The diagnostic accuracy of strategies used to identify early pregnancy: a systematic review.

Review question/objective The objective of this review is to summarise and synthesise the evidence on the diagnostic accuracy of screening strategies used to identify early pregnancy in patients in the health care setting More specifically, the objectives are to identify the diagnostic accuracy of: 1. patient history to identify early pregnancy; including verbal questioning, written questionnaires and use of last menstrual period dates, and 2. pregnancy testing in identifying early pregnancy. Background Diagnosing or excluding early pregnancy is an important step in the evaluation of women of child-bearing age in primary care and emergency departments. Early pregnancy can be defined as the period known as organogenesis which last from weeks two to eight post conception1, 2. The foetus is highly sensitive to the harmful effects of teratogenic drugs and ionising radiation in the early stages of pregnancy so it is often necessary to determine a patient's pregnancy status prior to performing surgical and diagnostic procedures or prescribing treatment34. This allows for appropriate clinical management and prenatal care to commence. Preoperative pregnancy testing has been a topic of interest in anaesthesiology for some time5. Although the possible teratogenic and abortive effects of commonly used anaesthesia are widely acknowledged, routine pregnancy testing prior to surgery has not been mandated. The WHO Guidelines for Safe Surgery 2009 do not include information regarding pre-operative pregnancy testing. Routine pregnancy testing prior to elective surgery is also not an American Society of Anaesthesiologists standard. However, clinical guidelines for preoperative testing recommend pregnancy checking or pregnancy testing for women prior to surgery4, 6. An abstract from Euroanaesthesia 2011 reported that pregnancy checking should be performed preoperatively and that a “clear definition of what makes an adequate preoperative pregnancy status check would help in assessment and achievement of standards as well as reducing interhospital variability”7. The developing foetus is most sensitive to the biological effects from exposure to ionising radiation in the early period of pregnancy known as organogenesis which occurs during weeks two to eight after conception. Exposure to ionising radiation from medical imaging procedures in this early stage of pregnancy may induce teratogenic, mutagenic or carcinogenic changes. The type and extent of damage that may occur is dependent on factors such as the total dose, dose rate, type of radiation used, and the stage of foetal development. Women are often unaware they are pregnant in the early stage of pregnancy. In order to protect the foetus the regulatory and professional bodies involved with the use of ionising radiation for medical imaging recommend checking patient pregnancy status prior to performing procedures using ionising radiation8. These documents do not provide definitions for “child-bearing age” or guidelines on how to determine the pregnancy status of the patient. Certain medications can induce adverse effects on a foetus if taken by the mother in the early stage of pregnancy. Isotretinoin is an effective acne treatment widely prescribed for severe cystic acne in Australia and the UK. As a member of the retinoid family, it is a known teratogen causing foetal abnormalities in rodent and primate models9. Craniofacial deformities, cardiac and eye abnormalities and spontaneous abortion are some of the major adverse effects of ingestion of isotretinoin during pregnancy. The Australian College of Dermatologists recommend that prior to commencing treatment, all women of child bearing age should be undergo a pregnancy test and be counselled regarding pregnancy, contraception and the possible dangers of falling pregnant during treatment. Other known teratogenic medications, such as ACE (angiotensin converting enzyme) inhibitors, high dose Vitamin A, some anticonvulsant medications, some antibiotics, chemotherapy drugs, lithium and warfarin can adversely affect the health of the foetus so pregnancy checking is recommended before treatment10. There is a perception that the ability of the patient to self-diagnose pregnancy is unreliable. Without a clinical examination or pregnancy test, explicit questions regarding the patient's sexual activity, menstrual history and use of contraceptives must be undertaken. In some cases, such as young teenagers, verbal questioning may not illicit a truthful or accurate response, particularly if a parent is present at the time of questioning11. Ramoska (1989)12 suggested that patient history was unreliable in diagnosing or excluding early pregnancy. However later studies performed by Strote and Chen (2006)13 and Minnerop et al (2011)14 reported that sexual history and self-assessment were effective predictors of pregnancy. All studies recommended the liberal use of a laboratory pregnancy test in most cases. Home pregnancy testing (HPT) kits have been found to be ineffective in identifying early pregnancy, especially if used prior to the date of missed menses. Manufacturers of HPT claim a greater than 97% accuracy in the detection of pregnancy however, when performed in the home, the real accuracy may be as low as 77%15. In particular, there is a high rate of false-negatives reported in early pregnancy. There are many reported possible causes for this including testing prior to the date of missed menses, tests with differing detectable concentrations of human chorionic gonadotrophin (HCG), operator error, and dilution of the urine sample16, 17. Pregnancy tests (both urine and serum) performed in the laboratory have been found to be highly accurate in determining pregnancy during early pregnancy. However, it is impractical and expensive to perform serum pregnancy testing on all female patients of child-bearing age presenting for medical imaging or elective surgery. A preliminary search performed of the JBI Library of Systematic Reviews and Cochrane Library did not find any systematic reviews regarding diagnosis of early pregnancy. A review article by Bastian and Piscitelli15 looked at articles published between 1966 and 1996 concerning the diagnosis of pregnancy focussing on the use of patient history and clinical examination to rule out early pregnancy. Their search identified nine studies and they used the findings from these to create likelihood ratios to predict the likelihood of pregnancy in certain situations. The review recommended laboratory pregnancy testing be used, rather than clinical history or examination, to accurately rule out pregnancy prior to any treatment or procedure that could adversely affect a foetus. Inclusion criteria Types of participants This review will consider studies that include adolescent and adult female patients in the health care setting of childbearing age. Focus of the review The focus of the review is the correct identification of early pregnancy. This review will consider studies that compare the diagnostic accuracy of screening strategies to identify early pregnancy compared to HCG pregnancy test. Types of outcomes This review will consider studies that report on the sensitivity, specificity and accuracy of medical history and/or physical examination (index test), and urine or serum HCG pregnancy test (reference test) to identify early stage pregnancy. Likelihood ratios and/or predictive values will be generated to further describe the diagnostic test accuracy. Types of studies This review will consider any quantitative studies that examine the diagnostic accuracy of the index and/or reference tests. Search strategy The search strategy aims to find both published and unpublished studies. A three-step search strategy will be utilised in this review. An initial limited search of MEDLINE and CINAHL will be undertaken followed by analysis of the text words contained in the title and abstract, and of the index terms used to describe article. A second search using all identified keywords and index terms will then be undertaken across all included databases. Thirdly, the reference list of all identified reports and articles will be searched for additional studies. Studies published in English will be considered for inclusion in this review. Studies published from January 1975 will be considered for inclusion in this review because monoclonal antibody pregnancy assays were introduced during that year18. The databases to be searched include: MEDLINE, CINAHL, EMBASE, SCOPUS, Web of Science, Cochrane Library The search for unpublished studies will include ProQuest, NOVA, TROVE, MedNar Initial keywords to be used will be: Female Humans Adolescent All adult Pregnancy tests Pregnancy Medical history Physical Examination Preoperative care Diagnosis Diagnostic accuracy Sensitivity Specificity Likelihood ratio Predictive value Assessment of methodological quality Papers selected for retrieval will be assessed by two independent reviewers for methodological quality prior to inclusion in the review. The Quality Assessment of Diagnostic Accuracy studies (QUADAS)19 checklist will be used as an appraisal instrument. The checklist is presented in Appendix I. Any disagreements that arise between the reviewers will be resolved through discussion, or with a third reviewer. Data collection Data will be extracted from papers included in the review using the Standards for Reporting STudies of Diagnostic Accuracy (STARD)20 checklist (Appendix II). The data extracted will include specific details about the populations, setting, study methods, sensitivity, specificity and accuracy of the index test as compared with serum HCG pregnancy test results. Data synthesis The reported sensitivity, specificity and accuracy from individual studies will be tabulated. Predictive values and likelihood ratios will be calculated for each included study. Where statistical pooling is not possible the findings will be presented in narrative form including tables and figures to aid in data presentation where appropriate. Revman 5.1 and Microsoft Excel software will be used for data management and analysis. Conflicts of interest Nil Acknowledgements As this systematic review will form part of a PhD in Medical Radiation Science thesis, the secondary reviewer will be used for critical appraisal of studies only.