Maternal, fetal, and neonatal lidocaine levels followinglocal perineal infiltration

Elliot H. Philipson, M.D., Betty R. Kuhnert, Ph.D., and Carol D. Syracuse, B.S.

Cleveland, Ohio

Local infiltration of the perineum is a simple and commonly used technique for providing pain relief for

episiotomy. It has always been considered safe and effective because a small amount of local anesthetic

agent could be administered quickly and accurately to the perineum just prior to vaginal delivery and cord

clamping. Because of the short time interval between local infiltration and delivery, very little anestheticwas thought to reach the fetus. However, the maternal and neonatal disposition of lidocaine following local

perineal infiltration has not been well studied. The purpose of this study was to document placentaltransfer or nontransfer of lidocaine following local perineal infiltration. Fifteen normal parturient women atterm and their infants were studied. After local perineal infiltration, the concentrations of lidocaine and two

metabolites-monoethyl glycine xylidide and glycine xylidide-were determined in maternal plasma, in theumbilical cord vein at delivery, and in maternal and neonatal plasma or urine for 2 days post partum.

Lidocaine and its metabolites were quantitated by gas chromatography-mass spectrometry. Thepharmacologic results indicated the .following: First, lidocaine is detected in maternal plasma as early as 1

minute after injection, and peak plasma concentrations occur within 3 to 15 minutes. Second, there is rapid

placental transfer of lidocaine; the mean fetal/maternal ratio of 1.32 was significantly higher than that found

following epidural anesthesia. Third, lidocaine and its active metabolites persisted in neonatal urine for at

least 48 hours after delivery. This study suggests that local perineal infiltration with lidocaine for episiotomy

should be considered similar to any other anesthetic technique in that it may result in significant neonatal

drug exposure. (AM. J. OBSTET. GVNECOL. 149:403, 1984.)

Local infiltration of the perineum is a rapid, simplemethod of providing pain relief for episiotomy.' It is awidely used technique that involves small doses of alocal anesthetic agent and a short drug injection-to­delivery time. As a result, it might be anticipated thatvery little or no drug would reach the fetus. However,the disposition of anesthetic agents following localperineal infiltration has not been well documented de­spite many years of use in obstetrics. Nevertheless, it iswell documented that local anesthetic agents cross theplacenta when given by other routes of administra­tion, for example, epidural, paracervical, or pudendalblock.":" Therefore the purpose of this study was todescribe the maternal and neonatal disposition oflidocaine following local perineal infiltration.

Material and methods

Patients. Fifteen normal gravida women at termfrom the obstetric service at Cleveland Metropolitan

From the Department ofObstetricsand Gynecology and the PerinatalClinical Research Center, Cleveland Metropolitan General Hospi­tal/Case Western Reserve University.

Supported in part by United States Public Health Service PerinatalClinical Research Center Grant No. M01-RR00210.

Receivedfor publication July 14, 1983; revised August 30,1983;accepted December 7, 1983.

Reprint requests:Elliot H. Philipson, M.D., Department ofObstetricsand Gynecology, Cleveland Metropolitan General Hospital, 3395Scranton Road, Cleveland, OH 44109.

General/Highland View Hospital were studied. In­formed consent for sample collection was obtained inearly labor. Local infiltration of the perineum was ad­ministered as needed for episiotomy. No patients had ahistory of liver or kidney disease or of known drugabuse.

Procedure. Local perineal infiltration was performedwith a 22-gauge needle. Lidocaine (I % or 2%) withoutepinephrine was injected superficially into the area ofthe posterior fourchette of each patient during thecrowning phase of the second stage of labor. Lidocainewas injected as needed and the amount injected wasdetermined by the physician. The total dose oflidocaine(milligrams) and the time (minutes) from administra­tion to delivery (the drug-delivery interval) wererecorded.

Sample collection. Maternal blood samples (3 ml)

were collected from an indwelling cannula ina periph­eral arm vein. Maternal samples were collected beforeand then 1, 3, 5, 7, 10, 20, and 30 minutes afterperineal infiltration or until delivery, at which time thefinal sample was drawn. A maternal sample was ob­tained at delivery coincident with the damping of the

infant's cord. All samples were drawn into heparinanticoagulant tubes and immediately placed on ice.

The plasma was removed following centrifugation andfrozen until assayed by gas chromatography-massspectrometry. Neonatal blood samples were obtained

403

404 Philipson, Kuhnert, and Syracuse June 15, 1984Am. J. Obstet. Gynecol.

78

oo

Characteristic

23 ± 6

132

79 ± 3 (30-160)7.8 ± 7 (3-26)

39 ± I3347 ± 308

MothersAge (yr)*Parity

NulliparousMultiparous

NeonatesGestational age (wk)Birth weight (gm)Apgar score <7

I min5 min

SexMaleFemale

Intrapartum anesthesiaTotal dose (mg)Drug-to-Delivery interval (min)

Classification

Table I. Maternal, neonatal, and intrapartumanesthesia characteristics of the study group(N = 15)

2510 15 20TIME (min)

5

1000

...... • c.v .E

500.......Olc......Z0I-« Del.e::I-ZWoz 1000o

Fig. 1. Maternal plasma concentration of lidocaine after localperineal infiltration. C.V. = Cord vein.

*Mean ± SD with range in parentheses.

at birth from a doubly clamped section of umbilicalcord and at 24 and 48 hours of age by either heel stickor venipuncture. These samples were obtained andprocessed in the same manner as the maternal bloodsamples.

Maternal urine samples were collected continuouslyuntil 48 hours after delivery. Neonatal urine sampleswere collected in plastic urine collection bags (HollisterNewborn U-bag). Collections of four consecutive 12­hour samples were attempted from both mother andbaby, but neonatal collections were stopped if signs ofskin irritation occurred. All urine samples were frozenuntil analyzed.

Sample analysis. All plasma and urine samples werequantitated for lidocaine and its two pharmacologicallyactive metabolites-monoethylglycinexylidide and gly­cinexylidide with use of the method of Hignite et al.7 asreported previously." Analysis was performed on aHewlett-Packard 5995A quadrupole table-top massspectrometer interfaced to the gas chromatograph witha glass jet separator. The chromatograph was fittedwith a 1.3 m by 2 mm (inside diameter) acid-washed

dimethylchlorosilane treated glass coil packed with 3%SE-30: OV -17 (6: 1) coated on an 80- to l Otl-meshSupelcoport (Supelco, Bellefonte, Pennsylvania). Theinstrument conditions were as follows: carrier gas flowrate, 20 ml/min; injection port temperature, 250 0 C;oven temperature, programmed from 1760 C (0 min­utes) at 140 C/min to 220 0 C (0.5 minutes); and entirerun time, 3.6 minutes. The optics of the mass spec­trometer were optimized by autotuning at mass/charge

ratio (m/z) 100.

Standard curves were prepared and the sampleswere quantitated by use of the Hewlett-Packard soft­ware for automatic quantitation of selected ion-moni­toring data by area normalization on mlz 86 (forlidocaine), mlz 100 (for monoethylglycinexylidide),and mlz 114 (for glycinexylidide and the internal stan­dard, W-12174).8

Pharmacokinetic analysis. Semilogarithmic plotsshowing the appearance and disappearance of lido­caine from maternal plasma were obtained by plottingconcentrations of lidocaine in maternal plasma againstminutes after the injection of lidocaine. Fetal maternalratios were calculated by dividing the amount oflidocaine in the cord vein at birth by the amount in thematernal vein at birth. Linear plots showing the disap­pearance from urine of lidocaine and two active me­tabolites, monoethylglycinexlyidide and glycinexylidide,were obtained by plotting drug concentration in mi­crograms of drug excreted per milligram of creatinineagainst hours after delivery for both mother and neo­nate. Urine data are expressed in terms of creatininebecause of the inherent difficulty in obtaining completeurine samples from neonates. Creatinine was deter­mined by the picric acid method."

Statistical analysis. The statistical calculations weredone with a Hewlett-Packard 41C programmable cal­culator. Student's t test was used to test for differences

between calculated fetal/maternal ratios after localperineal infiltration and the fetal/maternal ratios ob­tained after lidocaine epidural anesthesia (N = 25)from a previous study. Linear regression was used totest for any relationships between fetal/maternal ratioand total dose of lidocaine administered, drug-deliveryinterval, peak maternal plasma concentration, umbili-

Volume 149Number 4

Lidocaine levels following perineal infiltration 405

1.91~y 2.18

7.0

(8)

6.0

~5.0

Z~<wa:0 4 0Q •

ed.::L

Zo~<3.0a:I-ZWoZo02.0

MOTHER

(9)

III LIDOCAINE

o MEG X

~ GX

NEONATE

(7)

24

(4)

48 12

TIME (HOURS)

(9)

2412

1.0

Fig. 2. Maternal and neonatal excretion of lidocaine and metabolitesfor 48 hours after localperinealinfiltration (mean ± SEM). MEGX = Monoethylglycinexylidide. GX = Glycinexylidide.

cal cord vein pH, length of second stage of labor, andother clinical characteristics of the study group. A pvalue of <0.05 was considered significant.

Results

Patient population. The maternal, neonatal, and in­trapartum anesthesia characteristics of the 15 studypatients are shown in Table I. The mothers were atterm and without any underlying obstetric problems.All neonates were term with Apgar scores of ?7 at 1and 5 minutes. The mean pH of the umbilical cord veinat delivery was 7.32 + 0.04. There were no signs oflidocaine toxicity in any of the neonates.

Plasma lidocaine levels. Appearance of lidocainewas rapid. As early as 1 minute after local perinealinfiltration, measurable levels of lidocaine (nanogramsper milliliter) were detected in maternal plasma. Peakconcentrations ranged from 60 to 2400 ng/ml and oc­curred from 3 to 12 minutes following the localperineal infiltration (Table II). A representative curve

of the appearance and disappearance of lidocaine inone patient who received 8 ml of 1% lidocaine (80 mg)for local perineal infiltration is shown in Fig. I.

Detectable amounts of lidocaine were found at deliv­ery in all maternal venous and cord vein samples, asshown in Table II. The highest concentrations oflidocaine in maternal venous and umbilical cord veinsamples were found to be 1474 and 1380 ng/ml, re­spectively. Lidocaine was not detectable in postpartumneonatal plasma.

Table II also shows the fetal/maternal ratio of

lidocaine at delivery. This ranged from 0.05 to 4.66with a mean of 1.32. In five of the 15 infants studied,the concentration of lidocaine at delivery was greater in

cord vein than in maternal vein. No correlation couldbe found between the fetal/maternal ratio and dosageof lidocaine, drug-delivery interval, peak maternal

plasma concentration, umbilical cord vein pH, or otherclinical characteristics of the study patients. However,the fetal/maternal ratio was found to be significantly

406 Philipson, Kuhnert, and Syracuse June 15, 1984Am. J. Obstet. Gynecol.

*Ratio of level in cord vein to level in maternal vein atdelivery (mean of individual ratios, not ratio of means).

Table II. Lidocaine concentration (nanogramsper milliliter) at delivery in maternal plasmaand umbilical cord vein after local perinealinfiltration (N = IS)

related (r = 0.73, P < 0.01) to the length of the second

stage of labor. The equation of the regression line wasy = -0.34 + 0.03x.

Urine drug levels. The patterns of urinary excretionand metabolism of lidocaine by the mothers and neo­nates are shown in Fig. 2. Lidocaine and monoethyl­glyeinexylidide were detectable in both maternal andneonatal urine for at least 48 hours post partum.Glycinexylidide was detectable in maternal urine butonly in trace quantities in neonatal urine. In the first 36hours post partum, more monoethylglycinexylididethan lidocaine was excreted in maternal urine. Inneonatal urine, monoethylglycinexylidide was nothigher than lidocaine until after the first 12 hours andthereafter stayed relatively constant while lidocaineexcretion decreased dramatically.

Comment

The purpose of this study was to describe the dispo­

sition of lidocaine in mother and neonate followinglocal perineal infiltration. The key finding of this study

indicates that lidocaine rapidly crosses the placenta andreaches the fetus in considerable amounts after localinfiltration of the perineum. This transplacental trans­fer of lidocaine occurs despite administration of rela­tively small doses of lidocaine and short drug-to­delivery intervals.

Of particular interest is the additional finding of an

elevated fetal/maternal ratio following local perinealinfiltration. The mean fetal/maternal ratio (1.32) fol­lowing local perineal infiltration was significantly

higher than the mean fetal/maternal ratio of lidocaine

reported after lumbar epidural anesthesia for vaginaldelivery (0.56) or cesarean section (0.66).2 The mean

fetal/maternal ratio after local perineal infiltration was

also higher than the fetal/maternal ratio of lidocaine

reported after paracervical block (0.58)5 and pudendalblock (0.4).6 Possible explanations for this high fetal/maternal ratio warrant further comment. One ex pla-

nation may be the anatomic proximity of the perineumto the fetus. Injection of a local anesthetic agent into ahighly vascular area may result in rapid drug distribu­tion to the fetus." However, it has been reported thatthe fetal/maternal ratio is not affected by route of ad­ministration or maternal blood levels."

The strong correlation between the fetal/maternal

ratio and the length of the second stage of labor sug­gests an alternative and more probable explanation.The high fetal/maternal ratio may be related to a lowerfetal tissue pH in the second stage of labor. The mean

fetal tissue pH has been shown to decrease linearlyfrom 7.30 to 7.20 after 55 minutes of the second stageof labor.!" This was found despite maintenance of amean umbilical cord vein pH of 7.34 at delivery. Alower fetal pH can increase the ionized form of localanesthetic agent in the fetus and may increase thefetal/maternal ratio." It is well known that elevatedconcentrations of local anesthetic agents occur as a re­sult of "ion trapping" in an acidotic environment. II, 12

Acidosis can also decrease protein binding of local an­esthetic agents, which may also affect the fetal/mater­nal ratio.!" Further investigation with tissue pH mea­surements may be helpful to verify this speculation.

Nevertheless, the pharmacologic results of this studyclearly demonstrate considerable placental transfer oflidocaine. The neonatal urinary excretion and meta­bolic pattern demonstrate the persistence of lidocaineand its pharmacologically active metabolites for at least48 hours post partum.

Since local infiltration of the perineum for episiot­omy is a widely accepted and practiced procedure,what clinical implications, if any, do the results of thisstudy have? These results suggest that even a simpleprocedure that appears innocuous may have unantici­pated results. For example, lidocaine toxicity withneonatal seizures can result from direct accidental in­jection of lidocaine into the fetal scalp." Although thisis unlikely, even appropriate perineal administration oflidocaine can have the following consequences: (1)

Rapid placental transfer of lidocaine to the fetus may

occur; (2) the concentration of lidocaine in umbilicalcord blood may be higher than concurrent concentra­tion in maternal blood; (3) the neonate may excretelidocaine and its pharmacologically active metabolitesfor as long as 48 hours after delivery. This study alsosuggests that it may be possible to minimize this fetal

exposure by use of alternative local anesthetic agents.

A rapidly metabolized, ester-linked agent such as2-chloroprocaine (Nesacaine) is currently being inves­

tigated in our laboratory. At present, local perineal

infiltration with lidocaine for episiotomy should beconsidered similar to any other anesthetic technique inthat it may result in significant neonatal exposure.

Range

60-2400

33-147445-1380

0.05-4.66

648 ± 666

Concentration

Mean ± SD

548 ± 468420 ± 4061.32 ± 1.46

Sample

Maternal plasma (peak concen-tration)

Maternal plasma (at delivery)Umbilical cord veinFetal/maternal ratio*

Volume 149Number 4

REFERENCES

1. Pritchard JA, MacDonald PC. Williams' obstetrics,16th ed. East Norwalk, Connecticut: Appleton-Century­Crofts, chap. 18, 1980:445.

2. Kuhnert BR, Knapp DB, Kuhnert PM, Prochaska AK.Maternal, fetal and neonatal metabolism of lidocaine.Clin Pharmacol Ther 1979;26:213.

3. Ralston DH, Shnider SM. The fetal and neonatal effectsof regional anesthesia in obstetrics. Anesthesiology 1978;48:34.

4. Covino BG. Comparative clinical pharmacology of localanesthetic agents. Anesthesiology 1971;35: 158.

5. Petrie RH, Paul WL, Miller FC, et al. Placental transfer oflidocaine following paracervical block. AM J OBSTETGYNECOL 1974;120:791.

6. Zador G, Lindmark G, Nilsson BA: Pudendal block innormal vaginal deliveries. Acta Obstet Gynecol Scand(Suppl) 1974;34:51.

7. Hignite CE, Tschanz C, Steiner J, et al. Quantitation oflidocaine and its deethylated metabolites in plasma andurine by gas chromatography-mass spectrometry. JChromatogr 1978;161:243.

Lidocaine levels follOWing perineal infiltration 407

8. Leibrand R, Jackson D. Automatic quantitation of GC/MSdata obtained in the selected ion mode. Hewlett PackardCo., Paper No. WE-27, 1977.

9. Henry RJ, Cannon DC, WinkeimanJW. Clinical chemis­try: Principles and techniques, 2nd ed. Hagerstown,Maryland: Harper & Row, 1974:545.

10. Weber T, Hahn-Pedersen S. Normal values for fetal scalptissue pH during labour. Br J Obstet Gynaecol 1979;86:728.

11. Biehl D, Shnider SM, Levinson G, Callender K. Placentaltransfer of lidocaine: Effects of fetal acidosis. Anes­thesiology 1978;48:409.

12. Brown WU, Bell GC, Alper MH. Acidosis, local anes­thetics, and the newborn. Obstet Gynecol 1976;48:27.

13. Tucker GT, Mather LE. Pharmacokinetics oflocal anaes­thetic agents. Br J Anaesth 1975;47:213.

14. Kim WY, Pomerance JJ, Miller AA. Lidocaine intoxica­tion in a newborn following local anesthesia for episi­otomy. Pediatrics 1979; 64:643.

Bound volumes available to subscribersBound volumes of the AMERICAN JOURNAL OF OBSTETRICS AND GYNECOLOGY are avail­

able to subscribers (only) for the 1984 issues from the Publisher, at a cost of $70.95($90.00 international) for Vol. 148 (January-April), Vol. 149 (May-August), and Vol. 150(September-December). Shipping charges are included. Each bound volume contains asubject and author index and all advertising is removed. Copies are shipped within 30days after publication of the last issue in the volume. The binding is durable buckram with

the JOURNAL name, volume number, and year stamped in gold on the spine. Paymentmustaccompany all orders. Contact The C. V. Mosby Company, Circulation Department, 11830Westline Industrial Drive, St. Louis, Missouri 63146, USA; phone (800) 325-4177, ext. 351.

Subscriptions must be in force to qualify. Bound volumes are not available in place ofa regular JOURNAL subscription.