Clinical pharmacology and efficacy of vancomycin in pediatric patients

PEDIATRIC PHARMACOLOGY AND THERAPEUTICS PaulS. Lietman, Editor

Clinical pharmacology and efficacy of vancomycin in pediatric patients The emergence of staphylococci resistant to multiple antibiotics and the lack of information on pharmacology, efficacy, and safety of vancomycin in pediatric patients prompted this study. In vitro susceptibility studies showed that 33 to 50% o f staphylococci tested were resistant to nafcillin, methicillin, cephalothin, and gentamicin, whereas none were resistant to vancomycin. Pharmacokinetic data o f vaneomycin were evaluated in 55 pediatric patients and was characterized by the two-compartment open-system kinetic model. Mean peak serum concentrations in it{rants and children after lO and 15 mg/kg doses ranged from 25.2 to 32.5 i~g/ml. The 10 mg/kg dose in newborn infants resulted in substantially lower peak values. Mean serum elimination phase half-h/'e vahtes correlated invet'sely with gestational and chronologic age and ranged from 2.2 to 9.8 hours. The CSF penetration in three itT~tnts with staphylococcal ventriculoperitoneal shunt injections ranged from 7 to 21%. Sixteen patients with staphylococcal diseases were successfitlly treated with vancomycin. In these patients peak serum vancomycin concentrations greater than 25 ixg/ml and trough concentrations less" than 12 #~g/ml produced satisfactory inhibitory and bactericidal titers. Laboratory studies and auditory fimclion tests did not reveal any drug-related abnormalities. Dosage schedules were formulated fi'om these data. Indications and precautions for vancomycin therapy in infants and children are presented.

Urs B. Sehaad, M.D.,* George H. MeCraeken, Jr., M.D., a n d J o h n D. N e l s o n , M . D . , D a l l a s , T e x a s

V A N C O MYC I N, an antibiotic derived from Streptomyces orientalis, was introduced over 20 years ago as the first bactericidal antibiotic for treatment of disease due to penicillin-resistant staphylococci. For several reasons it had a brief clinical life-span in pediatrics. It is not absorbed from the gastrointestinal tract and is very irritating to soft tissue and muscle; therefore, it must be given intravenously. In the late 1950s drugs were given intravenously by "slow" injection over a one- or twominute period; in the case of vancomycin, this frequently

From the Department of'Pediatrics, The Universi O' of Texas Health Science Center at Dallas, Southwestern Medical School. Supported by the John A. Hartford Foundation Inc., and by Lilly Research Laboratories. *Reprint address." Department of Pediatrics, The University of Texas Health Science Center at Dallas, 5323 Harry Hines Blvd, Dallas, TX 75235. Dr. Sehaad is an Infectious Disease Research Fellow supported by the Swiss" National Science Foundation.

0022-3476/80/010119+08500.80/0 9 1980 The C. V. Mosby Co.

caused severe phlebitis. It also had a reputation for causing nephrotoxicity and ototoxicity. When methicillin became available, vancomycin was abandoned. It reentered adult medicine in the late 1960s as an agent for treating bacterial endocarditis in patients allergic to penicillin. With the appearance in the late 1970s of methicillin-resistant Staphylococcus aureus and multiply-resistant Staphylococcus epidermidis strains as clinical problems, there has been renewed interest in vancomycin, since it is generally active against such strains. >~ Abbreviations used MIC: minimal inhibitory concentration MBC: minimal bactericidal concentration

~]

Much is known of the mechanism of action and pharmacokinetics of vancomycin from studies in adults. The agent is bactericidal against most gram-positive cocci and rods but is ineffective against most gram-negative bacteria. It exerts its bactericidal effect by interfering with the phospholipid cycle of cell-well synthesis s and it also The Journal o f P E D I A T R I C S Vol. 96, No. 1, pp. 119-126

119

1 20

Schaad, McCracken, and Nelson

The Journal of Pediatrics January 1980

Table I. Susceptibilities of 20 S. aureus and 6 S. epidermidis strains to vancomycin, nafcillin, methicillin, cephalothin, and gentamicin

Inhibitory~bactericidal concentrations (izg/ ml)* Organism (No. of strains)

Antimicrobial agent

Range MIC

Range MBC

] M1C.,-,

[

q

] MBC2:,

MICro [ MBC~o

MIC.~,

MBC~,,

w

S. aureus (20)

S. epidermidis (6)

Vancomycin Nafcillin Methicillin Cephalothin Gentamicin Vancomycin Nafcillin Methicillin Cephalothin Gentamicin

0.32-1.25 0.16-20 , 0,63-80 ~0,63-10 7<0,16-> 80 0.63-1.25 0.08-40 1.25-160 ~0.63-80 ~0.16-80

0.32-10 0.16-> 1.25-> 0.63-> 0.63-> 0.63-10 0.08- > 1.25-> 1.25-> 0.16->

40 160 320 80 40 160 320 80

0.32 0.16 1.25 ~0.63 0.32 0.63 0.16 1.25 ~0.63 ~0.16

0.63 0.32 2.5 2.5 1.25 0.63 1,25

0.63 0.32 1.25 ~0.63 2.5 0.63 0,16

1,25 1.25

1,25 ~0,63

0.32

~0.16

1.25 1.25 5 5 10 1.25 1.25 2.5 1.25 0.32

1.25

5

20 40 5 > 80 1.25 40 160 80 80

> 40 > 160 80 > 80 10 > 40 > 160 > 320 > 80

*MIC~,,,MBC_,;,MIC:.o,MBCy,,,MIC,,0, MBC~,,: Concentrations required to inhibit and kill 25, 50, or 90% respectively,of test strains.

alters plasma m e m b r a n e function '~ and inhibits ribonucleic acid synthesiss Vancomycin is not metabolized and is excreted primarily in the urine. It has a serum proteinbinding affinity of 50 to 60% and diffuses readily into body fluid compartments, including the cerebrospinal fluid in patients with inflamed meningesY- ~'The incidence of adverse reactions is reported to be low. 1-~, ' There is little information on the pharmacology, efficacy, and safety of vancomycin in infants and children. ~.... This study was prompted by the lack of data in pediatric patients and the recent development of nursery outbreaks of infections caused by multiply-resistant S. aureus and the problems posed by ventriculoperitoneal shunt infections due to resistant S. epidermidis strains. MATERIALS

AND

METHODS

Study patients. Pharmacokinetic studies were performed in 55 patients treated at Parkland Memorial Hospital or Children's Medical Center, Dallas. There were 21 neonates, 16 infants from one to 12 months of age, and 18 older children. The sex distribution was 31 males and 24 females. In 43 patients (21 neonates, 9 infants, 13 children) a single dose of vancomycin was substituted for the penicillin or penicillin analog that the patient was receiving for proved or suspected bacterial infection. The remaining 12 patients (7 infants and 5 children) were treated with a full course of vancomycin for staphylococcal disease. Written, informed parental consent was obtained for all study patients. Vancomycin was administered intravenously over a 30or 60-minute infusion period in dosages of 10 or 15 m g / k g body weight. The contents of the vancomycin vial (500 mg) were first dissolved in 10 ml of sterile water. This solution was further diluted ten- to twentyfold in the

infusion solution (physiologic saline or glucose 5%) to concentrations of 2.5 or 5 mg/ml. Serum samples for bioassay were collected either by finger-prick or from an indwelling heparin lock at 0 (end of infusion), 0,5, 1.0, 2,0, 4.0, and 6.0 hours after the first dose. In seven premature infants serum samples were also obtained at 9.0 and 12.0 hours after the dose. In 11 patients treated for staphylococcal infection, serum samples for assay were obtained after the first or second dose and on several occasions during therapy. Vancomycin concentrations were measured in 12 cerebrospinal fluid samples from three patients, in six stool samples from five patients treated intravenously, and in five stool samples from two patients given vancomycin by mouth, Adverse oi toxic reactions to vancomycin were assessed by clinical observation and by obtaining laboratory studies (complete blood cell count, urinalysis, SCOOT, and B U N a n d / o r creatinine) and auditory function tests (impedance audiometry, pure tone audiometry or brainstem evoked response) before and after treatment. Clinical and bacteriologic responses to therapy were also evaluated. Vaneomycin assay. Serum, CSF, and stool specimens were stored frozen at - 2 0 ~ until assayed within three days of collection. Vancomycin concentrations were measured by a microbioassay technique using Bacillus subtilis (Difco Laboratories, 0453-36) as the test organism. I'- '" The lowest detectable concentration by this technique was 0.9/,g/ml. Standards and, when necessary, samples were diluted in 1% phosphate buffer (pH 6.0) after it had been established that vancomycin diffusion in agar was the same whether dilutions were made in pooled human plasma, human CSF. or 1% phosphate buffer. If the patient received a penicillin before administra-

Volume 96 Number 1

Clinical pharmacology and efficacy of vancomvcin

12 1

T a b l e 1I. Characteristics and observed serum concentrations of vancomycin in infants and children i

Patient

NO.

categop~v patients Newborn infants

7

(kg)

(rnin)

12 4

4.3 mo

5.2

15/60

5

3 11/2 yr

15.5

15/60

7

5 7/12 yr

20.0

10/60

6

7 7/12 yr

26.7

10/30

7

Children

tional age) 3.3 days (32 wk) 4.7 days (34 wk) 2.6 days (40 wk) 3.1 mo

7

Infants

Average age Average (mg/kg)/time Dose I (average gesta- weight (of infusion) ~

1.23

10/30

1.57

15/30

3.07

15/30

4.9

10/60

Serum concentrations (izg/ml) at time (hrs) after inJhsion 0

t

0

16.8 _+ 1.3" (11.0-20.5) 25.2 • 2.1 (17.1-32.0) 29.8 + 1.4 (24.4-34.0) 26.1 • 1.2 (19.7-32.1) 28.0 • 1.! (25.7-30.5 32.5 • 2.0 (28.0-40.2) 27.2 • 1.2 (23.2-32.8) 30.5 • 3.1 (23.5-42.0)

.

5

13.3 • 1.1 (10.3-17.9) 19.7 • 1.6 (14.1-25.1) 22.8 • 1.5 (19.0-29.0) 17.5 • 1.4 (10.9-24.4) 18.5 • 1.9 (15.5-23.9) 17.7 • 1.0 (15.0-19.7) 16.9 • 1.2 (13.9-23.3) 18.2 J- 1.9 (13.4-25.6)

1

1

1

12.8 • 1.1 (9.9-17.8) 18.4 • 1.5 (12.8-21.6) 19.5 • t.1 (16.2-23.3) 14.6 _+ 1.2 (9.6-20.6) 15.6 + 1.4 (13.4-19.5) 13.6 • 0.9 (10.6-15.6) 12.4 • 0.6 (10.1-15.2) 14.4 _+ 1.8 (10.6-21,0)

2

1

4

11.5 • 0.8 (9.2-14.6) 17.5 • 1.3 (12.4-19.6) 17.7 + 1.1 (13.9-20.0)

6

10.2 • 0.8 (8.0-13.4) 13.3 _+ 1.2 (8.6-15.0) 14.2 • 1.2 (9.6-18.2)

9.0 • 0.7 (7.7-11.6) 10.4 • 0.6 (7.3-12.9) 11.5 + 0.8 (7.8-14.4) 11.5 • 1.1 8.3 • 1.1 5.9 • 0.8 (6.2-16.2) (3.8-13.9) (2.4-9.8) 11.1 _ 2.0 6.9 • 2.1 5.2 ~- 1.5 (7.6-16.5) (3.8-12.5) (2.9-9.2) 9.5 _+ 0.6 5.1 • 0.5 3.0 • 0.3 (7.5-10.8) (3.8-6.2) (2.3-3.9) 7.6 + 0.6 4.1 • 0.4 2.6 • 0.4 (6.1-10.8) (2.8-5.9) (1.3-3.9) 8.9 • 1.0 4.9 • 0.6 2.6 _+ 0.2 (5.9-12.8) (3.1-6.8) (2.1-3.4)

*Mean value • standard error of mean. Figures in parentheses represent r a n g e of observed values.

tion of vancomycin, the penicillin was inactivated by incubating the specimen for 15 minutes with 50,000 U / m l of penicillinase (Difco Laboratories, 0345-62) in 1% phosphate buffer. When patients received concomitant aminoglycoside therapy, 3% sodium chloride was added to the seeded agar to eliminate aminoglycoside activity. These two procedures did not affect the measurement of vancomycin. Analysis of ten vancomycin reference curves demonstrated an average accuracy for the bioassay of _+ 4.8% (range _+ 3.5 to 6.7%) for the standard concentrations from 1.25 to 30 #g/ml. P h a r m a c o k i n e t i c d e t e r m i n a t i o n s . Serum concentrationtime curves were analyzed by the N O N L I N least squares regression computer program. '~ The curves were best described by the two-compartment open-system kinetic model. 18-2o

was taken as the lowest concentration of drug producing 99.9% bacterial killing: this was determined by quantitative subcultures from each clear tube inoculated onto 5% sheep blood agar plates. S e r u m titers. Serum inhibitory' and bactericidal titers against the individual pathogens causing disease in 20 children were determined by microtiter using serial twofold dilutions of serum in Mueller-Hinton broth. ~1The 20 patients were treated with vancomycin for proved staphylococcal disease (15 S. attreus and 5 S. epidermidis). Serum specimens and pathogens from four of these patients were provided by Robert Hall, M.D., Children's Mercy Hospital, Kansas City, Missouri.

S u s c e p t i b i l i t y studies. In vitro susceptibility testing was performed with 20 S. aureus strains and 6 S. epidermid& strains isolated from blood, CSF, or pus from 22 children and four adults. An inoculum of 5 • 10:' to 1 x l0 G colony-forming units per milliliter was added to MuellerHinton broth containing serial two-fold dilutions of the following antimicrobial agents: Vancomycin (80 to 0.16 ptg/ml), nafcillin (40 to 0.08 /zg/ml), methicillin (160 to 0.32 #g/ml), cephalothin (320 to 0.63/~g/ml), and gentamicin (80 to 0.16 ptg/ml). After 18 hours of incubation at 37~ the minimal inhibitory concentration was determined as the smallest amount of antibiotic inhibiting visible growth. The minimal bactericidal concentration

nafcillin, methicillin, cephalothin, and gentamicin are presented in Table I. The activity was comparable for the five antimicrobial agents regarding concentrations required to inhibit and kill 25 and 50% of test strains. However, the MIC,,, and MBC~, values showed that 33 to 50% of these staphylococci were resistant in vitro to concentrations of nafcillin, methicillin, cephalothin, and gentamicin achieved in serum with usual dosages. For all 26 strains the MIC of vancomycin was from 0.32 to 1.25 /tg/ml and the MBC 0.32 to 10/zg/ml. (Two S. aureus and o n e S. epiderrnidis strain required 10 /~g/ml for 99.9% killing.) S e r u m c o n c e n t r a t i o n s . The 55 pediatric patients in

RESULTS Susceptibility

studies. The susceptibilities of 20 S.

aureus and six S. epidermidis strains to vancomycin,

Schaad, McCracken, and Nelson

122

The Journal of Pediatrics January 1980

10 rng/kg IN 7 PREMATURE INFANTS

50

,a~ L~ =~'k~

~

,~k-x~

(D~ " ~ (.~

~ ~ 2[~

lm

15 mg/kg IN 7 PREMATURE INFANTS

x

AGE - 5.3 days GA -3Z wks

~ ~

BSA-O.~2r~ 2

~ .

15 mg/kg IN 7 FULLTERM INFANTS

-',

4.7 days~ ~ 34 wks ,~

0.~4~ 2

..l

2.6 days 4O wks

,~ \ J

Omm 2

i

T'l/2a - 0.'15hr T~/2/9 ~ 9.8 hrs AU'C--~92 ~ug/rnl.hrs Vr,(area)-756 ml/kg " C I p ' - 1 5 I'nl/min/'l.75m'

~"x~,l ~:~i ~,1 ~'~[ ~I

0.05 hr 5.9 hrs "18"1 #g/ml'hrs 706 ml/kg '27 ml/min/'l."f3m 2

x~ ,~ ~ ~ ~

0.26 hr 6.7 hrs 2"1"1 #ug/ml.hrs 690 ml/k 9 :50 ml/rnin/t.75m 2

Io Observed mean values l

~ C~tculated mean valuesl

N 0

2

4

6

8

10

42

0

i

2

3

4

5

6

0

~

2

3

4

5

6

HOURS A F T E R INFUSION

Fig. 1. Population characteristics, serum concentration-time curves, and pharmacokinetic values for vancomycin in 21 newborn infants. GA = Gestational age, WT = weight, BSA = body surface area; Tl/aa = alpha half-life; T]/=fi = beta half-life; AUC = area-under-the-serum-concentration curve; VD = volume of distribution; Clp = plasma clearance. whom serum concentration-time curves were determined were divided into eight categories according to age, dosage, and duration of intravenous infusion, Table II and Figs. 1 to 3 display the characteristics of the different patient categories and the observed serum concentrations. In every patient the pea!< serum concentration occurred at Completion of the infusion (time 0). Vancomycin was infused over 30 minutes in the 21 newborn infants and was well tolerated. At the beginning of the study vancomycin was given as a 30-minute infusion to six older children. Four of these six patients developed an erythematous, macular rash near completion of the infusion (see Safety and Efficacy Section). Thereafter, all infants 9and childi'en received vancomycin over a 60-minute period. Newborn i'nfants: In seven premature infants with an average gestational age of 32 weeks and an average Weight of 1.23 kg, a 10 mg/kg vancomyein dose resulted in peak serum concentrations from 11,0 to 20.5/~g/ml (mean 16.8 /~g/ml). The values 12 hours later were from 5.2 to 6.5 /~g/ml (mean 5.7 ~g/ml). 'A 15 mg/kg dose given to seven premature infants of similar gestational age and weight and to seven term newborn infants with an average weight of 3.07 kg produced substantially highe ~"peak values of 17. l to 32.0 ~g/ml' (mean 25.2 ~g/ml) and 24.4 to 34.0 ~g/ml (mea n 29.8/~g/ml), respectb?ely. I,nfants~ Twelve infants from one to 12 months of ag e were given 10 mg/kg and four infants received a dose of ,15 mg/kg intravenously as a 60-minute infusion. Peak

serum concentrations of vancomycin were only slightly higher with the larger dosage, Children. The peak serum levels in five children (average age 311/1,_,years) after a 15 mg/kg dose given over 60 minutes were higher (28,0 to 40.2 /~g/ml, mean 32.5 ~g/ml) than those in seven children (average age 57/1~ years) given a 10 mg/kg dose infused over 60 minutes (23.2 to 32.8/~g/ml, mean 27.2/xg/ml), but concentrations at the other time periods were similar. Thirty-nine serum specimens from 11 patients were obtained on the fifth to twenthieth day of vancomycin therapy and the serum concentrations in these samples were compared to those obtained at identical times after the dose given on the first day. The mean values after the initial dose and after later doses, respectively, were as follows at the indicated times: 32.0 and 30.6 /xg/ml (0 hour), 17.5 and 17.9 ~g/ml (0.5 hours), 18.4 and 18.9 /~g/ml (1 hour), 10.9 and 11.6 txg/ml (4 hours), and 7.8 and 8.5 /~g/ml (6 hours). Thus, there was no evidence of accumulation with repeated dosing. Pharmacokinetic calculations. Pharmacokinetic data based on the mean serum concentration-time curve results are presented for the eight patient categories in Figs. 1 to 3. The serum concentration-time curves were best adapted to the two-compartment open-system pharmacokinetic model. The correlation coefficients for the fit of all eight curves were from 0.997 to 1.00. The alpha, or distribution phase, half-life values were between 0.05 and 0.50 hours, representing the initial rapid fall in serum concentrations. The shortest Tit a values

Volume 96 Number 1

Clinical pharmucoloRy and el~'caql' of vanco~Tvcin

10 m g / k g IN 12 INFANTS

50 40

15 m g / k g IN 4 INFANIS

AGE 3.'1 rnos WT 4.9 kg BSA 0.28 m z

30

123

4.5 rnos 5.2 kg 0.30 rn2

N \\\NN\N

_~ 2O

E

~\\\\'R

%. (:~

.",,\\\\N~ ~xxx~x ~x\\\xx

g to

,\.'%.:..x." ~\xxx\x

VD15 (~!~ !2i~ 7? /i~~

g

.x\\\\\,

CIp -

50

ml/rnin/i.75m 2

J , Calculated mean values

0

hrs

4.i 93 964 841

hr

I

I

I

I

1

I

"1

2

3

4

5

6

Observed

J_

)Jg/ml.hrs ml/kg ml/min/t.73m 2

values}

mean

~ _ _ L 0

"1

2

I__- L

|

___1

3

5

6

4

HOURS AFTER INFUSION Fig. 2. Population characteristics, serum concentration-time curves, and pharmacokinetic values for vancomycin in 16 infants one to 12 months of age. See Fig. 1 for abbreviations. 15 m g / k g IN 5 CHILDREN

10 m g / k g IN 7 CHILDREN

AGE - 5tl/~z yrs WT -'15,5 kg BSA - 0,65 mz

4o 3o

10 m g / k g IN 6 CHILDREN

5 ?/t2 yrs 20,0 kg 0.78 m2

R'..".~ ~.,"N~[k',,'~

7 7/~2 yrs 26.7 kg

2o v

~o

AUC- 65 #g/ml'hrs Vo(areQ)-8"18 ml/kg CIp-t65 ml/min/'L73m z

U')

o

~

2

hrs

"~ J-

I

I

I

4

5

6

72! k\\\\\Xl

m!',"!

....

2

T

ml/rnin/i]Sm" I

0

T

"I

L

2

HOURS A F T E R

l

I

I

..L

'1'

3

4

5

6

~

0

I

I __.~l__L

-1

2

._

5

4

~

....

5

~J

6

INFUSION

Fig. 3. Population characteristics, serum concentration-time curves and pharmacokinetic values for vancomycin in 18 children 13/,~to 11'A._,years of age. See Fig. 1 for abbreviations.

were e n c o u n t e r e d in n e w b o r n infants. The average beta, or elimination phase, half-life values were from 5.9 to 9.8 hours in n e w b o r n infants. 4.1 hours in older infants, and from 2.2 to 3.0 hours in children. The a r e a - u n d e r t h e - s e r u m - c o n c e n t r a t i o n curve was inversely p r o p o r t i o n a l to chronologic age. The values were approximately 200 /~g/ml. hours in neonates. 100 ffg/ml" hours in infants, a n d 60 ffg/ml 9 h o u r in children. The area-derived volume

of distribution showed relatively little variation a m o n g the different patient categories a n d ranged from 538 to 964 ml/kg. Plasma clearance rates expressed in relation to body surface area correlated directly with chronologic age. The calculated rates were from 15 to 30 m l / m i n u t e / 1.73 m ~ in neonates, from 50 to 8l m l / m i n u t e / 1 . 7 3 m ~ in infants, and from 131 to 163 m l / m i n u t e / l . 7 3 m-' in children.

124

The Journal of Pediatrics January 1980

Schaad, McCracken, and Nelson

SERUM ANTISTAPHYLOCOCCAL ACTIVITY IN 2 0 C H I L D R E N T R E A T E D W I T H V A N C O M Y C I N `1:64 1:32

cO

rr

i I

~x

`1:16 `1:8

I.d I--

i

`1:4 1.1.1

"1:2 o Bacteriostatic A Bactericidal

~,

,~4,2 __~.~l___s

5

I

"10

55

20

~

25

I

I

I

50

55

40

45

SERUM VANCOMYCIN CONCENTRATION (,,ug/ml)

Fig. 4. Serum bacteriostatic and bactericidal titers in relation to vanc0mycin concentrations. *S. aureus strain demonstrating tolerance to vancomycin (MIC, 0.63 /tg/ml; MBC, 40/~g/ml).

CSF and fecal concentrations. Twelve CSF specimens from three vancomycin-treated infants with staphylococcal ventriculoperitoneal shunt infections were available for assay. The CSF samples were obtained on the third to eleventh day of therapy. The specimens contained from 15 to 531 WBC/mm :~and the protein concentrations were from 22 to 265 mg/dl. The concentrations of vancomycin were from 1.2 to 4.8 /~g/ml (mean 3.1 ~tg/ml) and the penetration into CSF, expressed as a percentage of the concurrent serum concentration, ranged from 7 to 21% (mean, 14%). The concentrations of drug in six fecal samples from five intravenously treated patients were from 4.1 to 35.8 /xg/gm ",vet stool (mean 12.5 ~g/gm). Two infants were treated with vancomycin given orally (15 to 20 m g / k g / d a y ) for S, aureus enterocolitis. The concentrations in stool were from 85 to 540/.~g/ml liquid stool in five samples (mean 350 /~g/ml). Serum specimens obtained during oral therapy contained 0.9 and 1.05 /~g/ml. Serum titers. The relation between antistaphylococcal activity and vancomycin concentration in serum is shown in Fig 4. The two specimens containing < 1:2 bactericidal activity were obtained from an immunologically compromised infant treated for recurrent S. attreus sepsis. At the time that these titers were determined the S. aureus strain (phage type 53) showed tolerance to vancomycin (MIC, 0.63/~g/ml: MBC, 40 ~g/ml). The tolerance disappeared after storage of the organism on a blood agar plate

at 4~ (The high MIC value was not recorded in the susceptibility studies (Table I) that were performed at a later date.) With the exception of one of these two samples, a bactericidal titer of 1:8 or greater against the individual staphylococci was observed when the vancomycin concentration was over 12 ~g/ml. Bacteriostatic activity of 1:8 or greater was noted in all but one specimen. This single exception was from an infant with S. epidermidis sepsis in whom the vancomycin level was 8.4 p,g/ml, bacteriostatic titer was 1:4, and bactericidal titer was 1:2. The S. epidermidis strain had an MIC of 2.5 ~g/ml and an MBC of 5.0 #g/ml. Safety and efficacy. Four of the six children who received a vancomycin infusion in 30 minutes or less developed an erythema multiforme-like reaction associated with intense pruritus during the last ten minutes of the infusion. The distribution of the rash was unique in that it involved the face, neck, upper trunk and back, and upper arms with sparing of the rest of the body. No systemic manifestations were observed and the rash cleared in three to five hours. In the remaining patients the vancomycin infusions, administered over 30 minutes in newborn infants and over 60 minutes in older infants and children, were well tolerated; pain or discomfort at the infusion site was not encountered. In the 16 patients treated for staphylococcal illnesses, no adverse or toxic reactions to vancomycin were detected. Laboratory studies did not reveal abnormalities

Volume 96 Number l

Clinical pharmacology and effica~, of vancoo(vcin

of hematopoiesis or of renal and hepatic function. Auditory function tests showed no evidence of drug-related ototoxicity. Vancomycin appeared to be effective in the treatment of staphylococcal diseases in these 16 infants and children. The diseases included sepsis (3), osteomyelitis (3), ventriculoperitoneal shunt infections (3), and cutaneous and muscle infections (7). In most patients, treatment was changed to an orally administered antistaphylococcal agent from seven to 20 days after starting vancomycin; however, during the period of vancomycin therapy all patients showed clinical improvement as evidenced by diminution of the signs and symptoms of their staphylococcal infection, reduced fever, and sterilization of blood or cerebrospinal fluid cultures.

19.4%-'J For comparison, the penetration of nafcillin into CSF of three adult patients with meningeal inflammation similar to that in our patients was 4.8 to 21%/~ The concentrations of vancomycin in feces of five children with intact intestinal function treated intravenously exceeded the MIC values (0.2 to 0.4 /,g/ml) of Clostridium difficile, the probable etiologic agent of pseudomembranous enterocolitis in man. z:' The fecal concentrations in the two infants who received vancomycin orally were considerably lower than those ( > 1,000 /xg/ml) measured in adults after oral therapy.'-' ...... Our data suggest that both oral and intravenous vancomycin therapy should produce fecal levels that exceed the MBC values for most, if not all, Stapt~vlococcus aureus and Clostridium diJflcile strains, In the present study, no staphylococcal strain tested was resistant in vitro to vancomycin. Tolerance-'; was found in the S. aureus from one infant with sepsis. The organism had an MIC of 0.63 and MBC of 40 /~g/ml on initial microtiter broth technique; this could not be reproduced after storage of the organism for several weeks. The failure to demonstrate tolerance in other strains was probably due to storage of the organisms on blood agar plates at 4~ as reported by Bradley et al. -'~ Peak serum concentrations of vancomycin greater than 25/,g/ml produced a median inhibitory titer of 1:32 and a median bactericidal titer of 1: 16. Trough concentrations of less than 12 /,g/ml vancomycin resulted in median inhibitory and bactericidal titers of 1:8 and 1:4, respec-tively. This degree of activity should be satisfactory for management of staphylococcal diseases. Sixteen children have been treated with vancomycin for staphylococcal diseases. No neonates have been treated to date. On the basis of this limited experience and historical evidence in adults, vancomycin appears to be a safe and efficacious drug in the management of staphylococcal diseases in pediatric patients. No hematologic, renal. hepatic, or otologic abnormalities were observed. Adequate dilution and infusion over one hour reduces the likelihood of the erythema-multiforme-like reaction noted in some patientsY 4. ~' Implications. On the basis of these clinical and laboratory data, the following dosage schedules for vancomycin in pediatric patients are suggested: (1) 15 mg/kg doses.. given intravenously over 30 or 60 minutes to newborn infants every 12 hours during the first week of life (30 mg/kg/day) and every 8 hours in those 8 to 30 days of age (45 mg/kg/day): (2) 10 mg/kg doses given intravenously over 60 minutes every 6 hours (40 mg/kg/day) to older infants and children: (3) 15 mg/kg doses given intravenously over 60 minutes every 6 hours (60 mg/kg/day) to infants and children with staphylococcal central nervous

DISCUSSION The serum concentration-time curves for the eight patient groups declined in a biphasic manner and were best adapted to the two-compartment open-system kinetic model. '-~-~"The excellent fit between the values calculated by the two-compartment computer program '~ and the observed serum concentrations was expressed by correlation coeff• of 0.997 to 1.00. To our knowledge, two-compartment model pharmacokinetics have not been previously applied to vancomycin in pediatric patients. Preliminary data in adults have suggested a decline in the serum concentration time-curves that would be best defined by a two or three compartment pharmacokinetic m o d e l / _~2The beta or elimination phase half-life correlated inversely with chronologic or gestational age, weight and body surface area. Half-life times of two to three hours in children were shorter than those (four to six hours) reported in adults.' :~-e-' Moellering~ recently reported half-life values from five to 11 hours in four adults. A possible explanation for the longer half-life times in adults is the larger apparent volume of drug distribution or reduced drug elimination, or both. The volume of distribution in four adults reported by Moellering" ranged from 490 to 1,250 ml/kg, compared to 538 to 818 ml/kg in children. Plasma clearances of vancomycin in adults were 1.09 to 1.37 ml/minute/kg, compared to 2.78 to 3.97 ml/minute/kg in our children. The mean CSF penetration of 14% was remarkable in view of the low-grade meningeal inflammation in the patients with ventricutoperitoneal shunt infections. The three infants were clinically and bacteriologically cured: in two patients sterilization of CSF occurred before the shunt was revised. There are no previously published data on penetration of vancomycin into CSF in human beings. In a rabbit model of Staphylococcus aureus meningitis, the CSF penetration of vancomycin ranged from 4.5 to

12 5

1 26

Schaad, McCracken, and Nelson

system infections. T h e larger dosage might also be considered for patients with o v e r w h e l m i n g staphylococcal sepsis in the absence of central nervous system involvement. M e a s u r e m e n t of serum concentrations a n d of serum bactericidal titers are r e c o m m e n d e d in order to ascertain that peak v a n c o m y c i n levels are in the therapeutic and safe range of 25 to 4 0 / ~ g / m l and that peak bactericidal titers are at least 1:8. In children with i m p a i r e d renal function, dosage and intervals of a d m i n i s t r a t i o n are best d e t e r m i n e d by m e a s u r e m e n t of serum levels. In cases of long-term t r e a t m e n t a n d in any patient with k n o w n n e p h r o p a t h y , auditory a n d renal function should be monitored. Indications for v a n c o m y c i n use in pediatric patients at the present time include: infections caused by Staphylococcus aureus, Staphylococcus epidermidis, a n d Streptococcus pneumoniae strains multiply resistant to penicillins, cephalosporins, and aminoglycosides. The authors gratefully acknowledge the assistance of the following individuals: Dr. Craig Brater of the Department of Pharmacology assisted in the analysis of the pharmacokinetic data. Norma Threlkeld, R.N., and Helen Kusmiesz, R.N., coordinated the clinical studies and Marion Thomas, B.A., performed the assays.

The Journal of Pediatrics January 1980

12.

13.

14. 15.

16. 17.

18.

19. 20. 21.

REFERENCES

1. Cook FV, and Farrar WE: Vancomycin revisited, Ann Intern Med 88:813, 1978. 2. Geraci JE: Vancomycin, Mayo Clin Proc 52:631, 1977. 3. Alexander MR: A review of vancomycin after 15 years of use, Drug Intelligence and Clinical Pharmacy 8:520, 1974. 4. Esposito AL, and Gleckman RA: Vancomycin: A second look, JAMA 238:1756, 1977. 5. Anderson JS, Matsuhashi M, Haskin MA, and Strominger JL: Lipidphosphoacetylmuramyl-pentapeptide and lipidphosphodisaccharide-pentapeptide: Presumed membrane transport intermediates in cell wall synthesis, Proc Natl Acad Sci USA 53:881, 1965. 6. Jordan DC, and Mallory HDC: Site of action of vancomycin on Staphylococcus aureus, Antimicrob Ag Chemother 4:489~ 1964. 7. Jordan DC, and lnniss WE: Selective inhibition of ribonucleic acid synthesis in Staphylococcus aureus by vancomycin, Nature 184:1894, 1959. 8. Moellering RC: Pharmacokinetics ofvancomycin in normal subjects and patients with reduced renal function, Proc Syrup on Vancomycin. San Francisco, 1978 (in press). 9. Riley HD Jr: Vancomycin and novobiocin, Med Clin North Am 54:1277, 1970. 10. Spears RL, and Koch R: The use of vancomycin in pediatrics, Antibiotics Annual 1959-1960, pp 798-803. 11. Riley HD Jr, and Ryan N J: Treatment of severe staphylo-

22.

23.

24.

25.

26.

27.

28.

coccal infections in infancy and childhood with vancomycin, Antibiotics Annual 1959-1960, pp 908-916. George RM, Cochran CP, and Wheeler WE: Epidemic mehin~tis of the newborn caused by flavobacteria. I1. Clinical manifestations and treatment, Am J Dis Child 101:296, t961. Plotkin SA, and McKitrick JC: Nosocomial meningitis of the newborn caused by a flavobacterium, JAMA 198:194, 1966. Hawley HB, and Gump DW: Vancomycin therapy of bacterial meningitis, Am J Dis Child 126:261, 1973. Sabath LD, Casey JI, Ruch PA, Stumpf LL, and Finland M: Rapid microassay of gentamicin, kanamycin, neomycin, streptomycin, and vancomycin in serum or plasma, J Lab Clin Med 78:457, 1971. Walker CA, and Kopp B: Sensitive bioassay for vancomycin, Antimicrob Ag Chemother 13:30, 1978. Metzer CM: 1969. NONLIN: A computer program for parameter estimation of nonlinear situations, Technical Report No. 7292/69/7292/005, The Upjohn Co., Kalamazoo, Mich. Loughnan PM, Sitar DS, Ogilvie RI, and Neims AH: The two-compartment open-system kinetic model: A review of its clinical implications and applications, J PED1ATR88:869, 1976. Greenblatt DJ. and Kock-Weser J: Clinical pharmacokinetics (first of two parts), N Engl J Med 293:702, 1975. Greenblatt D J, and Kock-Weser J: Clinical pharmacokinetics (second of two parts), N Engl J Med 293:964, 1975. Nelson JD, Howard JB, and Shelton S: Oral antibiotic therapy for skeletal infections of children. I. Antibiotic concentrations in suppurative synovial fluid, J PEDIATR 92"131, 1978. Quintiliani R: The use of vancomycin in the prophylaxis and treatment of anephric patients undergoing dialysis, Proc Symp on Vancomycin, San Francisco, 1978 (in press). Murray T, Strausbaugh L, and Sande M: Comparative penetration of antibiotics into the CSF of rabbits with experimental staphylococcal meningitis. Abstract No. 246A. In Program and Abstracts, 16th Interscience Conference on Antimicrobial Agents and Chemotherapy, Chicago, illinois, American Society for Microbiology, 1976. Kane JG, Parker RH, Jordan GW and Hoeprich PD: Nafcillin concentration in cerebrospinal fluid during treatment of staphylococcal infections. Ann Intern Med 87:309, 1977. Fekety R: The use ofvancomycin in treatment of antibioticassociated enterocolitis. Proc Syrup on Vancomycin, San Francisco, 1978 (in press). Tedesco F, Markham R, Gurwith M, Christie D, and Bartlett JG: Oral vancomycin for antibiotic-associated pseudomembranous colitis. Lancet 2:226, 1978. Sabath LD, Wheeler N, Laverdiere M, Blazevic D, and Wilkinson BJ: A new type of penicillin resistance of Staphylococcus aureus~ Lancet 2:443, 1977. Bradley H, Weldy PL, and Hodes DS: Tolerance in Staphylococcus aureus, Lancet 1:150, 1979.