Home > DARE Reviews > Are patient characteristics associated...

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].

Are patient characteristics associated with the accuracy of hysterosalpingography in diagnosing tubal pathology? An individual patient data meta-analysis

KA Broeze, BC Opmeer, N Van Geloven, SF Coppus, JA Collins, JE Den Hartog, PJ Van der Linden, P Marianowski, EH Ng, JW Van der Steeg, P Steures, A Strandell, F Van der Veen, and BW Mol.

Review published: 2011.

Link to full article: [Journal publisher]

CRD summary

This individual patient data (IPD) review based on seven studies concluded that hysterosalpingography was a useful tubal patency screening test for all infertile couples irrespective of their characteristics. This conclusion cannot be considered reliable as 90% of eligible trials were excluded, analytical methods were not described and test sensitivity was low.

Authors' objectives

To assess whether the diagnostic accuracy of hysterosalpingography was associated with different patient characteristics.


MEDLINE and EMBASE were searched from July 1994 to January 2010 to update an existing meta-analysis (see other publications of related interest). References were also searched and authors were contacted for additional material. Language restrictions were not applied and search terms were reported.

Study selection

To be eligible for inclusion, individual patient data (IPD) must have been available including patient characteristics, hysterosalpingography results and the results of diagnostic laparoscopy (tubal pathology absent or present). Authors were asked to indicate whether tubal pathology was unilateral or bilateral (definitions presented). Hysterosalpingography was used in all trials for tubal patency testing

Fifth and 95th age percentiles ranged from 23 to 40 years, with median subfertility from 1.4 to three years. Bilateral tubal pathology prevalence ranged from nine to 21%, with any tubal pathology prevalence from 12 to 38%.

Two independent reviewers screened studies for eligibility.

Assessment of study quality

The QUADAS quality assessment tool was used to assess methodological quality. The consistency of IPD was checked with comparison to published results and authors were contacted to address missing data and discordance.

The number of reviewers assessing study quality was unreported.

Data extraction

IPD were obtained from trials that reported sensitivity and specificity. IPD were available for seven of 71 trials. Summary data was not extracted from the 64 relevant trials where IPD was unavailable. Missing covariate data was handled using multiple imputation. Authors who provided IPD were also asked to provide any additional unpublished information.

The number of reviewers extracting data was unreported.

Methods of synthesis

Sensitivities and specificities were pooled using an unreported method. The impact of patient characteristics were explored by subgroup analysis and using an unspecified logistic regression approach. The impact of multiple imputation was assessed as a sensitivity analysis.

Results of the review

Seven trials (4,521 women) were available. Trial quality was considered sufficient by the authors, but most studies did not avoid partial or differential verification.

Hysterosalpingography had a sensitivity of 53% (95% CI 50 to 57) and specificity of 87% (95% CI 86 to 88) for diagnosing tubal pathology and 46% (95% CI 41 to 51) and 95% (95% CI 0.94 to 0.95) for bilateral tubal pathology.

"Low-risk clinical history" increased sensitivity from 38% to 61% for any tubal pathology (p=0.005), and from 13% to 47% for bilateral tubal pathology (p=0.01) but other factors (age, duration, body mass index, type (primary/secondary), HxPID, and CAT) were not significantly associated.

Results were sensitive to imputation with sensitivities of 70% (95% CI 66 to 74) and 66% (95% CI 55 to 75) for tubal pathology prior to imputation. Of 4,521 data points, 2,632 were imputed as these women did not have laparoscopy data

Authors' conclusions

Hysterosalpingography was a useful tubal patency screening test for all infertile couples irrespective of their characteristics

CRD commentary

The search and inclusion criteria were appropriate to identify eligible trials and avoid selection biases. Availability bias could have had potentially large effects as IPD was available for only seven of 71 eligible trials. The quality of available evidence also raised some concerns, particularly about missing data. The difference in results between complete case analysis and inference after multiple imputation made the results sensitive to violations of the assumptions underlying the imputation. Lack of clarity on methods for statistical synthesis added to uncertainty about reliability of the results.

The conclusion that hysterosalpingography was a useful test relies on the additional assumption that moderate specificity was useful even where sensitivity was barely better than random chance. Therefore, there were considerable potential biases introduced and the conclusions did not fully reflect the evidence. The reliability of the conclusion that different subgroups had similar sensitivity and specificities was unclear because of availability bias, the sensitivity of results to imputation and lack of clarity regarding analytical method. These uncertainties impinge on the conclusion that the test was useful. This uncertainty extends to the assumption that a test that identifies true positives no better than chance has utility. Therefore, this conclusion is unreliable.

Implications of the review for practice and research

Practice: The author's stated that hysterosalpingography could be used as a screening test for all infertile couples.

Research: The author's did not state any implications for research



Bibliographic details

Broeze KA, Opmeer BC, Van Geloven N, Coppus SF, Collins JA, Den Hartog JE, Van der Linden PJ, Marianowski P, Ng EH, Van der Steeg JW, Steures P, Strandell A, Van der Veen F, Mol BW. Are patient characteristics associated with the accuracy of hysterosalpingography in diagnosing tubal pathology? An individual patient data meta-analysis Human Reproduction Update 2011; 17(3): 293-300. [PubMed: 21147835]

Other publications of related interest

Swart P, Mol B W, van der Veen F, van Beurden M, Redekop W K, Bossuyt P M. The accuracy of hysterosalpingography in the diagnosis of tubal pathology: a meta-analysis. Fertility and Sterility 1995; 64(3): 486-491.

Indexing Status

Subject indexing assigned by NLM


Adult; Age Factors; Fallopian Tube Diseases /radiography; Fallopian Tube Patency Tests /methods; Female; Humans; Hysterosalpingography /methods; Infertility, Female /radiography; Regression Analysis; Risk Factors; Sensitivity and Specificity; Time Factors



Database entry date


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: 21147835

PubMed Health Blog...

read all...

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...