Printer Friendly

Cardiac output and propofol concentrations in prone surgical patients.

The prone position is commonly used in lumbar spine surgery. Advantages of the position include excellent access to the surgical site and minimal bleeding if optimal positioning is obtained. Risks of the position include nerve and eye injury, respiratory insufficiency and cardiovascular instability (1). A reduction in cardiac output (CO) has been reported (2-8), possibly resulting from abdominal compression and venous pooling in the lower limbs, and this may lead to increased plasma concentrations of anaesthetic drugs and exaggerated drug effects, including deeper than intended anaesthesia (9).

Two previous studies have investigated the effect of prone positioning on cardiac output during propofol-based anaesthesia (5,7). Both reported a reduction in cardiac output once patients were positioned prone. However, Takizawa et al (5) did not monitor anaesthetic depth during their study. Furthermore they did not administer opioids, reducing the generalisability of their results. Bennarosh et al (7) adjusted the propofol target concentration to maintain a constant depth of anaesthesia, resulting in lower targets in the prone position, but did not measure plasma propofol concentrations. The aim of our study therefore, was to compare CO and measured plasma propofol concentrations at pseudo-steady state in the supine and prone positions, in healthy patients having elective lumbar spine surgery positioned on a Wilson frame. Our hypotheses were that CO would be lower and measured plasma propofol concentrations would be higher in the prone position than in the supine position.

MATERIALS AND METHODS

Prospective approval was obtained from the Human Research Ethics Committee of the Royal Melbourne Hospital. The study was not randomised with respect to initial patient position and was not blinded. Patients were eligible if they were aged 18 to 70 years, of American Society of Anesthesiologists physical status I to II and were presenting for elective lumbar spine surgery positioned on the Wilson frame. Those with inadequate English comprehension due to a language barrier, cognitive deficit or intellectual disability, those with significant cardiorespiratory impairment and those with a body mass index above the parameter limits of the Schnider (10) and Minto (11) pharmacokinetic/dynamic models were excluded. All eligible patients were approached and recruited patients provided written informed consent prior to commencement of the study.

Patients fasted for at least six hours prior to induction of anaesthesia. All patients were fitted with venous compression stockings preoperatively, but were not prescribed any sedative or analgesic premedication. Intravenous (16 G, forearm vein) and intra-arterial (20 G, radial artery) cannulae were inserted under local anaesthesia and intravenous infusion of Hartmann's solution 100 ml/hour was commenced. Monitoring with electrocardiography, invasive arterial blood pressure, oxygen saturation, inspired and expired gases and the bispectral index (BIS) was commenced. The arterial line transducer was maintained at the level of the heart in the supine and prone positions.

Anaesthesia was induced with the patient in the supine position using target-controlled infusions of propofol (5 [micro]g/ml target effect-site concentration using the model of Schnider et al (10)) and remifentanil (5 ng/ml target effect-site concentration using the model of Minto et al (11)). After loss of consciousness, rocuronium (0.6 mg/kg) was administered and the trachea was intubated. Mechanical ventilation was instituted using 100% [O.sub.2] to maintain the end-tidal carbon dioxide partial pressure in the range of 35 to 45 mmHg. After intubation, the propofol target was decreased to 3 [micro]g/ml, the remifentanil target was decreased to 3 ng/ml and the 20-minute supine phase of the study began. These targets were selected with the goal of maintaining appropriate anaesthetic depth without haemodynamic instability in the absence of surgical stimulation. We chose a 20-minute sampling period to ensure effect-site equilibration with respect to hypnosis and haemodynamic changes consequent upon the position (12). After 20 minutes, the patients were turned prone onto the Wilson frame, with the frame being adjusted by the surgeon for optimal positioning, and the prone phase of the study began. The surgeons were allowed to apply skin preparation and the surgical drapes and place a marker under radiological control during this period, but were prohibited from injecting local anaesthetic or adrenaline or starting the operation. After 20 minutes in the prone position, the study was complete and surgery commenced. Patients were interviewed after surgery about their recall of events during anaesthesia.

Cardiac output was measured during each 20-minute period by a partial rebreathing method based on the Fick principle (13) using the NICO[R] Cardiopulmonary Management System (Philips Healthcare, Suresnes, France). The NICO[R] sensor was connected directly to the endotracheal tube.

Five millilitre samples of arterial blood for propofol assay were obtained at 15, 17.5 and 20 minutes of each 20-minute period. The samples were stored at 4[degrees]C for up to 10 weeks before analysis (propofol concentrations decrease at <0.2% /week at 4[degrees]C). They were analysed using a high-performance liquid chromatography assay modified from the method of Plummer (14). This assay is linear to at least 20 [micro]g/ml, has a detection limit of 0.025 [micro]g/ml, an inter-assay coefficient of variation of 4.1% and an intra-assay coefficient of variation of 4.9%.

Significant hypotension, defined as a systolic blood pressure (SBP) <85 mmHg, or a decrease in SBP of concern to the anaesthetist, was treated with metaraminol 0.5 mg intraveneously. Excessively deep anaesthesia, defined as a BIS value <30 with significant hypotension, was treated with a decrease in propofol target concentration in 0.5 [micro]g/ml increments and treatment of hypotension with metaraminol 0.5 mg intraveneously until it was controlled. A BIS value >60 with or without other signs of inadequate anaesthetic depth, was treated with an increase in propofol target concentration in 1 [micro]g/ml increments. Patients in whom propofol concentration was changed were removed from the study. Remifentanil target concentrations were not altered.

The primary endpoints were CO and measured plasma propofol concentration and secondary variables included mean arterial pressure (MAP) and BIS. The study of Takizawa et al (5) was used as the basis for our sample size calculation. In their study, CO decreased from 5.2 (0.8) l/minute in the supine to 4.3 (1.0) l/minute in the prone position, and the measured plasma propofol concentration increased from 2.05 (0.20) to 2.38 (0.28) [micro]g/ml. We calculated that with 20 patients we would achieve 81% power to detect this change in CO and 96% power to detect this change in measured plasma propofol concentration.

Statistical analyses were undertaken using Stata 10.1. Continuous data were assessed for normality using graphical methods. Normally distributed data were summarised using mean (standard deviation), skewed data were summarised using median (range) and categorical data were summarised using the count (percent). Paired two-tailed t-tests or Wilcoxon signed-rank tests were used to compare variables in the supine and prone positions at 20 minutes. Mixed models, with a subject-specific random effect to allow for within-subject correlation of repeated measurements, were used to assess the differences in CO, MAP and BIS values over the 20-minute period and between the supine and prone positions, and to assess changes in measured plasma propofol concentrations between 15 and 20 minutes of infusion in each position. The variability of within-subject errors was allowed to differ between the two groups and was assumed to follow an autoregressive structure of order within groups over time. Group and time were both included in the mixed model as categorical fixed effects. Interaction terms between group and time were included to assess changes in the difference between the two positions over time. Normality of random effects and residuals was assessed using graphical methods. Results are presented as differences in means with 95% confidence intervals and Wald test P values. All P values presented are two-tailed, with P values of <0.05 considered statistically significant.

[FIGURE 1 OMITTED]

RESULTS

Twenty-five patients were recruited and 15 were included in the analyses. Two patients were excluded due to technical difficulties with CO measurement and eight were excluded because target propofol concentration was increased due to BIS >60 (all patients were in the supine position when the initial increase was made). With 15 patients we had 69% power to detect a 0.9 l/minute change in CO and 90% power to detect a 0.33 /g/ml change in measured plasma propofol concentration. No patients reported recall of events during the study postoperatively. Included patients were aged 50 (13) years and had a body mass index of 27 (6) and eight were male.

Cardiac output was similar at 20 minutes of infusion in the supine and prone positions whereas MAP was significantly higher in the prone position (Table 1). Measured plasma propofol concentration and BIS values were not significantly different at 20 minutes.

Figure 1 shows the estimated marginal means of CO, BIS, MAP and plasma propofol concentration in each position over time.

In the mixed model for CO there was no evidence of an interaction between group and time (P=0.53), so we included only main effects for group and time in the model. There was no evidence of a difference in CO between the supine and prone positions (estimated mean difference 0.3 (95% CI -1.0 to 0.3), P=0.34). Cardiac output did not change significantly over the 20 minutes of infusion in either group (P=0.37).

There was a statistically significant decline over time in MAP in both the supine and prone positions (supine P=0.02 and prone P <0.001). Mean arterial pressure was initially significantly higher in the prone position (estimated difference in mean -25.3 mmHg, 95% CI -35.8 to -14.9), but by 10 minutes the difference decreased to approximately 13 units and remained at that level (Table 2).

For BIS, there was no evidence of an interaction between group and time (P=0.20). BIS decreased over time (P=0.02) but there was no evidence of a group difference (P=0.12).

Six patients received metaraminol, 0.5 mg intravenously, to treat SBP <85 mmHg during the study (two during the supine position alone, one during the prone position alone and three in both positions). There was no significant difference in CO at 20 minutes in the supine and prone position in these patients (5.2 [1.5] vs 5.1 [2.2] l/minute; P=0.67).

Measured plasma propofol concentrations varied between 15 and 20 minutes of infusion by median 1% (range -23 to +6%) in the supine position and -1% (-16 to +20%) in the prone position, and between 20 minutes and 17.5 minutes by 0% (-9 to + 16%) and 2% (-15 to +13%) respectively. There was no evidence of an interaction between group and time (P=0.74). There was no evidence of a change over time (P=0.84) or between groups (P=0.92).

DISCUSSION

Cardiac output, measured plasma propofol concentration and BIS were not significantly different at pseudo-steady state in the supine and prone positions in healthy patients having elective lumbar spine surgery positioned on a Wilson frame. The 95% confidence intervals of the differences in CO, propofol concentration and BIS indicate that clinically significant differences in these variables are also unlikely, despite the slightly lower than expected power of our study. Mean arterial pressure was greater at 20 minutes of infusion in the prone patients compared to the supine patients but was adequate in both groups. This difference could be due to increased systemic vascular resistance which was not measured in our study. We conclude that effect-site targets for propofol do not need to be decreased in patients who are turned prone onto a Wilson frame for elective lumbar spine surgery.

Our results are similar to the study of Toyota et al (15) in which there was no change in cardiac index in spinal surgery patients anaesthetised with isoflurane. Like ours, these patients were healthy and were positioned on longitudinal bolsters which are similar to the Wilson frame in that they support the patient while minimising abdominal (and possibly inferior vena caval) compression.

Our results contrast with previous reports in which a 17 to 35% decrease in CO or cardiac index was reported (2-8), including the two previous studies in which patients were anaesthetised with propofol and remifentanil (5,7). No short- or long-term adverse effects of decreased CO were noted in these studies. Only one previous study has investigated the Wilson frame: Dharmavaram et al (6) compared the haemodynamic effects of five support systems in American Society of Anesthesiologists physical status I to III patients presenting for elective lumbar spine surgery and CO decreased from 6.4 (SEM 0.3) l/minute to 5.2 (SEM 0.3) l/minute in the Wilson frame group. Only one previous study has utilised a depth of anaesthesia monitor: Bennarosh et al7 titrated BIS to 40 to 50 and this resulted in decreased target plasma propofol concentrations when the patients were prone: we targeted constant propofol concentrations and demonstrated steady BIS values. Finally, only Takizawa et al measured actual propofol concentrations (5) which is important as target-controlled infusions devices may not deliver target concentrations accurately (16). Indeed, the four-fold range in measured propofol concentrations (Figure 2), despite all patients receiving dose regimens to deliver identical effect-site concentrations, highlights the inter-patient variability between predicted and actual propofol concentrations inherent in even a healthy population. However, Takizawa et al did establish the stability of propofol concentration before CO measurement and did not measure anaesthetic depth (5). The factors that may have resulted in a different result in these studies compared with ours include our exclusion of patients with significant cardio-respiratory illness, maintenance of similar and constant plasma propofol concentrations in the supine and prone positions, tight control of depth of anaesthesia (which itself alters CO), and allowing drug concentrations to come to pseudo-equilibrium with the effect site.

[FIGURE 2 OMITTED]

The goal of our protocol was to maintain stable plasma and effect-site propofol concentrations throughout the study, to avoid inadequate and excessively deep anaesthesia and to avoid hypotension. To achieve this, we chose effect-site steered target-controlled infusions, waited for 20 minutes for equilibration in each position and monitored anaesthetic depth to establish that the depth of hypnosis was comparable. We piloted propofol and remifentanil maintenance effect-site concentrations of 2 mg/ml and 2 ng/ml but increased our targets to 3 mg/ml and 3 ng/ml respectively because of high BIS values. Even at these concentrations, we needed to exclude eight patients from the trial because of BIS values >60, but needed to administer metaraminol to six patients for SBP <85 mmHg. Variance in cardiac output was larger than predicted in our sample size calculation, and this, along with the loss of patients, has reduced the power of our study to find a difference in CO.

For patient safety reasons, hypotension must be treated, but the choice of agent is challenging as all methods will alter CO. Our study is limited by the fact that we did not assess volume status objectively and ensure normovolaemia before starting. However, other investigators have done this and reported a decrease in CO (6,7,15). A fluid bolus once hypotension develops may not increase blood pressure expediently and [alpha]- and [beta]-agonists affect CO. Excluding these patients would decrease our sample size further and there was no difference in the effect size or direction between patients who received metaraminol and those who did not.

A further limitation of our study may be our choice of the NICO device for CO estimation. Comparative studies (17-19) report a relatively loose agreement between CO measured using NICO and thermodilution, the method most commonly considered as the clinical 'gold standard'. The NICO device assumes complete mixing of gases in the alveoli. However, in reality, there is likely to be incomplete mixing, as seen by breath to breath variations in inspiratory and expiratory alveolar concentrations of anaesthetic agents. Furthermore, the presence of intrapulmonary shunts or haemodynamic instability may affect the accuracy of the CO estimations. The volume of dead space and degree of shunt may change when the prone position is assumed, although this would be difficult to estimate (1). We chose the NICO because it provided a non-invasive estimate of CO in patients in whom invasive monitoring could not be justified. Using a less accurate method of cardiac output measurement means that any calculated 95% confidence intervals will be an overestimate of the ability to exclude a clinically meaningful difference.

There is a plausible physiologic rationale for an inverse relationship between CO and propofol concentrations. Upton et al (20) reported that initial propofol concentrations were decreased in sheep with higher CO induced by carbon dioxide and increased in sheep with lower CO induced by metaraminol infusion. Similarly, Myburgh et al (9) induced elevated CO in sheep with adrenaline, noradrenaline and dopamine and reported reduced propofol concentrations. This observation was further confirmed in pigs by Kurita et al (21). The proposed mechanism is increased first-pass dilution and increased clearance with increased CO (and vice versa for decreased CO) (22), and it is an effect with influences propofol concentrations for some time after induction of anaesthesia. However, the lack of an effect on the present study relates to the minimal CO effects of positioning changes in this surgical model. A significant positioning effect on CO and propofol requirements, however, cannot be excluded in other patient groups, such as the elderly, those with cardiovascular co-morbidities, and those on medications such as anti-hypertensives.

In conclusion, CO and measured plasma propofol concentration were not significantly different at pseudo-steady state in the supine and prone positions in healthy patients having elective lumbar spine surgery positioned on a Wilson frame. Effect-site targets for propofol do not need to be decreased when turning healthy patients prone.

ACKNOWLEDGEMENTS

We would like to thank research nurses, Ms Jacklyn De Gabriele and Julia Pearce, the neuroanaesthetists and neurosurgeons at the Royal Melbourne Hospital, and Ms Kelly Storer of Mayo Healthcare Pty Ltd (Rosebery, NSW) for their assistance with this project.

REFERENCES

(1.) Edgcombe H, Carter K, Yarrow S. Anaesthesia in the prone position. Br J Anaesth 2008; 100:165-183.

(2.) Backofen J. Hemodynamic changes with prone positioning during general anethesia. Anesth Analg 1985; 64:194.

(3.) Yokoyama M, Ueda W, Hirakawa M, Yamamoto H. Hemodynamic effect of the prone position during anesthesia. Acta Anaesthesiol Scand 1991; 35:741-744.

(4.) Hatada T, Kusunoki M, Sakiyama T, Sakanoue Y, Yamamura T, Okutani R et al. Hemodynamics in the prone jackknife position during surgery. Am J Surg 1991; 162:55-58.

(5.) Takizawa D, Hiraoka H, Nakamura K, Yamamoto K, Horiuchi R. Influence of the prone position on propofol pharmacokinetics. Anaesthesia 2004; 59:1250-1251.

(6.) Dharmavaram S, Jellish WS, Nockels RP, Shea J, Mehmood R, Ghanayem A et al. Effect of prone positioning systems on hemodynamic and cardiac function during lumbar spine surgery: an echocardiographic study. Spine (Phila Pa 1976) 2006; 31:1388-1393, discussion 1394.

(7.) Bennarosh L, Peuch C, Cohen J, Dauzac C, Guigui P, Mantz J et al. Effects of the knee-chest position on cardiac index and propofol requirements during bispectral index (BIS)-guided spine surgery. Ann Fr Anesth Reanim 2008; 27:158-163.

(8.) Poon KS, Wu KC, Chen CC, Fung ST, Lau AW, Huang CC et al. Hemodynamic changes during spinal surgery in the prone position. Acta Anaesthesiol Taiwan 2008; 46:57-60.

(9.) Myburgh JA, Upton RN, Grant C, Martinez A. Epinephrine, norepinephrine and dopamine infusions decrease propofol concentrations during continuous propofol infusion in an ovine model. Intensive Care Med 2001; 27:276-282.

(10.) Schnider TW, Minto CF, Shafer SL, Gambus PL, Andresen C, Goodale DB et al. The influence of age on propofol pharmacodynamics. Anesthesiology 1999; 90:1502-1516.

(11.) Minto CF, Schnider TW, Egan TD, Youngs E, Lemmens HJ, Gambus PL et al. Influence of age and gender on the pharmacokinetics and pharmacodynamics of remifentanil. I. Model development. Anesthesiology 1997; 86:10-23.

(12.) Kazama T, Ikeda K, Morita K, Kikura M, Doi M, Ikeda T et al. Comparison of the effect-site k(eO)s of propofol for blood pressure and EEG bispectral index in elderly and younger patients. Anesthesiology 1999; 90:1517-1527.

(13.) Capek JM, Roy RJ. Noninvasive measurement of cardiac output using partial CO2 rebreathing. IEEE Trans Biomed Eng 1988; 35:653-661.

(14.) Plummer GF. Improved method for the determination of propofol in blood by high-performance liquid chromatography with fluorescence detection. J Chromatogr 1987; 421:171-176.

(15.) Toyota S, Amaki Y. Hemodynamic evaluation of the prone position by transesophageal echocardiography. J Clin Anesth 1998; 10:32-35.

(16.) Masui K, Upton RN, Doufas AG, Coetzee JF, Kazama T, Mortier EP et al. The performance of compartmental and physiologically based recirculatory pharmacokinetic models for propofol: a comparison using bolus, continuous, and target-controlled infusion data. Anesth Analg 2010; 111:368-379.

(17.) van Heerden PV, Baker S, Lim SI, Weidman C, Bulsara M. Clinical evaluation of the non-invasive cardiac output (NICO) monitor in the intensive care unit. Anaesth Intensive Care 2000; 28:427-430.

(18.) Nilsson LB, Eldrup N, Berthelsen PG. Lack of agreement between thermodilution and carbon dioxide-rebreathing cardiac output. Acta Anaesthesiol Scand 2001; 45:680-685.

(19.) Maxwell RA, Gibson JB, Slade JB, Fabian TC, Proctor KG. Noninvasive cardiac output by partial CO2 rebreathing after severe chest trauma. J Trauma 2001; 51:849-853.

(20.) Upton RN, Ludbrook GL, Grant C, Martinez AM. Cardiac output is a determinant of the initial concentrations of propofol after short-infusion administration. Anesth Analg 1999; 89: 545-552.

(21.) Kurita T, Morita K, Kazama T, Sato S. Influence of cardiac output on plasma propofol concentrations during constant infusion in swine. Anesthesiology 2002; 96:1498-1503.

(22.) Upton R. Relationships between steady state blood concentrations and cardiac output during intravenous infusions. Biopharm Drug Dispos 2000; 21:69-76.

K. LESLIE *, C. Y.-X. WU ([dagger]), A. R. BJORKSTENX, D. L. WILLIAMS ([double dagger]), G. LUDBROOK **, E. WILLIAMSON ([dagger][dagger])

Department of Anaesthesia and Pain Management, Royal Melbourne Hospital, Departments of Pharmacology and Medicine, University of Melbourne, Melbourne, Victoria and Department of Anaesthesia, University of Adelaide, Adelaide, South Australia, Australia

* M.B., B.S., M.D., M.Epi., F.A.N.Z.C.A, Head of Research, Department of Anaesthesia and Pain Management and Honorary Professorial Fellow, Department of Pharmacology, University of Melbourne.

([dagger]) Medical Student, Department of Anaesthesia and Pain Management.

([double dagger]) B.Sc., Ph.D., Medical Scientist, Department of Anaesthesia and Pain Management.

[section] M.B., B.S., F.A.N.Z.C.A., Director of Anaesthesia, Department of Anaesthesia and Pain Management, Royal Melbourne Hospital and Clinical Associate Professor, Department of Medicine, University of Melbourne.

** M.B., B.S., Ph.D., F.A.N.Z.C.A., Professor, Department of Anaesthesia, University of Adelaide.

[dagger][dagger] Ph.D., Post-doctoral Fellow, Department of Epidemiology and Preventive Medicine, Monash Univeristy and Molecular, Environmental, Genetic and Analytic Epidemiology, School of Population Health, University of Melbourne.

Address for correspondence: Professor K. Leslie. Email: kate.leslie@ mh.org.au

Accepted for publication on April 14, 2011.
TABLE 1
Measured variables 20 minutes after positioning in the supine and
prone positions in 20 patients. Supine and prone values are median
(range) or mean (SD). Mean difference values are mean (SD) [95%
confidence interval].

 Supine Prone

Oxygen saturation, % * 100 (97-100) 100 (98-100)
End-tidal C[O.sub.2] partial 39 (5) 39 (4)
 pressure, mmHg
Heart rate, bpm 59 (12) 60 (9)
Mean arterial blood pressure, 65 (10) 78 (15)
 mmHg
Cardiac output, l/min 6.1 (1.6) 6.1 (1.9)
Measured propofol, [micro]g/ml 2.55 (0.89) 2.53 (0.90)
Bispectral index 41 (7) 40 (9)

 Mean difference P value

Oxygen saturation, % * 0 [-0.3 to 0.3] 0.40
End-tidal C[O.sub.2] partial -0.1 (4.1) [-2.4 to 2.2] 0.95
 pressure, mmHg
Heart rate, bpm -0.5 (5.9) [-3.7 to 2.8] 0.76
Mean arterial blood pressure, -13 (15) [-21 to -5] 0.003
 mmHg
Cardiac output, l/min 0.1 (2.2) [-1.1 to 1.3] 0.87
Measured propofol, [micro]g/ml 0.02 (0.9) [-0.5 to 0.5] 0.93
Bispectral index 1 (11) [-5 to 7] 0.80

* Median (range with bootstrapped confidence interval for median
difference).

TABLE 2
Estimated within-subject mean differences over the 20 minute period
from mixed models

Outcome Time Estimated difference, P value
 supine--prone
 (difference [95% CI])

Cardiac output, l/min -- 0.32 (-0.34 to 0.98) 0.34
Bispectral index -- 2.8 (-0.7 to 6.4) 0.12
Mean arterial pressure, 0 -25.3 (-35.8 to -14.9) <0.001
 mmHg
 5 -5.9 (-16.4 to 4.5)
 10 -10.8 (-21.3 to -0.3)
 15 -13.5 (-24.0 to -3.1)
 20 -13.3 (-23.8 to -2.9)
Measured propofol -- 0.02 (-0.41 to 0.45) 0.92
 concentration,
 [micro]g/ml

For cardiac output, propofol and bispectral index, the model included
main effect terms for group and time only. For mean arterial pressure,
the model also included a group-by-time interaction. CI=confidence
interval.
COPYRIGHT 2011 Australian Society of Anaesthetists
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2011 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Leslie, K.; Wu, C.Y.-x.; Bjorksten, A.R.; Williams, D.L.; Ludbrook, G.; Williamson, E.
Publication:Anaesthesia and Intensive Care
Article Type:Report
Geographic Code:8AUST
Date:Sep 1, 2011
Words:4186
Previous Article:Initial lactate level and mortality in septic shock patients with hepatic dysfunction.
Next Article:The association between intraoperative electroencephalogram-based measures and pain severity in the post-anaesthesia care unit.
Topics:

Terms of use | Privacy policy | Copyright © 2019 Farlex, Inc. | Feedback | For webmasters