Secondary Logo

Journal Logo

Prediction of Spontaneous Preterm Delivery in Women With Preterm Labor

Analysis of Multiple Proteins in Amniotic and Cervical Fluids

Holst, Rose-Marie, MD, PhD1; Hagberg, Henrik, MD, PhD1,2; Wennerholm, Ulla-Britt, MD, PhD1; Skogstrand, Kristin, PhD3,4; Thorsen, Poul, MD, PhD4,5; Jacobsson, Bo, MD, PhD1,3,6

doi: 10.1097/AOG.0b013e3181ae6a08
Original Research
Free

OBJECTIVE: To analyze whether specific proteins in amniotic and cervical fluids, alone or in combination with risk factors, can identify women in preterm labor with intact membranes who will deliver spontaneously within 7 days of sampling.

METHODS: In a cohort of 89 women in preterm labor, amniotic and cervical fluids were collected between 22 and 33 weeks of gestation. Twenty-seven proteins were analyzed simultaneously using multiplex technology. Individual levels of each protein were compared and calculations performed to find associations among different proteins, background variables, and spontaneous preterm delivery within 7 days of sampling. The area under the curve (AUC) was calculated using receiver operating characteristic curve analysis, and prediction models were created based on stepwise logistic regression.

RESULTS: We found two multivariable models that predicted spontaneous preterm delivery better than any single variable. One combined multivariable prediction model was based on amniotic macrophage inflammatory protein-1β, cervical interferon-γ, and monocyte chemotactic protein-1. This model predicted outcome with 91% sensitivity, 84% specificity, 78% positive predictive value, and 94% negative predictive value, with a likelihood ratio of 5.6 and AUC of 0.91. An alternative, noninvasive model based on cervical length, cervical interferon-γ, interleukin-6, and monocyte chemotactic protein-1 predicted delivery within 7 days with 85% sensitivity, 82% specificity, 74% positive predictive value, and 90% negative predictive value, with a likelihood ratio of 4.7 and AUC of 0.91.

CONCLUSION: A combination of proteins from amniotic fluid and cervical fluid or cervical length can help determine which women will deliver preterm.

LEVEL OF EVIDENCE: II

A combination of proteins from amniotic fluid and cervical fluid or cervical length can help determine which women will deliver preterm.

From the 1Perinatal Center, Department of Obstetrics and Gynecology, Institute of Clinical Sciences, Sahlgrenska Academy, Sahlgrenska University Hospital/East, Göteborg, Sweden; the 2Imperial College, Institute of Reproductive and Developmental Biology, Queen Charlotte’s & Chelsea Hospital, Hammersmith Campus, London, United Kingdom; the 3Department of Clinical Biochemistry, Statens Serum Institut, Copenhagen, Denmark; 4NANEA at the Department of Epidemiology, Aarhus University, Aarhus, Denmark; the 5Department of Epidemiology, Rollins School of Public Health, Emory University, Atlanta, Georgia; and the 6Department of Obstetrics and Gynecology, Rikshospitalet, Oslo, Norway.

Supported by grants from the Swedish Medical Research Council (VR 2006-3396), Swedish Medical Society (SLS 2008-21198), and Swedish governmental grants to researchers in the public health service (ALFGBG-2863, ALFGBG-11522), the Göteborg Medical Society.

The authors thank Mattias Molin, who assisted with the statistical analyses.

Corresponding author: Dr. Bo Jacobsson, Perinatal Center, Department of Obstetrics and Gynaecology, Sahlgrenska Academy, Sahlgrenska University Hospital/East, 416 85 Göteborg, Sweden; e-mail: bo.jacobsson@obgyn.gu.se.

Financial Disclosure The authors did not report any potential conflicts of interest.

Research in biomedicine has developed rapidly over the past decade. New tools for the diagnosis of medical syndromes give researchers new options for solving old medical problems. Preterm delivery is a major unsolved obstetrical problems worldwide, contributing significantly to the incidence of perinatal mortality and long-term morbidity. In many countries, the rates of preterm delivery (United States 12.5%) continue to rise.1 However, in Sweden, the frequency has fluctuated around 5.5% for the past 30 years.2 As much as 75% of total perinatal mortality and nearly half of neurologic morbidity are attributed to preterm delivery.3–5

Spontaneous preterm delivery is a complex obstetric and perinatal problem, and the necessity of finding good prediction models is urgent. Much data indicates that increased production of cytokines and other proteins in maternal and fetal serum and in amniotic and cervical fluid is involved in both term and preterm labor.6–8 A new, multiplexed sandwich immunoassay has been developed based on flowmetric xMAP technology. This technology makes it possible to analyze an array of proteins simultaneously using only small sample volumes. The xMAP technology has been used to analyze multiple inflammatory markers and neurotrophins in neonatal dried blood spots and from maternal plasma and cervical–vaginal mucous in women with preterm labor.8,9 In the current study, we analyzed a panel of 27 selected proteins in amniotic and cervical fluid using this technology. Some of the proteins analyzed have been evaluated previously as markers for spontaneous preterm delivery in amniotic or cervical fluid, whereas others, to our knowledge, have not.

The aim of this study was to analyze whether these selected proteins (cytokines and neurotrophins) in amniotic and cervical fluid, alone or in combination with each other or with the risk factors cervical length, smoking, and previous preterm delivery, could predict spontaneous preterm delivery, identify women at high risk, and distinguish them from women at low risk of spontaneous preterm delivery. The primary end point was delivery within 7 days of sampling.

Back to Top | Article Outline

MATERIALS AND METHODS

In this prospective cohort study, 89 women with singleton pregnancies who were in preterm labor presenting at Sahlgrenska University Hospital/East, Gothenburg, Sweden, from 1996–2005, gestational age 22 0/7 weeks to 33 6/7 weeks, were enrolled. Preterm labor was defined as regular uterine contractions in combination with cervical changes and intact membranes. Inclusion and exclusion criteria for the cohort, together with data retrieval information and sampling procedures have been described previously.10 Owing to limited resources, we did not recruit women at nighttime, on weekends, at Christmas, and during summertime nor during busy hours at the delivery ward, which limited the number of women enrolled in the study.

Concentrations of inflammatory markers (interleukin [IL]-1β, IL-2, IL-4, IL-5, IL-6, IL-8, IL-10, IL-12, IL-17, IL-18, soluble IL-6 receptor α, interferon [IFN]-γ, tumor necrosis factor [TNF]-α, TNF-β, monocyte chemotactic protein-1, transforming growth factor-β, macrophage inflammatory protein-1α, macrophage inflammatory protein-1β, matrix metalloproteinasis-9, triggering receptor expressed on myeloid cells-1, brain-derived neurotrophic factor, granulocyte-macrophage-colony–stimulating factor [CSF], neurotrophin-3, neurotrophin-4, soluble TNF receptor I, migration inhibitory factor [MIF], regulated on activation, normal T-expressed and secreted) in amniotic and cervical fluid were determined at Statens Serum Institute (Department of Clinical Biochemistry, Denmark) using a multiplex sandwich immunoassay based on flowmetric xMAP technology as previously described.9 The xMAP technology is based on flowmetric analysis of microbeads that act as solid support for individual assay reactions incorporating a common fluorophore reporter. Assays for several analytes can be made simultaneously on different sets of beads with unique fluorescence characteristics. This technique makes it possible to measure a broad panel of proteins with small sample volumes.

The amniotic and cervical samples were measured undiluted in duplicates. Fifty microliters of sample was added to each filter plate well and 50 microliters of a suspension of capture-antibody-conjugated beads, 1,500 beads per analyte. After 1.5 hours of incubation, the beads were washed twice and subsequently reacted for 1.5 hours with a mixture (50 microliters) of relevant biotinylated detection antibodies, each diluted 1:1,000; next, 50 microliters of streptavidin-phycoerythrin, 20 micrograms/mL, were added to the wells. Incubation then was continued for an additional 30 minutes. Finally, the beads were washed twice and resuspended in 125-microliter buffer and analyzed on the platform.

In both amniotic and cervical fluid, the mean intraassay coefficient of variation was 6% and the mean interassay coefficient of variation was 12%. We used a defined working range as described by Skogstrand instead of the more commonly used signal-to-noise ratio (limit of detection) because this was considered to be a more precise way of defining the sensitivity because it was not possible to obtain amniotic and cervical fluid depleted of cytokines.9 Thus, the detection level was set as half the lowest concentrations in the working range (in both amniotic and cervical fluid): IL-β (40 pg/mL), IL-2 (4 pg/mL), IL-4 (4 pg/mL), IL-5 (4 pg/mL), IL-6 (40 pg/mL), IL-8 (40 pg/mL), IL-10 (10 pg/mL), IL-12 (4 pg/mL), IL-17 (4 pg/mL), IL-18 (40 pg/mL), soluble IL-6 receptor α (2,500 pg/mL), IFN-λ (4 pg/mL), TNF-α (4 pg/mL), TNF-β (4 pg/mL), monocyte chemotactic protein-1 (156 pg/mL), transforming growth factor-β (4 pg/mL), macrophage inflammatory protein-1α (40 pg/mL), macrophage inflammatory protein-1β (40 pg/mL), matrix metalloproteinasis-9 (5,000 pg/mL), triggering receptor expressed on myeloid cells-1 (100 pg/mL), brain-derived neurotrophic factor (10 mg/mL), granulocyte-macrophage–CSF (4 pg/mL), neurotrophin-3 (40 pg/mL), neurotrophin-4 (4pg/mL), soluble TNF receptor I (156 pg/mL), MIF (100 pg/mL), regulated on activation, normal T-expressed and secreted (40 pg/mL). The staff member (K.S.) who performed the protein analyses did not have any clinical information on the outcomes of the women.

The study was approved by the local ethics committee at the University of Gothenburg (No. 349–95, 476–05). All patients gave informed consent before enrollment.

For comparisons between women with and without delivery within 7 days, Mann-Whitney U-test was used for continuous variables and Fisher exact test for dichotomous variables. Receiver operating characteristic (ROC) curves and area under the curve (AUC) for prediction of delivery within 7 days were computed for each protein and for the background variables. Continuous variables were dichotomized from the ROC curve based on the concentration to obtain the optimal prediction of delivery within 7 days. Logistic regressions with these dichotomized variables as independent variables were used to obtain odds ratios (ORs). The variables in the univariable analyses significant at P<.001 were entered into a stepwise logistic regression. All analyses were performed for each amniotic and cervical marker and the combinations. Spearman’s rank correlation test was used for analysis of correlation between continuous variables. A P<.05 or a confidence interval (CI) not including 1 was considered statistically significant. All analyses were computed using SAS 9.1 or 9.2 (SAS Institute, Inc., Cary, NC) or StatView 5.01 (SAS Institute, Inc.).

Back to Top | Article Outline

RESULTS

In the current study, we examined 27 proteins in both amniotic and cervical fluid in a cohort of 89 pregnant women with symptoms of preterm labor together with cervical length assessed by transvaginal ultrasonography and the maternal characteristics smoking and previous late miscarriage or spontaneous preterm delivery or both. The predictive value for delivery within 7 days for each marker separately and combined was calculated.

The study population consisted of 89 healthy women with singleton pregnancies. Patients were included in the study at a median of 30 5/7 weeks (range 22 2/7 weeks to 33 5/7 weeks), and the median gestational age at delivery was 34 3/7 weeks (range 23 2/7 weeks to 43 0/7 weeks). Delivery within 7 days occurred in 38% (34/89) of the cases. Demographic characteristics are presented in Table 1. Women who delivered preterm (within 7 days of sampling) did not differ from those who delivered later in regard to age, parity, number of previous gestations, number of previous preterm deliveries, smoking habits, or gestational age at study inclusion. The median cervical length was significantly shorter in women who delivered within 7 days (5.5 mm) compared with that in women who did not (20 mm). The women who delivered within 7 days were given corticosteroids before sampling more often than were the women who delivered later.

Table 1

Table 1

The median concentrations of all proteins in amniotic and cervical fluid in relation to delivery within 7 days and delivery after 7 days and AUC together with OR, 95% CI, and P values are presented in Tables 2 and 3. The diagnostic indices for a single protein (data not shown) were not nearly as good as those for the combination of various markers in the multivariable models.

Table 2

Table 2

Table 3

Table 3

The IL-8 and matrix metalloproteinasis-9 results from analysis of both amniotic and cervical fluid were excluded from the analyses because of methodologic problems. The levels were in the upper range of the standard curve, but the limited amount of sample volume did not permit further testing.

Women who delivered within 7 days had significantly higher levels of several proteins, IL-1β, IL-6, IL-10, IL-17, IL-18, TNF-α, monocyte chemotactic protein-1, macrophage inflammatory protein-1α, macrophage inflammatory protein-1β, granulocyte-macrophage–CSF, and regulated on activation, normal T-expressed and secreted in their amniotic fluid than did the women who delivered later (Table 2). The highest values of AUC in amniotic fluid were detected for IL-6 (AUC=0.82), macrophage inflammatory protein-1β (AUC=0.80), and IL-1β (AUC=0.80).

In cervical fluid, the median concentrations of IL-1β, IL-4, IL-5, IL-6, IL-10, IL-17, soluble IL-6 receptor α, IFN-γ, TNF-α, monocyte chemotactic protein-1, macrophage inflammatory protein-1α, macrophage inflammatory protein-1β, triggering receptor expressed on myeloid cells-1, brain-derived neurotrophic factor, granulocyte-macrophage–CSF, neurotrophin-3, neurotrophin-4, MIF, and regulated on activation, normal T-expressed and secreted were significantly higher in women who delivered within 7 days than in women who delivered later (Table 3). The AUC for cervical length measured by transvaginal ultrasonography was 0.78; for both of the cytokines IL-6 and macrophage inflammatory protein-1α, the AUC was 0.74.

The correlations between the individual proteins in the two different compartments were generally low. There were, however, significant correlations between the two fluids for IL-1β (rho=0.42, P<.001), IL-5 (rho=0.211, P<.049), IL-6 (rho=0.384, P<.001), IL-10 (rho=0.229, P<.033), IL-17 (rho=0.249, P=.019), soluble IL-6 receptor α (rho=0.221, P<.050), TNF-β (rho=0.235, P=.028), macrophage inflammatory protein-1α (rho=0.453. P<.001), and macrophage inflammatory protein-1β (rho=0.414, P<.001).

To better predict which women with symptoms of preterm labor would deliver within 7 days of sampling, protein levels from both amniotic and cervical fluid along with the background variables were included in a multivariable analysis. Based on a stepwise multivariable logistic regression on dichotomous variables, the 10 most significant variables in the univariable analyses then were used to construct prediction models using variables in amniotic (Table 4) and cervical fluid (Table 5) separately and in a model combining proteins from the two different compartments (Table 6). In the multivariable model analyzing proteins in the amniotic fluid along with the background variables (Table 4), only high levels of macrophage inflammatory protein-1β (≥3.05 ng/mL) (OR 15.5, 95% CI 3.0–80.0) and IL-6 (≥10.0 ng/mL) (OR 7.8, 95% CI 1.9–31.5) contributed significantly to the prediction of spontaneous preterm delivery within 7 days of sampling. The AUC was 0.89. In the model using cervical fluid cytokines together with the background variables (Table 5), high levels of monocyte chemotactic protein-1 (≥2.7 ng/mL) (OR 8.1, 95% CI 2.1–30.5), IL-6 (≥36.6 ng/mL) (OR 6.2, 95% CI 1.5–25.3), and IFN-γ (≥0.021 ng/mL) (OR 5.8, 95% CI 1.3–24.8) and cervical length 10 mm or more (OR 0.13, 95% CI 0.03–0.55) contributed significantly to the prediction of spontaneous preterm delivery within 7 days. The AUC was 0.907 (Fig. 1). In the combined prediction model using both amniotic and cervical proteins along with the background variables (Table 6), high levels of amniotic fluid macrophage inflammatory protein-1β (≥3.05 ng/mL) (OR 25.6, 95% CI 4.4–150), cervical fluid IFN-γ (≥0.021 ng/mL) (OR 5.5, 95% CI 1.2–25.2), and cervical fluid monocyte chemotactic protein-1 (≥2.7ng/mL) (OR 7.2, 95% CI 1.9–26.8) contributed significantly to the prediction of spontaneous preterm delivery within 7 days. Using this combined model, 86.5% of the women who delivered within 7 days of sampling were identified (Table 6). The ROC curve showed an AUC of 0.913 (Fig. 2). We also calculated that we had to do 18 invasive amniocentesis procedures instead of only a noninvasive cervical sampling to predict one extra preterm delivery by calculating numbers needed to treat.

Table 4

Table 4

Table 5

Table 5

Table 6

Table 6

Fig. 1.Holst. Prediction of Spontaneous Preterm Delivery. Obstet Gynecol 2009.

Fig. 1.Holst. Prediction of Spontaneous Preterm Delivery. Obstet Gynecol 2009.

Fig. 2.Holst. Prediction of Spontaneous Preterm Delivery. Obstet Gynecol 2009.

Fig. 2.Holst. Prediction of Spontaneous Preterm Delivery. Obstet Gynecol 2009.

Back to Top | Article Outline

DISCUSSION

Prediction of a woman’s risk of spontaneous preterm delivery and treatment of women at high risk remain major unsolved problems in contemporary obstetrics. It is essential to find good predictors of spontaneous preterm delivery. An ideal prediction model for spontaneous preterm delivery would be a test making it possible to avoid invasive procedures to obtain the analytes and with predictive values high enough to enable physicians to avoid unnecessary interventions.

In our study of women in preterm labor, we found two models for predicting delivery within 7 days. The model that best predicted spontaneous preterm delivery consisted of a combination of proteins from both amniotic and cervical fluid. Amniotic macrophage inflammatory protein-1β together with cervical IFN-γ and cervical monocyte chemotactic protein-1 predicted spontaneous preterm delivery within 7 days with 91% sensitivity, 84% specificity, 76% positive predictive value, and 94% negative predictive value, with a likelihood ratio 5.6. The ROC-curve analyses of the combined prediction model had an AUC of 0.91. The model using markers collected noninvasively, including cervical length and cervical IFN-γ, IL-6, and monocyte chemotactic protein-1, did not achieve as good diagnostic results as the combination of markers from both compartments.

Others have studied some of the same proteins and background variables and their association with spontaneous preterm delivery as in the present data set. Goldenberg et al studied several markers of spontaneous preterm delivery in a nested, case–control study consisting of asymptomatic women in gestational weeks 23 to 24.6 The majority of markers were analyzed in maternal serum, some from cervix and vagina. They tried to construct a multi-marker test for spontaneous preterm delivery and found that there was very little overlap among the strongest biologic markers for spontaneous preterm delivery, suggesting that a combination of different markers would enhance the ability to predict spontaneous preterm delivery.

Vogel et al studied some of the same proteins that we did in maternal serum and cervical–vaginal fluid in asymptomatic high-risk women between 16 and 30 weeks of gestation.8 They also used the xMAP technology to analyze the protein levels. In their study, the best model of prediction for spontaneous preterm delivery before 35 weeks of gestation was a combination of high levels of TNF-α in maternal serum and high levels of cervical–vaginal-soluble IL-6 receptor α combined with a cervical length less than 25 mm. This model had a 69% sensitivity, 95% specificity, 82% positive predictive value, and 91% negative predictive value, thus slightly better than ours. However, they studied a different population—a cohort of asymptomatic high-risk women, predominantly African-American women with low socioeconomic status and with a history of spontaneous preterm delivery between 16 and 30 weeks of gestation. Our population consisted of a majority of socially and financially well-situated women—white (83%), Asian (10%), Hispanic (4%), and black African (2%).

We also analyzed soluble IL-6 receptor α in cervical fluid. This cytokine belongs to the IL-6 group. Interleukin-6 can either activate IL-6Rα bound to the membrane that stimulates gp130 and intracellular signaling or interact with soluble IL-6 receptor α, which probably also activates the intracellular cascade.11 We found in this study that high levels of soluble IL-6 receptor α in cervix was significantly associated with spontaneous preterm delivery in patients in preterm labor, which adds additional evidence that IL-6 family proteins are involved in spontaneous preterm delivery. The fact that soluble IL-6 receptor α with IL-6 are induced as part of the proinflammatory response in preterm labor also indicates that they might play an important role in IL-6 signaling. The soluble IL-6 receptor α levels early in pregnancy in nonlaboring women seem to have a different meaning because low levels of cervical–vaginal-soluble IL-6 receptor α in their study of asymptomatic pregnant women in midgestation was associated with an increased risk of spontaneous preterm delivery.

Grenache et al studied the potential usefulness of IL-6, TNF-α, and IL-2R in cervico–vaginal fluid in symptomatic women for prediction of spontaneous preterm delivery within 14 days of testing.12 The study was a retrospective cohort study of samples from cervical–vaginal fluid originally collected for fetal fibronectin testing. Only nine patients delivered within 14 days, and IL-6 was found to be the only factor significantly associated with preterm delivery. In our previous work, we have found that cervical IL-6 relates significantly to delivery within 7 days of assessment as well as to delivery before 34 weeks of gestation.13 This relationship was verified in the present study, in which we found that both IL-6 and TNF-α in both amniotic and cervical fluid were strongly associated with spontaneous preterm delivery within 7 days of assessment among symptomatic women.

In our strongest model of prediction, high levels of macrophage inflammatory protein-1β in amniotic fluid played an important role. This cytokine is one of four members of the macrophage inflammatory protein-1 CC chemokine subfamily. These proteins are produced by many cells, particularly macrophages, dendritic cells, and lymphocytes. Macrophage inflammatory protein-1 proteins, which act via cell surface receptors expressed by lymphocytes and monocytes/macrophages, are best known for their chemotactic and proinflammatory effects. Both macrophage inflammatory protein-1α and macrophage inflammatory protein-1β activate production of IL-1, IL-6, and TNF from monocytes and expression of beta1-integrins on endothelial cells. Dudley et al showed that human decidual cells in culture produced macrophage inflammatory protein-1α in response to other inflammatory cytokines. This was interpreted as evidence that decidual cell production of macrophage inflammatory protein-1α could be an important early event in the pathophysiology of infection-associated preterm labor. Microbial invasion of the amniotic cavity has been associated with increased levels of macrophage inflammatory protein-1α in both preterm and term gestations, and women who presented with clinically evident chorioamnionitis displayed the highest concentrations of amniotic fluid macrophage inflammatory protein-1α.14–16 This is in agreement with our findings; we found that levels of cervical and amniotic macrophage inflammatory protein-1α and macrophage inflammatory protein-1β were significantly higher in women who delivered preterm (within 7 days) as compared with women who delivered later.

Another cytokine in our prediction model was IFN-γ, which is secreted primarily by T cells (CD4+ T cells, CD8+ T cells, dendritic and natural killer cells). It has been viewed as having considerable potential as an immunotherapeutic agent owing to its ability to enhance the antimicrobial actions of host defenses against a diverse group of pathogens. Interferon-γ has antiviral, immunoregulatory, and antitumor properties. It alters transcription in up to 30 genes, producing a variety of physiologic and cellular responses. It also increases antigen presentation of macrophages, activates and increases lysozyme activity in macrophages, suppresses Th2 cell activity, causes normal cells to express class II major histocompatability complex molecules, promotes the adhesion and binding required for leukocyte migration, and promotes natural killer cell activity. Our data did not show any relations between amniotic IFN-γ and delivery within 7 days. In cervical fluid, however, there was a strong relationship between high levels of IFN-γ (an ROC-curve analysis gave a cutoff value of 0.021 ng/mL [OR 9.0, 95% CI 2.8–29]), an AUC of 0.71, (P<.001), and delivery within 7 days. This difference between the two compartments might be explained by the presence of more T cells in cervical than in amniotic fluid.

Our research group has shown previously that the chemokine monocyte chemotactic protein-1 in cervical and amniotic fluid is elevated in patients with preterm labor and that there is a correlation to intraamniotic infection/inflammation.17 The current study verifies our previous findings. Monocyte chemotactic protein-1 in both compartments was strongly associated with the interval between sampling and delivery within 7 days, and the level of monocyte chemotactic protein-1 in cervical fluid is part of both of our prediction models. This chemokine is produced by monocytes/macrophages, fibroblasts, B cells, endothelial cells, and smooth muscle cells and is a proinflammatory cytokine chemotactic for monocytes. It regulates cytokine production in monocytes and the expression of adhesion molecules on macrophages and augments cytostatic activity. Monocyte chemotactic protein-1 is enhanced during inflammation and infection and induces basophilic histamine release and promotes Th2 immunity.18 A study by Esplin showed that concentrations of amniotic fluid monocyte chemotactic protein-1 were raised in women in preterm labor both with and without intrauterine infection/inflammation.19 Their findings suggest that monocyte chemotactic protein-1 may play a role in preterm labor regardless of the presence of intrauterine infection/inflammation.

Cervical length assessed using transvaginal ultrasonography has been shown to be a marker of spontaneous preterm delivery in several studies. Cervical length also contributed to prediction in our noninvasive prediction model.

Back to Top | Article Outline

CONCLUSIONS

The x-MAP technology enabled us to study a panel of several specific proteins simultaneously using only small sample volumes. In our study population of symptomatic women in preterm labor, we found two different prediction models that might be useful to clinicians in discriminating between women with symptoms who will deliver preterm and women who will not. The reasonably high positive predictive value and negative predictive value means that both women at considerable risk of delivering preterm and who may be targeted for treatment and women at low risk who could be sent home can be identified. On the market today, there are tests with a high negative predictive value of preterm delivery that makes it possible to refrain from hospitalization, bed rest, and repeated monitoring, but there is not one available with a high positive predictive value. A test with a high positive predictive value could assist in, for example, the timing of corticosteroid administration. Furthermore, improved tests with high positive predictive values could help to target novel treatment (eg, progesterone) to high-risk groups in clinical trials. Selection bias cannot be ruled out because inclusions were made only during daytime.

Back to Top | Article Outline

REFERENCES

1. Ananth CV, Vintzileos AM. Epidemiology of preterm birth and its clinical subtypes. J Matern Fetal Neonatal Med 2006;19:773–82.
2. Morken NH, Kallen K, Hagberg H, Jacobsson B. Preterm birth in Sweden 1973–2001: rate, subgroups, and effect of changing patterns in multiple births, maternal age, and smoking. Acta Obstet Gynecol Scand 2005;84:558–65.
3. Slattery MM, Morrison JJ. Preterm delivery. Lancet 2002;360:1489–97.
4. Hack M, Fanaroff AA. Outcomes of children of extremely low birthweight and gestational age in the 1990’s. Early Hum Dev 1999 Jan;53:193–218.
5. McCormick MC. The contribution of low birth weight to infant mortality and childhood morbidity. N Engl J Med 1985;312:82–90.
6. Goldenberg RL, Goepfert AR, Ramsey PS. Biochemical markers for the prediction of preterm birth. Am J Obstet Gynecol. 2005;192:S36–46.
7. Romero R, Espinoza J, Kusanovic JP, Gotsch F, Hassan S, Erez O, et al. The preterm parturition syndrome. BJOG 2006;113:17–42.
8. Vogel I, Goepfert AR, Thorsen P, Skogstrand K, Hougaard DM, Curry AH, et al. Early second-trimester inflammatory markers and short cervical length and the risk of recurrent preterm birth. J Reprod Immunol 2007;75:133–40.
9. Skogstrand K, Thorsen P, Norgaard-Pedersen B, Schendel DE, Sorensen LC, Hougaard DM. Simultaneous measurement of 25 inflammatory markers and neurotrophins in neonatal dried blood spots by immunoassay with xMAP technology. Clin Chem 2005;51:1854–66.
10. Jacobsson B, Mattsby-Baltzer I, Andersch B, Bokstrom H, Holst RM, Wennerholm UB, et al. Microbial invasion and cytokine response in amniotic fluid in a Swedish population of women in preterm labor. Acta Obstet Gynecol Scand 2003;82:120–8.
11. Rose-John S, Waetzig GH, Scheller J, Grotzinger J, Seegert D. The IL-6/sIL-6R complex as a novel target for therapeutic approaches. Expert Opin Ther Targets 2007;11:613–24.
12. Grenache DG, Hankins K, Parvin CA, Gronowski AM. Cervicovaginal interleukin-6, tumor necrosis factor-alpha, and interleukin-2 receptor as markers of preterm delivery. Clin Chem 2004 Oct;50:1839–42.
13. Holst RM, Mattsby-Baltzer I, Wennerholm UB, Hagberg H, Jacobsson B. Interleukin-6 and interleukin-8 in cervical fluid in a population of Swedish women in preterm labor: relationship to microbial invasion of the amniotic fluid, intra-amniotic inflammation, and preterm delivery. Acta Obstet Gynecol Scand 2005;84:551–7.
14. Dudley DJ, Hunter C, Mitchell MD, Varner MW. Elevations of amniotic fluid macrophage inflammatory protein-1 alpha concentrations in women during term and preterm labor. Obstet Gynecol 1996;87:94–8.
15. Romero R, Gomez R, Galasso M, Munoz H, Acosta L, Yoon BH, et al. Macrophage inflammatory protein-1 alpha in term and preterm parturition: effect of microbial invasion of the amniotic cavity. Am J Reprod Immunol 1994;32:108–13.
16. Dudley DJ, Spencer S, Edwin S, Mitchell MD. Regulation of human decidual cell macrophage inflammatory protein-1 alpha (MIP-1 alpha) production by inflammatory cytokines. Am J Reprod Immunol 1995;34:231–5.
17. Jacobsson B, Holst RM, Wennerholm UB, Andersson B, Lilja H, Hagberg H. Monocyte chemotactic protein-1 in cervical and amniotic fluid: relationship to microbial invasion of the amniotic cavity, intra-amniotic inflammation, and preterm delivery. Am J Obstet Gynecol 2003;189:1161–7.
18. Borish LC, Steinke JW. 2. Cytokines and chemokines. J Allergy Clin Immunol. 2003;111:S460–75.
19. Esplin MS, Romero R, Chaiworapongsa T, Kim YM, Edwin S, Gomez R, et al. Monocyte chemotactic protein-1 is increased in the amniotic fluid of women who deliver preterm in the presence or absence of intra-amniotic infection. J Matern Fetal Neonatal Med 2005;17:365–73.
Figure

Figure

© 2009 by The American College of Obstetricians and Gynecologists. Published by Wolters Kluwer Health, Inc. All rights reserved.