Evaluation of the quantitative fetal fibronectin test and PAMG-1 test for the prediction of spontaneous preterm birth in patients with signs and symptoms suggestive of preterm labor

Preterm labor with intact membranes: a simple noninvasive method to identify patients at risk for intra-amniotic infection and/or inflammation

OBJECTIVE

To develop a noninvasive scoring system to identify patients at high risk for intra-amniotic infection and/or inflammation, which would reduce the need for amniocentesis.

METHODS

This prospective cohort study comprised patients admitted with preterm labor and intact membranes (20-34 weeks of gestation) who underwent a transabdominal amniocentesis and for whom concentrations of quantitative cervical fetal fibronectin and of maternal serum C-reactive protein (CRP) were determined. Intra-amniotic infection was defined as a positive amniotic fluid culture for microorganisms. Intra-amniotic inflammation was defined as an amniotic fluid matrix metalloproteinase-8 concentration >23 ng/mL. Multivariate logistic regression analysis was performed to identify intra-amniotic infection/inflammtion with noninvasive parameters that had a significant relationship with univariate analysis. With four parameters identified by multivariate analysis, we generated a noninvasive risk scoring system.

RESULTS

Of the study population consisting of 138 singleton pregnant women, (1) the overall rate of intra-amniotic infection/inflammation was 28.3% (39/138); (2) four parameters were used to develop a noninvasive risk scoring system [i.e. cervical fetal fibronectin concentration (score 0-2), maternal serum CRP concentration (score 0-2), cervical dilatation (score 0-2), and gestational age at presentation (score 0-1)]; the total score ranges from 0 to 7; 3) the area under the curve of the risk score was 0.96 (95% confidence interval (CI), 0.92-0.99), significantly higher than that of each predictor in the identification of intra-amniotic infection/inflammation (p < .001, for all); 4) the risk score with a cutoff of 4 had a sensitivity of 94.9% (37/39), a specificity of 90.9% (90/99), a positive predictive value of 80.4% (37/46), a negative predictive value of 97.8% (90/92), a positive likelihood ratio of 10.4 (95% CI, 5.6-19.5), and a negative likelihood ratio of 0.06 (95% CI, 0.15-0.22) in the identification of intra-amniotic infection/inflammation.

CONCLUSIONS

(1) The combination of four parameters (concentrations of cervical fetal fibronectin and maternal serum CRP, cervical dilatation, and gestational age) was independently associated with intra-amniotic infection and/or inflammation; and (2) the risk scoring system comprised of the combination of 4 noninvasive parameters was sensitive and specific to identify the patients at risk for intra-amniotic infection and/or inflammation.

AACC Guidance Document on Laboratory Testing for the Assessment of Preterm Delivery

Identifying women with preterm labor who will go on to deliver prematurely is crucial to improving outcomes for mother and baby and for saving healthcare resources. Even among those with symptoms, the number of women who deliver preterm is low, and thus the low positive predictive value (PPV) and high negative predictive value (NPV) associated with available biomarkers does not substantially reduce the uncertainty of the clinical diagnosis. While there is some promise in the use of fetal fibronectin (fFN), interleukin 6 (IL-6), or placental alpha microglobulin 1 (PAMG-1) for predicting preterm birth (PTB), their use is unlikely to provide considerable clinical value in populations with a low prevalence. To provide real clinical benefit, a biomarker must demonstrate a high PPV to allow identification of the minority of symptomatic women who will deliver prematurely. As none of the currently available biomarkers exhibit this performance characteristic, we do not recommend their routine clinical use in populations with a pre-test probability of PTB of <5%. Limiting biomarker testing to only high-risk women identified on the basis of cervical length or other characteristics will increase the pre-testprobability in the tested population, thereby improving PPV. PAMG-1 is associated with a higher PPV than fFN and may show clinical utility in populations with a higher pre-test probability, but further work is required to conclusively demonstrate improved outcomes in this patient group.

Machine Learning Methods for Preterm Birth Prediction: A Review

Preterm births affect around 15 million children a year worldwide. Current medical efforts focus on mitigating the effects of prematurity, not on preventing it. Diagnostic methods are based on parent traits and transvaginal ultrasound, during which the length of the cervix is examined. Approximately 30% of preterm births are not correctly predicted due to the complexity of this process and its subjective assessment. Based on recent research, there is hope that machine learning can be a helpful tool to support the diagnosis of preterm births. The objective of this study is to present various machine learning algorithms applied to preterm birth prediction. The wide spectrum of analysed data sets is the advantage of this survey. They range from electrohysterogram signals through electronic health records to transvaginal ultrasounds. Reviews of works on preterm birth already exist; however, this is the first review that includes works that are based on a transvaginal ultrasound examination. In this work, we present a critical appraisal of popular methods that have employed machine learning methods for preterm birth prediction. Moreover, we summarise the most common challenges incurred and discuss their possible application in the future.

Serum apolipoprotein A-II and alpha-2-antiplasmin levels in midtrimester can be used as predictors of preterm delivery

OBJECTIVE

To identify serum protein markers in midtrimester that predict preterm delivery.

METHODS

A retrospective case-control study randomly selected patients that experienced spontaneous preterm birth and healthy control patients that experienced a normal delivery at term. A proteomic analysis was undertaken using the data-independent acquisition method.

RESULTS

A total of 30 singleton pregnant women were randomly selected from 12 800 pregnant women: 15 women had a spontaneous preterm birth (group Y) and 15 age- and body mass index-matched women gave birth at term (group D). All of the patients provided serum at 15-20 weeks of gestation. A total of 39 differentially expressed proteins were identified. Compared with group D, 24 proteins were upregulated and 15 were downregulated in the preterm group Y. Using Kyoto Encyclopedia of Genes and Genomes pathway enrichment, the 24 upregulated proteins were significantly enriched in the complement and coagulation cascade pathways. Search Tool for the Retrieval of Interacting Genes Furthermore (STRING) analysis showed that apolipoprotein A-II (apoA-II) and alpha-2-antiplasmin (α2-AP), two upregulated proteins, were key nodes in the STRING protein-protein network.

CONCLUSIONS

These findings suggest that apoA-II and α2-AP might be new markers for predicting preterm delivery in the midtrimester.

Evaluation of placental alpha microglobulin-1(PAMG1) accuracy for prediction of preterm delivery in women with the symptoms of spontaneous preterm labor; a comparison with cervical length and number of contractions

Introduction: Since the capacity of tertiary centers is limited, identifying women with the risk of preterm delivery is crucial amongst women who refer to hospitals with threatened preterm labor.Objective: The purpose of the present study was to evaluate the accuracy of the placental alpha microglobulin-1 (PAMG-1) test for identifying women who referred to hospitals with signs of spontaneous preterm labor and ultimately delivered within ≤48 h and ≤7 of testing and ≤37 weeks of gestation and its comparison with cervical length of <25 mm and number of contractions.Materials and methods: A prospective observational study was performed on women with the signs of spontaneous preterm labor. Placental alpha microglobulin-1 (PAMG-1) was evaluated in vaginal secretion. Cervical length (CL) and number of contractions were evaluated and recorded. The test to- spontaneous- delivery interval was documented. Sensitivity (SN), specificity (SP), accuracy rate of the PAMG-1, cervical length of <25 mm and number of contractions in prediction of spontaneous preterm delivery within ≤48 h, ≤7 days and preterm delivery (≤37 weeks of gestation) were calculated.Results: One hundred eighty women finished the study. 44 women had positive PAMG-1 test and 58 women had cervical length of <25 mm. Women with a positive PAMG-1 were more likely to deliver within ≤48 h (p < .0001), ≤7 days (p < .0001), and before 37 weeks (p < .0001), compared to the women who had a negative test. For delivery within ≤48 h, ≤7 days and ≤37 weeks, specificity (SP) of the PAMG-1; was statistically higher than cervical length of <25 mm and contractions of 12-17/hr but not contractions of ≥18/hr. PAMG-1 showed a higher accuracy rate than cervical length of <25 mm, contractions of 12-17/hr and contractions of ≥18/hr for deliveries within ≤48 h (78.9, 55.3, 48.9, and 69.44%, respectively) and ≤7 days (76.1, 55.32, 55, and 65.56%, respectively). For deliveries before 37 weeks; the PAMG-1 test showed higher LR + than CL of <25 mm, contractions of 12-17/hr and contractions of ≥18/hr [10.24 (2.57-40.86), 2.01 (1.24-3.23), 1.30 (1.05-1.62), and 5.12 (1.24-21.11), respectively]. PPV for deliveries within ≤48 h and ≤7 days for PAMG-1 was greater than cervical length of <25 mm and number of contractions, however, NPV did not show a significant difference.Conclusion: PAMG-1 test showed a higher accuracy rate for prediction of delivery within ≤48 and ≤7 days in comparison with cervical length of <25 mm and number of contractions. Also, PAMG-1 test had a higher positive likelihood ratio for deliveries at ≤37 weeks. PPV for deliveries within ≤48 h and ≤7 days for PAMG-1 was greater than cervical length of <25 mm and number of contractions, however, NPV did not show a significant difference.

Placental alpha microglobulin-1 (PartoSure) test for the prediction of preterm birth: a systematic review and meta-analysis

OBJECTIVES

To assess the accuracy of the placental alpha microglobulin-1 (PAMG-1) to predict preterm birth (PB) in women with symptoms of PB through use of formal methods for systematic reviews and meta-analytic techniques.

METHODS

We performed a comprehensive search of medical bibliographic databases to identify observational studies that reported on the predictive accuracy of PAMG-1 for PB. Two investigators independently assessed studies, assessed quality of studies, and extracted data. Summary receiver operating characteristic (SROC) curves, pooled sensitivities, specificities, likelihood ratios (LR), and diagnostic odds ratio (DOR) were generated.

RESULTS

Seventeen studies involving 2590 women met the inclusion criteria. Meta-analysis of 15 studies (including 1906 women) revealed a pooled sensitivity of 66.2% (95% CI: 59.1, 72.7) and specificity of 96.1% (95% CI: 95.1, 97.0) with the SROC equal to 0.97 (95% CI: 0.95, 0.98) for prediction of delivery within 7 d of testing. The summary estimates were 15.26 (95% CI: 11.80, 19.75) for LR + and 0.31 (95% CI: 0.17, 0.55) for LR - for prediction of delivery within 7 d of testing. Pooled estimate of DOR for predicting delivery within 7 d of testing was 55.13 (95% CI: 35.32, 86.06). The sensitivity, specificity and the SROC of PAMG-1 pooled from 10 studies (including 1508 women) for prediction of delivery within 14 d of testing were 64.4% (95% CI: 56.8, 71.5), 96.9% (95% CI: 95.8, 97.7) and 0.97 (95% CI: 0.95, 0.98). The overall pooled LR + and LR - of PAMG-1 for predicting delivery within 14 d of testing among the included studies were 16.72 (95% CI: 12.03, 23.23) and 0.42.1 (95% CI: 0.31, 0.56), respectively. The pooled DOR of the PAMG-1 for prediction delivery within 14 d of testing was equal to 44.65 (95% CI: 26.30, 75.78).

CONCLUSION

Cervical PAMG-1 had a high accuracy to predict PB within 7 and 14 d of testing in symptomatic pregnant women.

A high concentration of fetal fibronectin in cervical secretions increases the risk of intra-amniotic infection and inflammation in patients with preterm labor and intact membranes

Objective To determine whether the risk of intra-amniotic infection/inflammation and spontaneous preterm delivery (SPTD) varies as a function of the concentration of cervical fetal fibronectin (fFN) in patients with preterm labor and intact membranes. Methods This prospective study included 180 patients with preterm labor and intact membranes who had a sample collected for quantitative fFN measurement and underwent amniocentesis. Amniotic fluid was cultured for aerobic and anaerobic bacteria and genital mycoplasmas. Intra-amniotic inflammation was defined as an amniotic fluid matrix metalloproteinase-8 concentration >23 ng/mL. Results (1) The prevalence of intra-amniotic infection/inflammation and SPTD within 7 days was 32.2% (58/180) and 33.9% (61/178), respectively; (2) The higher the fFN concentration, the greater the risk of intra-amniotic infection/inflammation and SPTD within 7 days (P<0.001, respectively); (3) An fFN concentration 150 ng/mL had a better diagnostic performance than an fFN 50 ng/mL in the identification of intra-amniotic infection/inflammation and SPTD within 7 days; (4) Among the patients with an fFN <50 ng/mL, intra-amniotic infection/inflammation was identified in 7.6% (6/79) of patients and 66.7% (4/6) delivered within 7 days. Conclusion The higher the concentration of fFN, the greater the risk of intra-amniotic infection/inflammation and SPTD in patients with preterm labor and intact membranes.

Prediction of spontaneous preterm delivery in women presenting with premature labor: a comparison of placenta alpha microglobulin-1, phosphorylated insulin-like growth factor binding protein-1, and cervical length

BACKGROUND

Placental alpha microglobulin-1 and phosphorylated insulin-like growth factor-binding protein-1 have been studied in patients at risk for preterm birth with signs and symptoms of preterm labor. However, a direct comparison between these 2 biomarkers, alone or in combination with cervical length measurement with an adequate sample size, has been lacking to date.

OBJECTIVE

The purpose of this study was to compare the placental alpha microglobulin-1 test and the phosphorylated insulin-like growth factor-binding protein-1 test alone and in combination with cervical length measurement for the prediction of imminent spontaneous preterm birth of testing in pregnant women with symptoms of preterm labor in a tertiary care setting.

STUDY DESIGN

Four hundred three patients with intact amniotic membranes and cervical dilation ≤3 cm, without recent intercourse or cerclage, between gestational weeks of 20⁺⁰ and 36⁺⁶ were recruited prospectively from 3 international centers. Placental alpha microglobulin-1 and phosphorylated insulin-like growth factor-binding protein-1 tests were conducted before cervical length measurement via transvaginal ultrasound scanning. Caregivers were blinded to the biomarker test results. Medically indicated deliveries within 14 days of testing were excluded. Standard performance statistics with 95% confidence intervals were calculated and compared based on pairwise estimates from a generalized model.

RESULTS

Of 403 subjects who were enrolled in the study cohort, 94% (383/403 women) met the inclusion criteria. Median gestational age and cervical length at presentation were 30⁺⁵ weeks and 27 mm, respectively; 6.8% women (26/383 women) had spontaneous birth ≤7 days from testing. The placental alpha microglobulin-1 test was positive in 7.8% of the women (30/383 women); the phosphorylated insulin-like growth factor-binding protein-1 test was positive in 29.5% women (113/383 women). Positive predictive value for placental alpha microglobulin-1, phosphorylated insulin-like growth factor-binding protein-1, and cervical length <25 mm for the prediction of spontaneous preterm birth in the overall cohort was 60.0% (18/30 women), 18.6% (21/113 women), 11.8% (18/152 women), respectively. The negative predictive value was 97.7% (345/353 women), 98.2% (265/270 women), 96.5% (223/231 women), respectively. The prevalence of spontaneous preterm birth in this group was 6.8% (26/383 women). The positive likelihood ratios were 20.6, 3.1, and 1.8, respectively. The negative likelihood ratio were 0.3, 0.3, and 0.5, respectively. Positive predictive values for placental alpha microglobulin-1 and phosphorylated insulin-like growth factor-binding protein-1 tests in patients with cervical length shortening of 15-30 mm for the prediction of spontaneous preterm birth were 60.9% (14/23 women) and 28.1% (16/57 women), respectively. The negative predictive values were 97.1% (168/173 women) and 97.8% (136/139 women), respectively. The prevalence of spontaneous preterm birth in the 15-30 mm cohort was 9.7% (19/196 women). The positive likelihood ratios were 14.5 and 3.6, respectively. The negative likelihood ratios were 0.3 and 0.2, respectively.

CONCLUSION

Placental alpha microglobulin-1 is significantly more specific than phosphorylated insulin-like growth factor-binding protein-1 for the prediction of spontaneous preterm birth ≤7 days (P<.0001), whereas both tests have comparable sensitivity. In patients with cervical length 15-30 mm, although placental alpha microglobulin-1 has a significantly higher positive predictive value and specificity compared with phosphorylated insulin-like growth factor-binding protein-1 for the prediction of spontaneous preterm birth at ≤7 days (P<.01), both tests have a comparable sensitivity and negative predictive value. In conclusion, placental alpha microglobulin-1 is a better predictor of imminent spontaneous preterm birth when compared with phosphorylated insulin-like growth factor-binding protein-1 alone or in combination with cervical length measurement. In patients with shortening of cervical length of 15-30 mm, the placental alpha microglobulin-1 test is a significantly better predictor of imminent spontaneous preterm birth within 7 days of testing than is phosphorylated insulin-like growth factor-binding protein-1.