative hypotension or fluid requirements. Methods. Fifty five patients undergoing non-emergent in-patient surgeryhad their IVCCI measured in the preoperative holding area and intraoperativehemodynamics and fluid administration recorded. Results.
We found IVCCI to be similar amongst patients whodeveloped hypotension intraoperatively and those who did not (33% +/-18% vs30.8%+/-15%, P=0.6). There was no correlation between IVCCI and intraoperativefluid administered. Conclusions.
IVCCIdid not demonstrate a strong correlation with intraoperative fluidadministration. With the range of IVCCI seen preoperatively, a much larger sample sizewould be needed to evaluate utility of IVCCI for predicting intraoperativehypotension. These findings call into question the routine preoperativeuse of IVC ultrasonography in the general elective surgery population. Key Words:preoperative transthoracic echocardiography, inferior vena cava collapsibility(IVCCI), intraoperative fluid administration, intraoperative hypotension. 1.
Introduction:Theuse of point of care ultrasound in the perioperative arena has recently gainedmomentum.1 Although there are studies showing the ability ofanesthesiologists to acquire echocardiographic images and interpret findings,2only a few have investigated specific applications of ultrasonography to alterperioperative care.3,4 Fluid management has a profound impact onpatient outcomes in various settings.
5 Inferior vena cava (IVC)collapsibility measured by ultrasound has been used for assessment of volumestatus and fluid responsiveness in the intensive care unit (ICU)6and emergency department.7 Preoperative fluid assessment is anotoriously difficult task for anesthesiologists and it has been advocated thatperioperative ultrasonography1 is the way to gain objective andreliable information. Intraoperative fluidadministration volumes have a wide variability in clinical practice.8Studies looking at patients undergoing intraabdominal and intrathoracicprocedures favor restrictive fluid administration.
9,10 However, itmight be useful to know which patients are hypovolemic prior to surgery, sincea restrictive fluid management approach may not be adequate for those patients.Unfortunately, there is no effective tool to determine the preoperativehydration status of patients prior to undergoing elective surgery and mostperioperative physicians have to use estimates of volume status.5Therange of IVC collapsibility with spontaneous respiration in patients presentingfor surgery has not been established. Recently, the preoperative measurement ofIVC collapsibility was correlated to hypotension after the induction of generalanesthesia showing a moderate correlation (r=0.46) between drop in meanarterial blood pressure after induction and IVCCI.11 However, theuse of IVC measurements in the preoperative setting has not been investigated toidentify hypovolemic elective surgery patients.
In the preoperative arena, ultrasound hasgained some popularity in evaluating cardiac function but has yet to becorrelated with intraoperative fluid administration. The primary aim of thisprospective observational study was to determine the range of IVCCI values inpatients undergoing elective surgery and the feasibility of using preoperativeIVC collapsibility to identify those who will have increased intraoperativefluid requirements and intraoperative hypotension. 2. Methods:2.1 Study Design: This prospective observational study was performed afterIRB approval from the University and the hospital’s Clinical Research ReviewCommittee. Written informed consent was obtained from all participants. Thestudy included 55 adult patients undergoing non-emergent surgery at our traumacenter who were expected to receive general anesthesia.
Patients were enrolledfrom January through May 2017. Patients excluded from the study were thoserequiring emergency surgery, patients requiring vasoactive drugs or intravenousfluid boluses for hemodynamic support preoperatively, and patients with woundsor dressings over the epigastric region (i.e., preventing the ultrasoundevaluation). 2.
2 Study Protocol: In the preoperative holding area, a single windowultrasound evaluation was performed by a critical care anesthesiologist trainedin point-of-care US using a Sonositeâ X-PorteUltrasound (Fujifilm Sonosite, Bothell, WA) with 5MHz- phased array probe inthe subcostal region while the patient was supine. The anesthesiologistperforming the ultrasound evaluation did not participate in the care of thepatient. Subcostal 4-chamber apical views were obtained to evaluate globalcardiac function and to exclude gross abnormalities of the valves orpericardium. The probe was then turned to obtain longitudinal views of the IVCin the retrohepatic region. Using M-Mode, measurements of the IVC were obtainedapproximately 3-5cm caudal to the diaphragm, distal to the hepatic vein convergencewith the IVC, throughout the respiratory cycle. With subjects taking normal effort spontaneous breaths, measurement ofthe IVC was recorded during a respiratory cycle. The maximum diameter duringexpiration (dmax) and the minimum diameter during inspiration (dmin) were usedfor the calculation of collapsibility.
The IVCCI was calculated using the followingformula: (dmax IVC – dmin IVC)/ dmax IVC. The anesthesia team was not informedof the findings from the preoperative ultrasound exam. After the ultrasound measurements were obtained, patientsproceeded with their scheduled surgery without any interventions from the studyteam. After the completion of the surgery the following intraoperative data wascollected: blood pressure values before and after induction of anesthesia, the cumulativedose of vasopressors used intraoperatively, cumulative intravenous fluidsadministered intraoperatively corrected for body weight and duration of surgery,and estimated blood loss and urine output throughout the case. 2.
3 Statistical AnalysisUsing previouslypublished data showing a moderate correlation (r=0.46) for IVCCI and decreasein mean arterial pressure (MAP) after the induction of general anesthesia,11a sample size of 45 was calculated to yield a power of 0.90 at alpha = 0.05 todetect an equally strong correlation to fluid administered.
We estimated 10-20%of patients would have unattainable IVC images or missing intraoperative datarequiring exclusion, therefore, 55 patients were enrolled. A plot of IVCCI and fluidadministered intraoperatively calculated as ml · kg-1 · hr-1was created using linear regression. Bivariate Pearson correlation wascalculated using IBM SPSS version 24 software. A two-sided Student’s t test wasused to compare difference between IVCCI in patients who experiencedhypotension (defined as MAP less than 60mmHg of any duration) to those thatwere normotensive throughout. 3. Results:Atotal of 55 patients were enrolled in the study. A total of 7 patients wereexcluded from the analysis.
The IVC was visualized and measurements obtained inall but 2 of the patients. There were 5 other patients excluded: 2 patients hadtheir surgery cancelled after enrollment, one patient did not receive GA, and 2patients had extended periods of intraoperative vitals missing from theelectronic medical record. Of the 48 patients used for analysis, the averageage was 42 (range 20-78), with 39 of the patients being classified as ASAphysical status 1 or 2 and an average BMI of 26. Of the 48 patients, only 3received blood products intraoperatively and no other colloids were used forvolume replacement in any patient. The IVCCI ranged from 1.9% to 74.1% with a meanCI of 32% (Table1). Only 7 of the 48 patients had greater than 50% collapsibilitywhich is commonly viewed as a cutoff for hypovolemia in healthy patients.
ThePearson correlation (R) for IVCCI to fluid administered intraoperatively was0.24, with wide variance and poor fit (Figure 1). Hypotensionafter induction, defined as a MAP of less than 60mmHg within 10 minutes ofinduction, was present in 10 of the 48 patients. Hypotension at any point inthe case was present in 26 of the 48 patients with a broad range of durationfrom 1 minute to 108 minutes of hypotension. The average IVCCI for thosepatients who did exhibit hypotension was 33% +/-18% (25.7-40.
4%) compared to30.8% +/-15% (23.8-37.7%) for those who did not experience intraoperativehypotension (P=.60). The IVCCI was equally distributed between the hypotensiveand normotensive patients (Figure 2). With the IVCCI measured in our population,an unrealistically large sample size study would need to be performed in orderto detect a difference in IVCCI.
The sample size required to achieve a Power of0.8 to detect a difference in means of at least 10-12% in IVCCI with alpha =0.05 between the participants that experienced hypotension and those who didnot would be well over a thousand. 4. Discussion: Thispilot study demonstrated a broad range of preoperative IVCCI in electivesurgery patients. Studies of acutely illpatients have demonstrated that IVCCI as low as 12% would indicate decreased enddiastolic volume12 or IVVCI greater than 25% would predict fluidresponsiveness.6 The mean IVCCI of 32% in our study was higherthan in the previously studied populations, but had poor correlation with fluidresponsiveness or hypotension.
Thismight indicate that the point of care echocardiography parameters establishedfor critically ill and acutely decompensated patients for fluid responsivenesswould not necessarily apply to patients presenting for elective surgery. Valuesof IVCCI in our study were lower than those obtained from healthy volunteers(47%) after an 8-hour fasting period.13 This might be explained bythe high percentage of inpatient participants who were receiving maintenancefluids during their fasting period. Therefore, it seems that preoperative patientsscheduled for elective surgery could have different parameters for determiningfluid status based on IVC dimensions than acutely ill patients. Inperioperative patients, a previous study has correlated IVCCI with changes in MAPafter induction which determined our pretest expectation of correlation withintraoperative fluids.11 In that study, by Zhang and colleagues,11there was good correlation between IVCCI and hypotension after induction, butthe association was not strong with static IVC diameter alone. The cutoff IVCCIof 43% was found to provide the best predictive value for hypotension afterinduction with AUC of 0.
9.11 However, this relationship with IVCCIdoes not seem to translate to changes as far as volume requirements intraoperativelyin our setting. In our study, when we look at hypotension defined as MAP lessthan 60mmHg throughout the entire case, the relationship with IVCCI is alsolost. IVCCI has been correlated to CVP14 and pulmonary artery pressures15in the critically ill.
The hope seems to be that the dynamic nature of the collapsibilityindex would be a useful monitor for preload and fluid responsiveness. However,the strong correlation to static pressures found in other studies14,15may not translate to a correlation in the perioperative setting for healthypatients. The selection of more acutely ill patients might provide a morerobust correlation between IVCCI and fluid requirements or hemodynamicparameters.Findingson the preoperative focused cardiac ultrasound may be more useful in theperioperative setting if used to answer specific clinical questions abouthemodynamics and alter management as it has been used in the ICU.
16Although in specific high-risk patients’ perioperative TTE is useful for diagnosisand management,17 this may not be so for all-comers as a screeningtool preoperatively. When preoperative focused ultrasonography is performed inurgent surgery, it was only able to provide some clinically useful informationto change management in sicker (ASA class ? 3) and older (Age 60+) patients.3The time and resources required to measure IVC diameters and asses cardiacfunction in all elective cases would likely not provide much clinical benefitover standard preoperative screening in the general population.
When IVCCI hasbeen studied in healthy volunteers to determine intravascular volume status andfluid responsiveness with passive leg raise, the sensitivity was found to bevery low with little clinical value.13 A more effective strategycould entail targeting perioperative ultrasonography to evaluate preoperativehydration in patients who are at increased risk of hemodynamic instability orwho present more challenging cases of fluid management. This presents an avenuefor further clinical investigation. Our study has several limitations. We could not control forconfounders such as the patients’ preoperative hydration strategies, theselection and the dose of induction agent, or intraoperative hemodynamic goals.
Also, the small sample size was notsufficient to detect a small difference in incidence of hypotension or fluidrequirements for different IVCCI values. Larger studies are needed usingtargeted point-of-care ultrasonography in specific populations where clinicalbenefit may be demonstrated. In conclusion, the preoperative IVVCI of the general electivesurgery population did not correlate to the volume of fluid administered intraoperativelyto the extent that changes in MAP have been noted. Our measured IVCCI wouldrequire a very robust patient enrollment to determine if there were anystatistically significant correlation between IVCCI and intraoperativehypotension in patients undergoing elective surgery calling into question anyclinical significance of this evaluation. Targeted preoperative transthoracicultrasound with IVC measurements may be a useful tool for specific clinicalquestions and in acutely ill patients, but may not be as high yield in thegeneral elective surgery population to determine hydration status. References:1- RamsinghD, Gudzenko V, Martin RD. Point-of-Careultrasound: Novel technology to routine perioperative assessment tool. AnesthAnalg.
2017;124(3):709-711.2- AndruszkiewiczP, Sobczyk D, Gorkiewicz-Kot I, Kowalik I, Gelo R, Stach O. Reliability offocused cardiac ultrasound by novice sonographer in preoperative anaestheticassessment: an observational study. Cardiovasc Ultrasound. 2015;13:45.3- BøtkerMT, Vang ML, Grøfte T, Sloth E, Frederiksen CA. Routine pre-operative focusedultrasonography by anesthesiologists in patients undergoing urgent surgicalprocedures.
Acta Anaesthesiol Scand. 2014;58(7):807-14.4- CantyDJ, Royse CF, Kilpatrick D, Bowyer A, Royse AG. The impact on cardiac diagnosisand mortality of focused transthoracic echocardiography in hip fracture surgerypatients with increased risk of cardiac disease: a retrospective cohort study.Anaesthesia. 2012;67(11):1202-9.
5- AditianingsihD, George YW. Guiding principles of fluid and volume therapy. Best Pract ResClin Anaesthesiol. 2014;28(3):249-60.
6- CorlKA, George NR, Romanoff J, Levinson AT, Chheng DB, Merchant RC, et al. Inferiorvena cava collapsibility detects fluid responsiveness among spontaneouslybreathing critically-ill patients. J Crit Care. 2017;41:130-137.7- DiptiA, Soucy Z, Surana A, Chandra S. Role of inferior vena cava diameter inassessment of volume status: a meta-analysis.
Am J Emerg Med.2012;30(8):1414-1419.8- LilotM, Ehrenfeld JM, Lee C, Harrington B, Cannesson M, Rinehart J. Variability inpractice and factors predictive of total crystalloid administration duringabdominal surgery: retrospective two-centre analysis. Br J Anaesth.2015;114(5):767-76.9- HanIW, Kim H, Heo J, Oh MG, Choi YS, Lee SE, et al. Excess intraoperative fluidvolume administration is associated with pancreatic fistula afterpancreaticoduodenectomy: A retrospective multicenter study.
Medicine(Baltimore). 2017;96(22):e6893. 10- EngOS, Arlow RL, Moore D, Chen C, Langenfeld JE, August DA, et al. Fluidadministration and morbidity in transhiatal esophagectomy. J Surg Res.2016;200(1):91-7. 11- ZhangJ, Critchley LA.
Inferior Vena Cava Ultrasonography before General AnesthesiaCan Predict Hypotension after Induction. Anesthesiology. 2016;124(3):580-9.
12- ZhaoJ, Wang G. Inferior Vena Cava Collapsibility Index is a Valuable andNon-Invasive Index for Elevated General Heart End-Diastolic Volume Index Estimationin Septic Shock Patients. Med Sci Monit. 2016;22:3843-48.13- GuiJ, Yang Z, Ou B, Xu A, Yang F, Chen Q, et al. Is the Collapsibillity Index ofthe Inferior Vena Cava an Accurate Predictor for the Early Detection ofIntravascular Volume Change? Shock. 2017. doi: 10.
1097/SHK.000000000000093214- IlyasA, Ishtiaq W, Assad S, Ghazanfar H, Mansoor S, Haris M, et al. Correlation ofIVC Diameter and Collapsibility Index With Central Venous Pressure in theAssessment of Intravascular Volume in Critically Ill Patients. Cureus. 2017;12;9(2):e1025.15- StawickiSP, Papadimos TJ, Bahner DP, Evans DC, Jones C. Correlations between pulmonaryartery pressures and inferior vena cava collapsibility in critically illsurgical patients: An exploratory study. Int J Crit Illn Inj Sci.
2016;6(4):194-199.16- FerradaP, Murthi S, Anand RJ, Bochicchio GV, Scalea T. Transthoracic focused rapidechocardiographic examination: real-time evaluation of fluid status incritically ill trauma patients. J Trauma. 2011;70(1):56-62.
17- KratzT, Steinfeldt T, Exner M, Dell Orto MC, Timmesfeld N, Kratz C, Skrodzki M, WulfH, Zoremba M. Impact of Focused Intraoperative Transthoracic Echocardiographyby Anesthesiologists on Management in Hemodynamically Unstable High-RiskNoncardiac Surgery Patients. J Cardiothorac Vasc Anesth. 2017;31(2):602-609.