Journal of Perinatology (2007) 27, 709712

r 2007 Nature Publishing Group All rights reserved. 0743-8346/07 $30

www.nature.com/jp

ORIGINAL ARTICLE

Reliability of salivary theophylline in monitoring the treatment

for apnoea of prematurity
P Chereches-Panta1, MV Nanulescu1, M Culea2 and N Palibroda3
1
3

The 3rd Pediatric Clinic, University of Medicine, Cluj-Napoca, Romania; 2University Babes-Bolyai, Cluj-Napoca, Romania and
Molecular and Isotopic Technology Institute, Cluj-Napoca, Romania

Objective: To evaluate the reliability of salivary levels of theophylline in

monitoring therapy of apnoea of prematurity.

Study Design: Aminophylline was administered intravenously in 13

infants with apnoea, in a loading dose of 5 mg/kg and maintenance dose
of 3 mg/kg, every 8 h. The patients were divided into two groups according
to their postconceptional age (PCA): group A, of infants with small PCA
(32.82.0 weeks; n 6 cases), and group B, infants with higher PCA
(37.10.8 weeks; n 7 cases).
Result: A total of 57 paired samples of serum and saliva were obtained in
all 13 infants. The mean serum level of theophylline was 7.85.8 mg/ml
and the ratio between serum and salivary concentration of theophylline
was 1.530.28. A strong correlation between the serum and salivary
concentration of theophylline (r 0.973) was found. Infants with small
PCA had significant higher serum concentration of theophylline than
those with higher PCA (10.6 vs 5.3 mg/ml; P 0.0002). The difference
between the mean ratios of serum/salivary theophylline levels in the two
groups was low (1.44 vs 1.62; P 0.0155).
Conclusion: The strong correlation of theophylline in serum and in
saliva recommends the salivary levels as a reliable method for monitoring
the treatment of apnoea of prematurity.
Journal of Perinatology (2007) 27, 709712; doi:10.1038/sj.jp.7211804;
published online 23 August 2007
Keywords: theophylline; saliva; apnoea of prematurity

Introduction
Intravenous and oral theophylline has long been the mainstay of the
treatment and prevention of apnoea in neonates. Theophylline is an
effective drug but a high serum concentration has the potential to
cause serious side effects, including seizures, cardiac arrhythmias
and death.1,2 Theophylline dosing in infants has not been well
Correspondence: Dr P Chereches-Panta, The 3rd Pediatric Clinic, University of Medicine and
Pharmacy, Campeni Street No 2-4, 400217 Cluj-Napoca, Cluj, Romania.
E-mail: pusacherechespanta@gmail.com
Received 7 February 2007; revised 28 June 2007; accepted 6 July 2007; published online
23 August 2007

defined.3,4 Small infants have a decreased capacity to metabolize

theophylline and the metabolic path is different from that in older
children and adults. Some authors demonstrated that a variability of
metabolism and subsequently of serum concentration of theophylline
correlated with postconceptional age (PCA) of neonates.5
Therefore, it is compulsory to measure serum levels for monitoring
the treatment of apnoea of prematurity. However, measuring the
salivary level of theophylline is a potentially useful alternative to
serum level, because sampling saliva is less invasive.6 There are
data in the literature consistent with the reliability of salivary
theophylline concentration for estimation of serum concentration,
but most of the studies were performed in older children or
adults with asthma. Only a few studies have been conducted in
premature infants and the results are not satisfactory.79
The purpose of this study was to evaluate the correlation
between serum and saliva concentration of theophylline in infants
with apnoea of prematurity. A secondary variable studied was the
influence of PCA on theophylline metabolism.

Methods
Patients
The study group included newborn infants according to the following
criteria: (a) gestational age <36 weeks; (b) episodes of apnoea that
lasted at least 20 s and (c) episodes of apnoea <20 s accompanied by
cyanosis and/or bradycardia. Exclusion criteria were as follows:
(a) any congenital abnormality; (b) infection (sepsis confirmed
by blood culture); (c) neurological disorders; (d) severe hepatic
diseases; (e) renal diseases and (f) concomittent medication that
might interfere with theophylline metabolism. Thirteen infants aged
between 2 and 10 weeks underwent the study. All parents were
informed of the nature of the study and written consent was
obtained. The study protocol was approved by the ethics committee.
Treatment with theophylline
We used ampoules of aminophylline (85% theophylline)
containing 240 mg/10 ml Miofilin 2.4% (Terapia SA, Cluj-Napoca).
A loading dose of 5 mg/kg of aminophylline was administered
intravenously (i.v.) over 10-min period, followed by maintenance

Salivary theophylline for apnoea of prematurity

P Chereches-Panta et al

710

Blood and saliva collection

Blood and saliva samples were obtained simultaneously at 2 and
4 h after the loading dose. During the next 2 days, simultaneous
samples were collected before the maintenance dose and at 4 h
after the administration of aminophylline. Venous blood (12 ml)
was collected from an arm different from the one in which
theophylline was administered. For the stimulation of mixed saliva
secretion, we placed citric acid crystals (12 mg) on the tongue.
Approximately 0.51 ml of saliva was then collected by aspiration
in a sterile recipient. The duration of saliva sampling varied
between 10 and 15 min.
Analysis
Serum and saliva samples were kept refrigerated for 4872 h at
201C before extraction. We measured the concentration of
theophylline by isotope dilution mass spectrometry, using as an
internal standard theophylline labelled with the 15N isotope.
Analysis in serum and saliva was performed on the molecular ion,
m/z 180 for the natural theophylline and m/z 181 for the
internal standard. This method does not allow interferences
between theophylline and caffeine measuring.
Statistics
We calculated the coefficient of correlation and analysed the
regression curve, to estimate the correlation between serum and
saliva concentration. We calculated the coefficient of variation or
relative standard deviation (s.d.) of the ratio between serum and
saliva concentration. The Students t-test was used and P<0.01 was
considered as statistically significant.
Results
The study group consisted of 13 premature newborn infants. The
infants were divided into two subgroups according to their PCA.
Group A consisted of six infants with PCA of 32.82.0 weeks
(between 30 and 35 weeks) and group B included seven infants
with PCA of 37.10.8 weeks (between 36 and 38 weeks).
Demographic data are presented in Table 1.
After obtaining a stable level of theophylline by i.v. loading dose,
a total of 78 blood samples were obtained. Saliva was collected
simultaneously with blood and 64 samples were obtained. From
these saliva samples, seven were excluded because of
contamination with blood during collection (from superficial
lesions of oral mucosa, probably due to manipulation during
collection) or with local debris.
The mean value of the serum concentration was 7.85.8 mg/ml
(means.d.) varying between 1.6 and 27.9 mg/ml, while salivary
concentration was 5.54.6 mg/ml.
Journal of Perinatology

Table 1 Mean demographic data of the two groups

Group A (n 6 cases)
Gender
Birth weight (g)

a
b

Group B (n 7 cases)

3 males; 3 females
4 males; 3 females
1800225.8a (between 1400 1657436.3a (between 950
and 2100)
and 2300)
30.22.8a (2735)
31.63.7a (2735)

Gestational age
(weeks)
Postconceptional age 32.82.0a (3035)
(weeks)b

37.10.8a (3638)

Means.d.
Age at which theophylline was first administered.

25.00
Salivary level of theophylline, g/ml

doses of 3 mg/kg given every 8 h for the next days. The mean
duration of therapy was 15 days and the outcome was good
in all cases.

20.00

y=0.6822x + 0.0726

15.00

10.00

5.00

0.00
0.00

5.00

10.00

15.00

20.00

25.00

30.00

Serum concentration of theophylline, g/ml

Figure 1 Correlation between serum and saliva concentration of theophylline;

r 0.973.

The ratio between the serum and saliva levels for the 57 paired
samples was 1.530.28. Calculation of relative s.d, revealed a value
of 5.44%, which reflects a small interindividual variation of the
ratio. The analysis of the regression curve showed a strong
correlation between serum and saliva concentration, with a
correlation coefficient of r 0.973 and a slope of B 0.6822
(Figure 1). We examined the ratio at different intervals after
administration of theophylline and it revealed no significant
differences. After loading dose, the ratio was 1.43 at 2 h
and 1.53 at 4 h. The coefficient of correlation between serum
and saliva levels of theophylline was 0.986 and 0.963 at these
two intervals. In the second and third days of therapy, the ratio
was 1.52 and 1.67 at 8 h after maintenance dose and 1.60 and
1.44, respectively, at 4 h after the dose administration. The
coefficient of correlation varied between 0.957 and 0.999 during
the second and third days of therapy at different intervals after
dose administration.
There were differences of theophylline concentration between
the two age groups. In group A, with small PCA, serum
concentration of theophylline was 10.67.4 mg/ml and saliva
concentration was 7.95.5 mg/ml. The ratio between serum and
saliva levels in 27 paired samples was 1.440.22 with a small

Salivary theophylline for apnoea of prematurity

P Chereches-Panta et al

711
Table 2 The differences between the two age groups

Serum theophylline concentration (mg/ml)

Saliva theophylline concentration (mg/ml)
Ratio serum/saliva concentration
Coefficient of correlation (r)

Group A
PCA of 32.82.0 weeks
(n 27 paired samples)

Group B
PCA of 37.10.8 weeks
(n 30 paired samples)

P-value

10.567.42
7.905.55
1.440.22
0.975

5.352.01
3.201.25
1.620.31
0.836

0.0002
0.0001
0.0155

Abbreviation: PCA, postconceptional age.

interindividual variation of 6.56%. The correlation between serum

and saliva concentration was strong (r 0.975).
In group B, with higher PCA, the serum concentration of
theophylline was 5.32.0 mg/ml and saliva concentration was
3.21.2 mg/ml. The ratio between serum and saliva levels in
30 paired samples was 1.620.31 with a small interindividual
variation of 5.22%. The correlation between serum and saliva
concentration was strong (r 0.836). The differences between the
two age groups were statistically significant for both serum and
saliva theophylline concentration. Regarding the differences
between the value of the serum/saliva concentration ratio in the
two groups, a significance of P 0.0155 was found (Table 2).
Discussion
Theophylline is a major drug in the management of idiopathic
apnoea of prematurity. There is a large interpatient variability in
dosage requirements due to interpatient variability in theophylline
clearance, which also varies with age. The metabolic path in
premature infants is different from that in older infants and
children.1013 There have been extensive studies of the evaluation
of theophylline dosing equations that are useful for this age
group.35 Therefore, close treatment monitoring by serum or saliva
is compulsory.
The few data in the literature on the value of saliva
concentration in premature infants are still controversial. Khanna
et al.7 reported that saliva levels are reliable for monitoring the
therapy with both theophylline and caffeine in premature infants.
Three years later, Toback et al.9 revealed a ratio between serum
and saliva of 1.020.09, which is smaller than in other studies.
They estimated that saliva concentration in premature infants is
approximately equal to serum concentration of theophylline.9 In
the study of Aviram et al.,8 the ratio was 1.780.22 and the
correlation coefficient r 0.96. These authors included both
premature infants and older children with asthma in their
study group.8
The results of our study confirm the strong correlation between
serum and saliva levels of theophylline (r 0.973). Moreover, the
small interindividual variation of the serum/saliva concentration

ratio would allow the use of saliva concentration multiplied by

1.53 (the value of the ratio) for monitoring the therapy.
In the past 30 years, there have been extensive studies on
the value of salivary concentration in monitoring therapy with
anticonvulsivant drugs (carbamazepine, phenytoin, phenobarbital,
primidone, ethosuximide valproic acid), antibiotics, caffeine,
paracetamol and cortisol.14,15 The results are variable for each of
these drugs. According to the transmembraneous transfer, drugs
may be classified into four groups based on their ratio between
salivary/serum levels (<1.0, 1 or >1.0) and the pH.16 The therapy
of apnoea of prematurity with caffeine citrate might also be
monitored using saliva.17 The salivary concentration of caffeine
predicts its serum level with a good reliability.
One factor that may influence the concentration of theophylline
is PCA. Kraus et al.5 showed that infants with PCA below 40 weeks
have a higher risk for achieving toxic levels of theophylline. Our
data also revealed significant higher concentration of theophylline
in both serum and saliva for infants with smaller PCA. Although
we found differences in the ratio between serum and saliva
concentration in the two age groups, these differences were not
significant (P>0.01).
In conclusion, measuring saliva theophylline concentration is a
reliable method for monitoring treatment with theophylline in
premature infants with apnoea. This technique is more acceptable
than measuring theophylline in serum, because it is painless,
simple, less invasive, cheaper and allows multiple sampling during
the day.
References
1 Calhoun LK. Pharmacologic management of apnea of prematurity (Review).
J Perinatal Neonatal Nurs 1996; 9(4): 5662.
2 Herzlinger R. Apnea In: Oski FA (ed). Principles and Practice of Pediatrics. JB
Lippincott Company: Philadelphia, PA, 1994, pp 381382.
3 Hogue SL, Phelps SJ. Evaluation of the three theophylline dosing equations for use in
infants up to one year of age. J Pediatr 1993; 123: 651656.
4 Bhatt-Mehta V, Johnson CE, Donn SM, Spadoni V, Schork MA. Accuracy and reliability
of dosing equations to individualize theophylline treatment of apnea of prematurity.
Pharmacother 1995; 15(2): 246250.
5 Kraus DM, Hatzoupoulos FK, Reitz SJ, Fischer JH. Pharmacokinetic evaluation of two
theophylline dosing methods for infants. Ther Drug Monit 1994; 16(3): 270276.

Journal of Perinatology

Salivary theophylline for apnoea of prematurity

P Chereches-Panta et al

712
6 Koysooko R, Ellis EF, Levy G. Relationship between theophylline concentration in
plasma and saliva in man. Clin Pharmacol 1974; 15: 454.
7 Khanna NN, Bada HS, Somani SM. Use of salivary concentrations in the prediction of
serum caffeine and theophylline concentrations in premature infants. J Pediatr 1980;
96(3): 494499.
8 Aviram M, Tal A, Ben-Zvi Z, Gorodisher R. Monitoring theophylline therapy using citric
acid-stimulated saliva in infants and children with asthma. Pediatrics 1987; 80(6):
894897.
9 Toback JW, Gal P, Erkan NV, Roop C, Robinson H. Usefulness of theophylline saliva
levels in neonates. Ther Drug Monit 1983; 5(2): 185189.
10 Milsap RL, Hill MR, Szefler SJ. Special pharmakokinetic considerations in children. In:
Evans WE, Schentag JJ, Jusko WJ (eds). Applied Pharmacokinetics, Principles of
Therapeutic Drug Monitoring, Third edition. Applied Therapeutics Inc.: Spokane, WA,
1992; 10-110-25.
11 Moore ES, Faix RG, Banagale RC, Grasela TH. The population pharmacokinetics of
theophylline in neonates and young infants. J Pharm Biopharm 1989; 17(1): 4666.

Journal of Perinatology

12 Nahata MC. Pediatrics. In: DiPiro JT, Talbert RL, Yee GC, Matzke GR, Wells BG, Posey
LM (eds). Pharmacotherapy, A Pathophysiologic Approach. Third edition. Appleton &
Lange: Stamford, CT, 1997; 7785.
13 Martin III ES. The population pharmacokinetics of theophylline during the early
postneonatal period. J Pharm Biopharm 1991; 19(3): 59S77S.
14 Liu H, Delgado MR. Therapeutic drug concentration monitoring using saliva samples.
Focus on anticonvulsants. Clin Pharmacokinet 1999; 36(6): 453470.
15 Vasudev A, Tripathi KD, Puri V. Correlation of serum and salivary carbamazepine
concentration in epileptic patients: implications for therapeutic drug monitoring.
Neurol India 2002; 50: 6062.
16 Gorodischer R, Burtin P, Verjee Z, Hwang P, Koren G. Is saliva suitable for therapeutic
monitoring of anticonvulsants in children. An evaluation in the routine clinical setting.
Ther Drug Monit 1997; 19(6): 637642.
17 Khanna NN, Somani SM, Boyer A, Miller Y, Chua C, Meuke JA et al. Cross validation of
serum to saliva relationship of caffeine, theophylline and total methylxantine in
neonates. Dev Pharmacol Ther 1982; 4(12): 1827.