Home > DARE Reviews > Accuracy of fibronectin tests for the...

PubMed Health. A service of the National Library of Medicine, National Institutes of Health.

Database of Abstracts of Reviews of Effects (DARE): Quality-assessed Reviews [Internet]. York (UK): Centre for Reviews and Dissemination (UK); 1995-.

Database of Abstracts of Reviews of Effects (DARE): Quality-assessed Reviews [Internet].

Accuracy of fibronectin tests for the prediction of pre-eclampsia: a systematic review

MM Leeflang, JS Cnossen, JA van der Post, BW Mol, KS Khan, and G ter Riet.

Review published: 2007.

CRD summary

The review assessed the accuracy of plasma fibronectin in predicting pre-eclampsia. It was generally well-conducted and well-reported, but the available data were limited and the index test, diagnostic threshold and definition of pre-eclampsia varied between the studies. The authors concluded that fibronectin shows promise but more research is required; this is a reasonable conclusion for the limited data available.

Authors' objectives

To determine the accuracy of fibronectin tests in predicting early pre-eclampsia.

Searching

MEDLINE (1953 to 2004), EMBASE (1980 to 2004), MEDION (1974 to 2004) and the Cochrane Library (Issue 3, 2004) were searched to identify relevant studies. The searches were updated in April 2006. The reference lists of reviews and studies included in the review were screened for additional articles. The search strategy included pre-eclampsia related terms and methodological filters and is available, in full, from the authors. No language restrictions were applied.

Study selection

Studies evaluating the accuracy of plasma fibronectin, measured before 25 weeks' gestation in pregnant women, in predicting pre-eclampsia were eligible for inclusion. The mean age of women in the included studies ranged from 19 to 31 years. In the included studies, the plasma fibronectin threshold used to define a positive test result ranged from 2.8 to 5.0 μg/mL for cellular fibronectin and from 293 to 370 μg/mL for total fibronectin. The occurrence of pre-eclampsia was treated as the reference standard and studies which did not discriminate between pre-eclampsia and other causes of gestational hypertension were excluded; the definition of pre-eclampsia varied across the included studies (details reported).The included studies were also required to report sufficient data to construct a 2×2 contingency table of test result versus reference standard.

Studies were selected for inclusion using two-stage screening of all retrieved references (one reviewer screening for studies on the prediction of pre-eclampsia and a second reviewer screening for fibronectin). Two reviewers independently screened all retrieved manuscripts, with any disagreements by consensus or through consultation with a third reviewer.

Assessment of study quality

The methodological quality of the included studies was assessed using the Quality Assessment of Diagnostic Accuracy Studies (QUADAS) tool, with topic-specific adaptations.

Two reviewers independently performed the quality assessment, with any disagreements resolved by consensus.

Data extraction

Data to populate 2×2 contingency tables were extracted. The sensitivity, specificity, and positive and negative likelihood ratios were calculated for each study.

Two reviewers independently extracted the data, with any disagreements resolved by consensus.

Methods of synthesis

The distribution of sensitivities and specificities, derived from the included studies, was examined by plotting the results in receiver operating characteristic (ROC) space. Meta-analyses were not conducted because of the observed heterogeneity; a narrative synthesis was presented.

Results of the review

Twelve studies were included in the review, of which five (with a total of 573 participants) reported sufficient data to calculate the accuracy of fibronectin. Three of these studies, which evaluated total fibronectin, were diagnostic cohorts and two, which evaluated cellular fibronectin, were matched case-controls. The studies reported multiple data sets for different gestation periods and fibronectin thresholds and fractions.

The quality assessment was based on 11 studies. All provided over 90% verification of diagnosis. Most of the studies provided an adequate description of the selection criteria, index test and reference standard. The studies were poorly compliant with other quality criteria; all were met by less than 50% of studies.

Where reported, the incidence of pre-eclampsia across all 12 studies ranged from 2.6 to 19.3%; in the studies reporting sufficient data to calculate accuracy, it ranged from 4.5 to 11.5%.

For cellular fibronectin (2 studies, 14 data sets), the sensitivity ranged from 50% (specificity 88% and 96%) to 100% (specificity 75%), and the specificity ranged from 72% (sensitivity 74%) to 96% (sensitivity 50%).

For total fibronectin (4 studies, 9 data sets), the sensitivity ranged from 50% (specificity 75%) to 83% (specificity 63%), and the specificity ranged from 43% (sensitivity 70%) to 94% (sensitivity 65%).

Authors' conclusions

Cellular fibronectin shows potential as a tool for the prediction of pre-eclampsia. Total fibronectin appears to produce more variable results. More high-quality and well-reported studies are needed to populate decision-analytic models.

CRD commentary

The review addressed a clearly stated research question and, although inclusion criteria were not explicitly stated, reasons for exclusion were given in a flow chart of the review process. A reasonably wide search, without language restrictions, was conducted in order to identify relevant primary studies. Methodological filters were included in the search, a strategy which is known to be problematic for studies of diagnostic tests and which can lead to the loss of relevant data. The review methodology included appropriate measures to minimise error and bias, and the methodological quality of the included studies was addressed using relevant topic- and study design-specific criteria. The decision not to undertake meta-analyses was appropriate given the differences in the study populations, index tests and thresholds, and definitions of pre-eclampsia used. Relevant data from the included studies were presented clearly and summarised. The authors' conclusions are appropriately cautious given the limited data available.

Implications of the review for practice and research

Practice: There were no recommendations for current practice.

Research. A formal decision analysis is required to quantify the role of fibronectin as an 'add-on' to clinical information usually available at the point of testing. More high-quality studies are required to populate decision-analytic models and thus to determine whether the accuracy of this test is sufficient to be clinically relevant.

Funding

UK NHS Health Technology Assessment (HTA) Programme, project number 01/64/04.

Bibliographic details

Leeflang M M, Cnossen J S, van der Post J A, Mol B W, Khan K S, ter Riet G. Accuracy of fibronectin tests for the prediction of pre-eclampsia: a systematic review. European Journal of Obstetrics and Gynecology and Reproductive Biology 2007; 133(1): 12-19. [PubMed: 17293022]

Indexing Status

Subject indexing assigned by NLM

MeSH

Biological Markers; Diagnostic Techniques, Obstetrical and Gynecological; Female; Fibronectins /blood; Pre-Eclampsia /diagnosis; Pregnancy; Sensitivity and Specificity

AccessionNumber

12007002425

Database entry date

03/11/2008

Record Status

This is a critical abstract of a systematic review that meets the criteria for inclusion on DARE. Each critical abstract contains a brief summary of the review methods, results and conclusions followed by a detailed critical assessment on the reliability of the review and the conclusions drawn.

CRD has determined that this article meets the DARE scientific quality criteria for a systematic review.

Copyright © 2014 University of York.

PMID: 17293022

Download

PubMed Health Blog...

read all...

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...