| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Technical Briefs |
, and Interleukin-2 Receptor as Markers of Preterm Delivery
1 Department of Pathology and Immunology, Division of Laboratory Medicine, Washington University School of Medicine, St. Louis, MO;
aaddress correspondence to this author at: Department of Pathology and Immunology, Division of Laboratory Medicine, Washington University School of Medicine, 660 S. Euclid Ave., Box 8118, St. Louis, MO 63110; fax 314-362-1461, e-mail gronowski{at}pathbox.wustl.edu
Preterm delivery is a common obstetric problem (1), and infants born prematurely are at increased risk for long-term disability and death (2). Clinical indicators of preterm delivery are neither sensitive nor specific (3). Fetal fibronectin (fFN) testing of cervicovaginal fluid is the only Food and Drug Administration-cleared test for prediction of preterm delivery. A negative fFN test (<50 µg/L) in a woman with symptoms of preterm labor has been associated with a >99% negative predictive value for delivery within 7 or 14 days of testing (4)(5). The negative result enables physicians to avoid unnecessary interventions, but the fFN test is costly (more than $100 per test).
Cytokines have been investigated as biomarkers of impending preterm delivery and may favor preterm birth by stimulating prostaglandin synthesis (6)(7). Increased serum and amniotic fluid concentrations of several cytokines have been reported in women with preterm labor (8)(9)(10)(11)(12). The use of cervicovaginal fluid has been suggested as an alternative to amniotic fluid testing because this fluid is thought to be released through mechanical or inflammation-mediated damage to the membranes or placenta before birth (13). Cytokines may be released into cervicovaginal fluid during the breakdown of the chorio-decidual adhesion or from an inflammatory reaction in the same area. Few studies have been performed on symptomatic women (14)(15)(16)(17)(18)(19)(20). The objective of this study was to examine the potential usefulness of cervicovaginal interleukin-6 (IL-6), tumor necrosis factor-
(TNF), and interleukin-2 receptor (IL-2R) in the prediction of delivery in symptomatic women within 14 days of testing.
This was a retrospective cohort study of samples for physician-ordered cervicovaginal fFN tests received by the Barnes-Jewish Hospital laboratory between February 2001 and December 2002. Cervicovaginal fluid was collected from women with regular contractions and without cervical changes, before cervical examination. Specimens were included in the study if they were collected from women who had the following: a singleton pregnancy between 240/7 weeks and 346/7 weeks of gestation; symptoms of preterm labor (contractions, manifested by pain, balling up sensation, back pain, lower abdominal pressure, vaginal or rectal pressure, approximately every 4–5 min); and a chart available for review. Specimens were excluded if collected from women who had chemically induced labor or elective caesarian section within 2 weeks of fFN collection, multiple gestation, cervical dilation 
3 cm, ruptured membranes, cervical cerclage, moderate or gross vaginal bleeding, placenta previa, or sexual intercourse within 24 h.
Among 592 specimens received, 400 were excluded by the study design, primarily because samples were from other hospitals and charts were not available for review or because of caesarian section or induced delivery. Another 27 samples had insufficient volume.
Specimens were collected by use of a Dacron® swab placed in the posterior fornix of the vagina for 10 s during a sterile speculum examination. The swab was inserted into a tube containing 1 mL of fFN extraction buffer containing proprietary concentrations of bovine serum albumin, Tween 20, EDTA, phenylmethylsulfonyl fluoride, and sodium azide (Adeza Biomedical); incubated for 10 min at 37 °C; and then filtered through a plunger filter (Adeza Biomedical). Institutional Review Board approval was obtained for this study.
fFN was measured by membrane immunoassay on a Rapid fFN TLiTM Analyzer (Adeza Biomedical) (21). Specimens with fFN concentrations >50 µg/L are interpreted as positive. After fFN analysis, samples were stored at –70 °C until cytokine measurements were performed (22). IL-6, TNF, and IL-2R were quantified on the DPC Immulite® (Diagnostic Products Corporation) by use of sequential, two-site, solid-phase, chemiluminescent enzyme immunometric assays (23)(24)(25).
We performed recovery studies to validate the cytokine assays for testing of samples in the Adeza buffer. We used recombinant human TNF (from Dr. Robert Schrieber, Washington University) or serum with a high concentration of IL-6 or IL-2R. In all instances the dilutions were <10% of the total volume to minimize changes in the buffer matrix.
For statistical analysis, we used GraphPad Prism, Ver. 4.00 (GraphPad Software, Inc.). Differences between groups were assessed by the Student t-test for continuous variables or the Fisher exact test for discrete variables. Correlation between gestational age and cytokine concentration was assessed by linear regression, and the slope was evaluated for significant deviation from zero with the F-test.
The recovery of IL-6 (200–1000 ng/L) was 94–108% (see Table 1 of the Data Supplement that accompanies the online version of this Technical Brief at http://www.clinchem.org/content/vol50/issue10/). The TNF assay demonstrated a linear but markedly nonparallel response to added TNF (3–250 ng/L). We thus used results only qualitatively as above or below a cut-point. For the IL-2R assay, results were within 15% of expected at 20–500 kIU/L.
Among the 165 samples, 9 (5.5%) were associated with births within 14 days of specimen collection. The mean (SD) gestational ages at time of sampling in women who delivered 14 days or longer after testing were not significantly different: 31.4 (3.0) weeks vs 30.3 (2.5) weeks (P = 0.17). Similarly, gestational age at time of sampling did not differ between patients with positive and negative fFN results [30.0 (2.3) vs 30.2 (2.6) weeks; P = 0.78].
Cytokine concentrations were not correlated with gestational age (see Fig. 1 of the online Data Supplement): for IL-6, TNF, and IL-2R, r2 = 0.004 (P = 0.41), 0.004 (P = 0.42), and 0.001 (P = 0.76), respectively. Cytokine concentrations in specimens from women with medical record documentation of chorionitis, chorioamnionitis, cervicitis, or bacterial vaginosis were not significantly different from those from women without documentation of infection. There were significant correlations between IL-6 and TNF (r2 = 0.53), IL-6 and IL-2R (r2 = 0.57), and TNF and IL-2R (r2 = 0.61; P <0.0001 for each).
Cervicovaginal cytokine concentrations and fFN in women who delivered in 
14 days and >14 days after testing are shown in Fig. 1
. The negative predictive value for fFN (96%; 95% confidence interval, 91–99%) was consistent with other studies (4)(5). A decision threshold concentration was chosen for each cytokine to match the sensitivity of fFN (44%). At 44% sensitivity (IL-6 cutoff of 250 ng/L), the specificity of 85% and negative (96%) and positive (14%) predictive values for IL-6 were nearly identical to those for fFN (81%, 96%, and 12%, respectively). IL-2R and TNF also had high negative predictive values but had lower specificities than did fFN.
|
A combination of IL-6 (250 ng/L) and IL-2R (24 kIU/L) provided a negative predictive value of 96%, specificity of 87%, and a positive predictive value of 16%. At an IL-6 cutoff of 160 ng/L, the sensitivity of the combined test was 56%, specificity was 81%, and negative predictive value was 97%. The mean (SD) gestational age at time of delivery was significantly lower for women with a positive fFN result [36.4 (3.5) weeks; range, 25.1–40.2 weeks] compared with a negative fFN result [38.0 (1.9) weeks; range, 31.5–42.4 weeks; P = 0.0004] and for women with an IL-6 concentration 250 ng/L [36.7 (3.3) weeks; range, 25.1–40.2 weeks] compared with those with a IL-6 concentration <250 ng/L [37.9 (2.1) weeks, range, 29.1–42.4 weeks; P = 0.02].
Only an IL-6 concentration 250 ng/L was significantly associated with delivery within 14 days (odds ratio, 4.4; 95% confidence interval, 1.1–17.6; P = 0.04), although a positive fFN result approached significance (odds ratio, 3.5; 95% confidence interval, 0.9–13.9; P = 0.08; Table 1
).
|
Consistent with other reports (14)(26), we found no statistically significant correlations between gestational age (between 240/7 and 346/7 weeks of gestation) and the concentrations of IL-6, TNF, or IL-2R in this population of symptomatic women. When we used an IL-6 cutoff of 250 ng/L, the sensitivity, specificity, and positive and negative predictive values were virtually identical to those of fFN. These findings are consistent with Coleman et al. (14) and LaShay et al. (16), although they used cutoffs of 35 and 100 ng/L, respectively, for IL-6. TNF and IL-2R had negative predictive values similar to those of fFN, but a lower specificity and positive predictive value. The combination of cervicovaginal IL-6 and IL-2R may perform better than fFN, but larger studies are needed to confirm this.
In our population, 44% (4 of 9) of women who delivered within 14 days had an IL-6 concentration 250 ng/L, and this was the only cervicovaginal fluid marker that was significantly associated with delivery within 14 days (odds ratio, 4.4; P = 0.04; Table 1
). The same number of women had a positive fFN result, although this was not significantly associated with delivery within 14 days (odds ratio, 3.5; P = 0.08). This was attributable to the slightly higher number of false-positive results observed with fFN (29 false positives for fFN and 24 false positives for IL-6), and it is likely that a significant association would also have been observed for this marker if the sample population were larger.
The fFN assay is performed on the TLi instrument, which can test a single sample during its 20-min assay time. Each single-use test strip costs 
$100. The quality control is electronic and therefore has no additional cost. The cytokine assays are performed on the DPC Immulite, which is an automated random-access instrument with a 40- to 70-min assay time (IL-2R, 40 min; IL-6, 60 min; TNF, 70 min). Assays are sold for $350 for 50 tests. Approximately 10 tests are required for calibration and quality-control samples, and calibration is maintained for 14 days. Reagent cost can therefore be estimated at less than $10 per reportable result, assuming that 40 assays are performed in a 14-day period.
This study is the largest to date of cervicovaginal cytokines as predictors of preterm labor in symptomatic women. Our data provide preliminary evidence that cervicovaginal IL-6 can be used as a less expensive marker of delivery within 14 days, with performance nearly identical to fFN. Because of the small sample size of this study, larger studies are needed to confirm these findings.
Footnotes
1 current address: Department of Pathology and Laboratory Medicine, University of North Carolina at Chapel Hill, School of Medicine, Chapel Hill, NC 27599; 
References
and its soluble receptor profile in term and preterm parturition. Am J Obstet Gynecol 1999;181:1142-1148.[CrossRef][Web of Science][Medline]
[Order article via Infotrieve]
2002 Diagnostics Products Corporation Los Angeles, CA. .
, interleukin-6, and fetal fibronectin in the lower genital tract during pregnancy: relation to outcome. Am J Obstet Gynecol 1994;171:5-10.[Web of Science][Medline]
[Order article via Infotrieve]The following articles in journals at HighWire Press have cited this article:
|
|
 |
|
  A. Woodworth, J. Moore, C. G'Sell, A. Verdoes, J. A. Snyder, L. Morris, C. Wares, D. G. Grenache, and A. M. Gronowski Diagnostic Accuracy of Cervicovaginal Interleukin-6 and Interleukin-6:Albumin Ratio as Markers of Preterm Delivery Clin. Chem., August 1, 2007; 53(8): 1534 - 1540. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
