Tcip(bdndd

v

c

z

v

c

R

a

s

I

v

c

J Oral Maxillofac Surg68:1007-1012, 2010

Desflurane Versus Sevoflurane to ReduceBlood Loss in Maxillofacial Surgery

Alessandro Rossi, MD,* Gabriele Falzetti, MD,†

Abele Donati, MD,‡ Giovanni Orsetti, MD,§

and Paolo Pelaia, MD�

Purpose: In our study, desflurane was hypothesized to reduce blood loss more than sevoflurane, bothused with targeted mild controlled hypotension.

Patients and Methods: A total of 20 American Society of Anesthesiologists Class I patients undergoingmaxillofacial elective surgery for maxillary and mandibular osteotomies were randomized to a desfluranegroup or a sevoflurane group. Anesthesia was performed with an end tidal value of the inhaled agent toobtain a bispectral index value �30 but without burst-suppression patterns (minimal alveolar concen-tration age-corrected between 0.7 and 0.9). Remifentanil was administered at a dose of 0.5 �g · kg�1 ·min�1 to obtain analgesia and a �2 surgical field level in Fromme’s modified scale. Sodium-nitroprussidewas administered on demand to have a surgical field level of �2 when the anesthesia plan was notsufficient to achieve this target. The minimal value of the mean arterial pressure achievable was 60 mm Hg.

Results: In the desflurane group, blood loss was more restricted. The hypotensive drug was used in 8patients in the sevoflurane group and 2 patients in the desflurane group.

Conclusions: Anesthesia with desflurane can reduce blood loss and could give an acceptable surgicalfield with mild controlled hypotension and with a substantial reduction in the vasoactive drug require-ment. These data need to be assessed with an enlargement of the statistical sample.© 2010 American Association of Oral and Maxillofacial Surgeons

J Oral Maxillofac Surg 68:1007-1012, 2010

ptrTcos

an

v

c

R

a

s

I

C

V

u

©

0

he vascular plexus of bone in the surgical fields forraniofacial surgery are highly vascularized, and bleed-ng during the surgical procedures can be an importantroblem for surgeon activity.1 Induced hypotensionmean arterial pressure [MAP] 50 mm Hg) to reducelood loss and improve surgeon activity could also be aangerous practice for the anesthesiologist.1 This tech-ique is related to many risks, including tissue perfusionecreases, oxygen delivery shortages, quick oxygenebt, and toxicity linked to drugs such as sodium-nitro-

*Clinica di Anestesia e Rianimazione, Azienda Ospedaliero-Uni-

ersitaria “Ospedali Riuniti” Umberto I, G Salesi, GM Lancisi, An-

ona, Italy.

†Dipartimento di Neuroscienze, Sezione di Anestesia e Rianima-

ione, Università Politecnica delle Marche, Ancona, Italy.

‡Clinica di Anestesia e Rianimazione, Azienda Ospedaliero-Uni-

ersitaria “Ospedali Riuniti” Umberto I, G Salesi, GM Lancisi, An-

ona, Italy; Dipartimento di Neuroscienze, Sezione di Anestesia e

ianimazione, Università Politecnica delle Marche, Ancona, Italy;

nd Emergency and Acceptance Department, Anesthesia and Inten-

ive Care Unit Section, Politechnical University of Marche, Ancona,

taly.

§Clinica di Anestesia e Rianimazione, Azienda Ospedaliero-Uni-

ersitaria “Ospedali Riuniti” Umberto I, G Salesi, GM Lancisi, An-

ona, Italy.d

1007

russide.2,3 In addition, some investigators have doubtedhe useful action of sodium-nitroprusside to promote theeduction of blood loss during induced hypotension.4

hese risks could be reduced with the use of mildontrolled hypotension (MAP �60 mm Hg) targetedn the basis of the surgeon’s request to achieve aurgical field useful for the procedure.

Recent findings have suggested that desflurane asn inhaled agent is useful for improving hemody-amic stability in splanchnic districts.5

�Clinica di Anestesia e Rianimazione, Azienda Ospedaliero-Uni-

ersitaria “Ospedali Riuniti” Umberto I, G Salesi, GM Lancisi, An-

ona, Italy; Dipartimento di Neuroscienze, Sezione di Anestesia e

ianimazione, Università Politecnica delle Marche, Ancona, Italy;

nd Emergency and Acceptance Department, Anesthesia and Inten-

ive Care Unit Section, Politechnical University of Marche, Ancona,

taly.

Address correspondence and reprint requests to Dr Donati,

linica di Anestesia e Rianimazione, Ospedale Regionale Torrette,

ia Conca 71, Torrette di Ancona 60020 Italy; e-mail: a.donati@

nivpm.it

2010 American Association of Oral and Maxillofacial Surgeons

278-2391/10/6805-0010$36.00/0

oi:10.1016/j.joms.2008.12.012

wumast

P

atpotrtdkr

tMtaftsrmvtacs

ttglmm

sbwaiig�s

a

wc

vwettt

tr

udsiraoMntP(s

R

2

O

M

RJ

1008 DESFLURANE VERSUS SEVOFLURANE TO REDUCE BLOOD LOSS

The aim of the present study was to determinehich inhaled agent (desflurane versus sevoflurane)sed during targeted mild controlled hypotension wasore suitable to reduce blood loss. The secondary

im was to evaluate the action of sodium-nitroprus-ide when used to induce and maintain mild con-rolled hypotension.

atients and Methods

Our institutional review board approved the study,nd all patients provided written informed consent. Aotal of 20 American Society of Anesthesiologistshysical status I patients, who were 18 to 40 yearsld, and scheduled for maxillary and mandibular os-eotomy, were randomly divided into 2 groups toeceive balanced anesthesia with remifentanil and ei-her desflurane or sevoflurane. Anesthesia was in-uced with fentanyl (1.5 �g · kg�1), propofol (2 mg ·g�1), rocuronium bromide (0.6 mg · kg�1), andemifentanil (0.25 �g · kg�1 · min�1).

The hypnotic status was assessed using the bispec-ral index (Aspect Medical Systems, Natick, MA).aintenance of hypnosis was performed with an end

idal value of the inhaled agent to obtain a minimallveolar concentration (MAC), age-corrected,6 of 0.8or the total length of the surgical procedure. Thus,he bispectral index was kept at �30, without burst-uppression patterns. Analgesia was maintained withemifentanil as a continuous infusion (0.5 �g · kg�1 ·in�1) to maintain a MAP of 20% lower than the basal

alue. This infusion rate was assumed to be sufficiento abolish the adrenergic response to surgical stressnd to avoid hypotension as a collateral effect ofardiocirculatory depression resulting from an exces-ive dosage of the analgesic drug.7-9

All patients received routine intraoperative moni-oring. The radial artery was cannulated to measurehe arterial pressure and to check the arterial bloodas tension, electrolytes, pH, base excess, and bloodactate levels every hour. The cardiac index (CI) was

easured using noninvasive cardiac output (Nova-etric Medical Systems, Wallingford, CT).The surgical procedures were performed using the

ame surgical equipment. Every 5 minutes after theeginning of the surgical procedure, the surgeons,ho were unaware of the blood pressure levels and

nesthetic technique, were asked to rate the operat-ng conditions on a scale from 0 to 3 using the mod-fied Fromme ordinal scale of assessment of the sur-ical field (Table 1).1 The target was to obtain a level2 of Fromme’s modified scale evaluation of the

urgical field.If the surgical field was at a 2 to 3 level 10 minutes

fter starting the procedure, sodium-nitroprusside s

as administered to obtain a Fromme modified surgi-al field level of �2.A MAP of 60 mm Hg was considered the lesser

alue tolerated during hypotension. The MAP and CIere registered at baseline (anesthesia induction),

very 5 minutes in maxillary time and in mandibularime, and at extubation. Blood loss was measured ashe blood in mL · kg�1 noted at the end of maxillaryime and mandibular time.

To avoid intersubject differences, the duration ofhe shorter surgical operation was assumed as theeference for the other procedures.

Continuous variables are reported as the mean val-es and intervals of standard deviation (� SD), me-ian, and minimal to maximal range, and are repre-ented as graphic box plots with the median,nterquartile range, and 5th and 95th percentiles, asequired. Noncontinuous variables are reported asbsolute values. Intergroups differences for continu-us variables were analyzed using the nonparametricann-Whitney U test, and intergroup differences foroncontinuous variables were analyzed using the �2

est. Statistical analysis was performed using Statisticalackage for Social Sciences, version 13, for WindowsSPSS, Chicago, IL). A P value less than .05 was con-idered statistically significant.

esults

No significant differences were found between thegroups concerning age, gender, weight, length of

Table 1. FROMME’S ORDINAL SCALE OFASSESSMENT OF SURGICAL FIELD AND FROMME’SMODIFIED SCALE

Level Aspect of Surgical Field

rdinal scale5 Massive uncontrollable bleeding4 Bleeding heavy but controllable that

significantly interferes with dissection3 Reasonable bleeding that moderately

compromises surgical dissection2 Moderate bleeding that does not

interfere with accurate dissection1 Bleeding so mild it is not even a surgical

nuisance0 No bleeding, virtually bloodless fieldodified scale3 Massive uncontrollable bleeding2 Bleeding with interference but accurate

dissection1 Bleeding without interference with

accurate dissection0 No bleeding

ossi et al. Desflurane Versus Sevoflurane to Reduce Blood Loss.Oral Maxillofac Surg 2010.

urgical procedures, Fromme’s modified level of sur-

gi

utdo

iacgrmttd

ltbrt

D

bgtrsr2

An

R J Oral

ROSSI ET AL 1009

ical field (Table 2), baseline MAP (Fig 1), or CI afterntubation.

The shorter surgical operation consisted of 70 min-tes for maxillary time and 15 minutes for mandibularime. All the other data collected during the proce-ures were reduced to match the time of the shorterperation.The MAP values were significantly greater (P � .05)

n desflurane group than in the sevoflurane group forll stages after 15 minutes from the start of the pro-edure (Fig 1A). The CI values were significantlyreater in the desflurane group than in the sevoflu-ane group during the first and second step of theaxillary time (Fig 1B). Sodium-nitroprusside to ob-

ain the surgical field target was used in 8 patients inhe sevoflurane group and only 2 patients in the

Table 2. PATIENT CHARACTERISTICS

Parameter Sevoflurane

Age (yr)Mean � SD 27 �Median (range) 27 (19-3

Weight (kg)Mean � SD 75 �Median (range) 78 (56-9

Tmx (min)Mean � SD 81 �Median (range) 80 (75-9

Tmd (min)Mean � SD 30 �Median (range) 30 (15-4

BL (ml·kg-1)Tmx

Mean � SD 7.4 �Median (range) 6.2 (5.0-

TmdMean � SD 2.2 �Median (range) 2.1 (0.8-

TotalMean � SD 9.6 �Median (range) 8.9 (6.3-

Total in Na-NP groupMean � SD 12.1 �Median (range) 10.1 (8.1.

Maxillary fieldMean � SD 1.8 �Median (range) 1.5 (1.0-

Mandibular fieldMean � SD 1.6 �Median (range) 1.0 (0.0-

Patients needing Na-NP (n) 8/10Gender (n)

Male 6Female 4

bbreviations: SD, standard deviation; Tmx, maxillary titroprusside.Field evaluated using Fromme’s modified scale of assessm

ossi et al. Desflurane Versus Sevoflurane to Reduce Blood Loss.

esflurane group (P � .01) (Table 2). The total blood f

oss in the desflurane group was lower (P � .05) thanhe blood loss in the sevoflurane group; the amount oflood loss in the 2 patients in the desflurane groupequiring sodium-nitroprusside was not different fromhat in the 8 patients in the sevoflurane group (Table 2).

iscussion

In this study, we obtained a greater reduction oflood loss and a better surgical field in the desfluraneroup than in the sevoflurane group. As reported inhe published data, desflurane seems to cause a lowereduction in cardiac output and a greater reduction inystemic vascular resistance index than does sevoflu-ane, although no studies have directly compared theinhaled agents.10-12 This point agrees with what we

p Desflurane Group P Value

24 � 4 NS23 (19-31)

76 � 9 NS77 (55-88)

78 � 9 NS75 (70-95)

36 � 11 NS35 (15-50)

4.8 � 1.4 .0084.4 (3.3-7.8)

2.4 � 1.7 NS1.7 (0.8-6.0)

7.2 � 2.8 .0386.6 (4.6-13.8)

12.4 � 0.7 NS12.4 (11-13.8)

1.5 � 0.2 NS1.0 (0.0-2.0)

1.5 � 0.2 NS1.0 (0.0-2.0)

2/10 .006

4 NS6

md, mandibular time; BL, blood loss; Na-NP, sodium-

f surgical field.

Maxillofac Surg 2010.

Grou

65)

100)

40)

125)

3.215.3)

0.83.5)

3.718.8)

2.9-18.8)

0.43.0)

0.22.0)

ime; T

ent o

ound in our study: the greater CI in the desflurane

gsvs

pbs

Fa llary tim

R J Oral

1010 DESFLURANE VERSUS SEVOFLURANE TO REDUCE BLOOD LOSS

roup at the same value of MAP in the first and secondtages of the surgical procedure resulted in a systemicascular resistance index greater than that in the

IGURE 1. A, MAP and B, CI trends in boxplots with medians, intend surgical steps. *P � .05 for Mann-Whitney U test. Tmx, maxi

ossi et al. Desflurane Versus Sevoflurane to Reduce Blood Loss.

evoflurane group. Thus, the craniofacial vascular s

lexus of the bone could be maintained dry from theeginning of the anesthesia. The myocardial depres-ion induced by sevoflurane caused congestion in the

les, and 5th to 95th percentiles in 2 groups during anesthesiologye; Tmd, mandibular time.

Maxillofac Surg 2010.

rquarti

urgical field. This point could explain the necessity

otctaccptldps

gwnimcgurT

cistptidCwofmvMdbimeacdrflflhcr

p

nbistcncrasvtprfivvhcbtslrtapbtwdtfi

snmrsgldoawtn

bcgg

ROSSI ET AL 1011

f administrating sodium-nitroprusside to induce con-rolled hypotension targeted to the level of the surgi-al field in the sevoflurane group. The differences inhe CI between the 2 groups can be explained by thebility of desflurane to mitigate Ca2� overload13 inardiovascular vessels. This pharmacologic actionould maintain greater cardiac inotropism and im-rove the reduction of differential arterial pressure ofhe splanchnic districts.5 Thus, drainage of the vascu-ar plexus of the bone, the first cause of blood lossuring maxillary and mandibular osteotomy, is im-roved, and the surgical field is acceptable to theurgeon.14

In the sevoflurane group, the finding that the sur-ical field was not good for the operating proceduresith an acceptable anesthesia plan required sodium-itroprusside administration to reduce blood loss and

mprove the surgical field. Sodium-nitroprusside ad-inistration induced MAP reduction and the CI in-

reased to same levels as those in the desfluraneroup. However, the total blood loss, and, in partic-lar, the blood loss in maxillary time, in the sevoflu-ane group was greater than in the desflurane group.his point could be related to some considerations.As reported by Degoute et al,15 remifentanil hydro-

hloride results in mild hypotension with notablentraoperative hemodynamic stability during stressfulurgical events. They suggested the use of the hypo-ensive action of remifentanil for mild controlled hy-otension and the exploitation of the synergistic ac-ion of remifentanil with hypnotic agents.15 Thenfusion rate of remifentanil used in our study wasetermined from the findings from Caverni et al.9

averni et al9 used a mean of 0.5 �g · kg�1 · min�1

ith an inhaled agent MAC of 0.5. This infusion ratef remifentanil can be considered as the median ef-ective dose for patients undergoing maxillary andandibular osteotomies. In our study, the end tidal

alue of the inhaled agents was maintained at 0.8AC, adjusted for age correction and bispectral in-ex. On the basis of bolograms built on synergismetween the potency of remifentanil and the MAC of

nhaled agents,16-18 the infusion rate of 0.5 �g · kg�1 ·in�1 for remifentanil could be considered the 95%

ffective dose when the end tidal value of the inhaledgents is 0.8 MAC. Generally, the synergistic pharma-ologic power of inhaled agents and opioids has airect action on the spinal A-� fibers, which are indi-ectly linked to the stress-induced hemodynamic re-ex. The synergistic power of remifentanil and des-urane together results in an improvement in theemodynamic stability that desflurane seems to exer-ise on the cardiovascular vessels,17 improving theeduction in blood loss.

The use of nitroprusside paradoxically could im-

rove blood loss. Nitroprusside produces direct ve- m

ous and arterial vasodilation with maintenance oflood flow adequate for all organs and improves the

ncrease in CI and reduces the systemic vascular re-istance index.19 Degoute et al15 have affirmed thathe hypotensive action of nitroprusside has directontrol on the vascular smooth muscle. The use ofitroprusside enhances catecholamine release, in-reasing the sympathetic tone and resulting in arte-iolar and precapillary vasoconstriction. Nitroprussidebolishes this reflex by its action on the vascularmooth muscle and maintaining the flow in vascularessels. However, the venous vasodilation increaseshe venous capacitance, congesting the venous bonelexus and promoting venous hemorrhage. Usingemifentanil as a “vasoactive drug” (in contrast to thendings from Degoute et al15) could reduce this reflexascular response, favoring a greater drainage of theascular plexus. Hersey et al4 found that nitroprussideas a more pronounced venous effect than does ni-ardipine. They have confirmed that blood loss inone surgery is highly dependent on venous conges-ion and that the venous action of nitroprusside re-ults in venous engorgement and an increase in bloodoss. In our study, the use of nitroprusside did noteduce blood loss. In addition, patients receiving ni-roprusside in the desflurane group had the samemount of blood loss as the patients receiving nitro-russide in the sevoflurane group. The difference inlood loss between the 2 groups in our study seemedo be related to the percentage of patients treatedith nitroprusside. What was not surprising about theifference (even if not statistically significant) be-ween the 2 groups regarding the level of the surgicaleld during the maxillary time.Our study had several potential weaknesses. The

tatistical sample size was small, and, thus, we couldot draw general conclusions. The use of the Frommeodified scale made the study easier but to the det-

iment of a reduction of the significance of the data,uch as the evaluation of the field level between the 2roups. The use of Fromme’s normal scale and aarger statistical sample size are needed to confirm theifference between the 2 groups regarding the levelf the surgical field. The interesting comparison of themount of blood loss for the patients treated with andithout nitroprusside in both the desflurane (2 pa-

ients) and sevoflurane (8 patients) groups includedumbers too low to demonstrate significance.However, despite these issues, in conclusion, we

elieve that desflurane at 0.8 MAC could be a goodhoice to decrease surgical bleeding, allowing forood hemodynamic stability, a reduction of the anal-esic and hypotensive drug requirement, and normal

etabolic status during 3 hours of anesthesia.

A

C

R

1

1

1

1

1

1

1

1

1

1

1012 DESFLURANE VERSUS SEVOFLURANE TO REDUCE BLOOD LOSS

cknowledgment

The authors wish to thank Patricia Cornwell (IIK Linguisticentre, Ancona, Italy) for her correction of the language in the text.

eferences1. Precious DS, Splinter W, Bosco D: Induced hypotensive anes-

thesia for adolescent surgery patients. J Oral Maxillofac Surg54:680, 1996

2. Enlund M, Ahlstedt B, Revends B, Krekmanov L, Ronquist G:Adverse effects on the brain in connection with isoflurane-induced hypotensive anesthesia. Acta Anesthesiol Scand 33:413, 1989

3. Tinker JK, Michenfelder JD: Sodium nitroprusside: Pharmacol-ogy, toxicology and therapeutics. Anesthesiology 45:340, 1976

4. Hersey SL, O’Dell NE, Lowe S, et al: Nicardipine versus nitro-prusside for controlled hypotension during spinal surgery inadolescents. Anesth Analg 84:1239, 1997

5. O’Riordan J, O’Beirne HA, Young Y, et al: Effect of desfluraneand isoflurane on splanchnic microcirculation during majorsurgery. Br J Anaesth 78:95, 1997

6. Lerou JGC: Nomogram to estimate age-related MAC. Br J An-aesth 93:288, 2004

7. McAtamney D, O’Hare R, Hughes D, et al: Evaluation ofremifentanil for control of haemodynamic response to trachealintubation. Anesthesia 53:1209, 1998

8. Munoz HR, Cortinez LI, Altermatt FR, et al: Remifentanil re-quirements during sevoflurane administration to block somaticand cardiovascular response to skin incision in children and

adults. Anesthesiology 97:1142, 2002

9. Caverni V, Rosa G, Pinto G, et al: Hypotensive anesthesia andrecovery of cognitive function in long-term craniofacial sur-gery. J Craniofac Surg 16:531, 2005

0. Weiskopf RB, Cahalan MK: Cardiovascular actions of desfluranein normocarbic volunteer. Anesth Analg 73:143, 1991

1. Weiskopf RB, Cahalan MK: Hemodynamic effects of desflu-rane/nitrous oxide anesthesia in volunteers. Anesth Analg 73:157, 1991

2. Rodig G, Keyl C, Kaluza M, et al: Effect of rapid increases ofdesflurane and sevoflurane to concentrations of 1.5 MAC onsystemic vascular resistance and catecholamine response dur-ing cardiopulmonary bypass. Anesthesiology 87:801, 1997

3. Guarracino F, Landoni G, Trirapepe L, et al: Myocardial damageprevented by volatile anesthetics: A multicenter randomizedcontrolled study. J Cardiothorac Vasc Anesth 20:477, 2006

4. Dolman RM, Bentley KC, Head TW, et al: The effect of hypo-tensive anesthesia on blood loss and operative time during LeFort I osteotomies. J Oral Maxillofac Surg 58:834, 2000

5. Degoute CS, Ray MJ, Manchon M, et al: Remifentanil andcontrolled hypotension; comparison with nitroprusside or es-molol during tympanoplasty. Can J Anaesth 48:20, 2001

6. Egan TD, Muir KT, Hermann DJ, et al: The electroencephalo-gram (EEG) and clinical measures of opioid potency: Definingthe EEG-clinical potency relationship (“fingerprint”) with ap-plication to remifentanil. Int J Pharm Med 15:11, 2001

7. Katoh T, Kobayashi S, Suzuki A, et al: The effect of fentanyl onsevoflurane requirements for somatic and sympathetic re-sponses to surgical incision. Anesthesiology 90:398, 1999

8. Sebel PS, Glass PSA, Fletcher JE, et al: Reduction of the MAC ofdesflurane with fentanyl. Anesthesiology 76:52, 1992

9. Friederich JA, Butterworth JF IV: Sodium nitroprusside:

Twenty years and counting. Anesth Analg 81:152, 1995