Predicting preterm delivery: comparison of cervicovaginal interleukin (IL)-1b, IL-6 and IL-8 with fetal ®bronectin and cervical dilatation Matthew A.G. Coleman a,* , Jeffrey A. Keelan b , Lesley M.E. McCowan a , Kevin M. Townend a , Murray D. Mitchell b a Department of Obstetrics and Gynaecology, University of Auckland School of Medicine, Private Bag 92019, Auckland, New Zealand b Department of Pharmacology and Clinical Pharmacology, University of Auckland School of Medicine, Private Bag 92019, Auckland, New Zealand Received 23 September 1999; received in revised form 11 May 2000; accepted 12 June 2000 Abstract Objective: To compare the cervicovaginal cytokines IL-1b, IL-6 and IL-8 with fetal ®bronectin (fFN) and cervical dilatation in the prediction of preterm delivery. Study design: Cervicovaginal cytokine concentration and fFN status were measured in 104 women with symptoms of preterm labour and intact membranes between 24 0 and 33 6 weeks and related to delivery within 2 and 7 days. Results: A group of 18% had cervical dilatation 1 cm and 18% were positive for fFN. Preterm delivery within 2 and 7 days occurred in 5 and 12%, respectively. Only IL-6 demonstrated any ability to predict delivery within 2 and 7 days (area under the ROC curve 0:63 and 0.75, respectively). Using 35 pg/ml (75th centile) as a cut-off, IL-6 had a sensitivity and speci®city of 60 and 77% for predicting delivery within 2 days, and 62 and 80% for predicting delivery within 7 days. This is similar to the performance of cervical dilatation or fFN status. Conclusions: Measurement of cervicovaginal cytokines has limited ability to predict imminent delivery apart from cervicovaginal IL-6 concentrations, which, in this population, is equivalent to that of fFN status and cervical dilatation 1 cm. # 2001 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Cervicovaginal cytokines; Preterm labour; Preterm delivery 1. Introduction Spontaneous labour is associated with activation of in¯ammatory reactions in gestational tissues [1,2]. Recruit- ment of in¯ammatory cells into the choriodecidual mem- branes and enhanced local production of cytokines and other immunomodulators characterise this process. Although labour at term, may be associated with a signi®cant in¯am- matory response [1,3±5], intrauterine infection-mediated cytokine release is thought to be a causative factor in a signi®cant proportion of preterm deliveries [6]. Pregnancies that deliver preterm have been associated with elevated amniotic ¯uid concentrations of the interleukins (IL): IL- 1b [7], IL-6 [8±10], IL-8 [11], IL-10 [12] and tumour necrosis factor-a (TNF-a) [13]. In particular, an elevated concentration of IL-6 in amniotic ¯uid appears to be a marker of intrauterine infection that will proceed to preterm delivery [8±10]. Recently, evidence has been presented that a fetal in¯ammatory response might also contribute to the increase in intrauterine cytokine concentrations that precede preterm labour [14]. Our current understanding of the pathophysiology of intra-uterine infection and the resultant cytokine response is incomplete. Furthermore, complete detection of all sig- ni®cant intrauterine infection before delivery is not possible with current technology. A more appropriate clinical diag- nostic strategy might be to attempt identi®cation of those pregnancies with symptoms of preterm labour that will deliver soon after presentation. This would allow more objective care and would avoid unnecessary treatment, particularly with tocolysis, of those women whose symp- toms will settle spontaneously and who are not at risk of immediate preterm delivery. While previous studies have focused primarily on amniotic ¯uid cytokine concentrations, amniocentesis is not a routine procedure in many institu- tions. Determination of cytokine concentrations in the cer- vicovaginal secretions is a safer and less-invasive technique. Recent research has demonstrated an association between amniotic ¯uid cytokine concentrations and those found in the cervicovaginal secretions in women with preterm labour European Journal of Obstetrics & Gynecology and Reproductive Biology 95 (2001) 154±158 * Corresponding author. Present address: Wessex Fetal Medicine Department, E level Princess Anne Hospital, Coxford Road, Southampton SO16 5YA, UK. Tel.: 44-23-8079-4228; fax: 44-23-8079-6207. 0301-2115/01/$ ± see front matter # 2001 Elsevier Science Ireland Ltd. All rights reserved. PII:S0301-2115(00)00450-4