Prenatal screening and diagnosis are integral to antenatal care worldwide. Prospective parents are offered screening for common fetal chromosomal and structural congenital malformations. In most developed countries, prenatal screening is routinely offered in a package that includes ultrasound scan of the fetus and the assay in maternal blood of biochemical markers of aneuploidy. Mistakes can arise at any point of the care pathway for fetal screening and diagnosis, and may involve individual or corporate systemic or latent errors. Special clinical circumstances, such as maternal size, fetal position, and multiple pregnancy, contribute to the complexities of prenatal diagnosis and to the chance of error. Clinical interventions may lead to adverse outcomes not caused by operator error. In this review I discuss the scope of the errors in prenatal diagnosis, and highlight strategies for their prevention and diagnosis, as well as identify areas for further research and study to enhance patient safety.
Introduction
‘I shall try to correct errors when shown to be errors, and I shall adopt new views so fast as they shall appear to be true views. (Abraham Lincoln).
An error is the failure of a planned action to be completed as intended, or the use of a wrong plan to achieve an aim. Errors are made in every facet of human endeavour as man is fallible and his exploits are not without mistakes. Errors arise during prenatal care. The diagnosis of fetal conditions, advising parents on these diagnoses, and proffering counselling on the basis of which parents are required to make informed choices about their pregnancies and birth, are complex and often intricate matters. Prenatal diagnosis is also uniquely challenging because of the need to diagnose the condition of a fetus while still in the intrauterine environment.
Mistakes arising during fetal prenatal screening or diagnostic procedures often lead to erroneous counselling, treatment options and parental decisions. Fetal interventions of an invasive nature may be attended by miscarriage, whereas non-intervention caused by errors in fetal screening and diagnosis may result in a profoundly handicapped child, raising the possibility of such legal claims as ‘wrongful birth and wrongful life’. Errors can include problems in practice, products, procedures, and systems. These could be errors of omission (whereby screening and diagnostic services are not offered in situations where they should have been offered), or of commission (whereby screening and diagnostic services are offered in situations where they should have been withheld, or inappropriate intervention suggested in a given situation).
Errors in prenatal diagnosis exact other significant tolls beyond the cost of pregnancy loss and survival with morbidity. These include increased costs for the additional care that may be necessitated by the errors, lost income and household productivity, and disability. Further costs include loss of trust in the healthcare system by patients, and diminished satisfaction by patients and healthcare professionals. Patients who experience disability as a result of errors pay with physical and psychological discomfort, whereas healthcare professionals often lose morale and suffer frustration at not being able to provide the best care possible. Societal costs include those resulting from lost worker productivity and lower levels of population health status.
In this review, I detail the range of errors in prenatal diagnosis, their relative prevalence, the system failures that are associated with the most common varieties, and the future scope for investigating, preventing, and mitigating them.
The nature of clinical errors in prenatal diagnosis
Most medical errors do not result from individual recklessness or the actions of a particular group. It is increasingly recognised that individual or personal clinical errors are often a result of system failures or so called ‘latent errors’— errors waiting to happen. Such errors are caused by faulty systems, processes, and conditions that lead people to make mistakes or fail to prevent them. Such a systems-based approach to error diagnosis and mitigation has led to improved safety in many high risk industries such as the aerospace and nuclear industries where systems redesigns have reduced the occurrence of human error. Application of these concepts to managing clinical risk, including situations in prenatal diagnosis and therapy, is leading to improvements in patient safety.
The types, timing, and domains of medical errors in prenatal diagnosis are summarised in Table 1 . Such errors may have deleterious effects on the fetus, the mother, or both. They may be minor or major, and therefore material or immaterial. Their severity and impact may depend on the gestational age at which they occur, and may be related to birth order, with several errors arising from dealing with multiple pregnancies. Accurate determination of fetal gender is crucial to care in several clinical situations: determining chorionicity in twins, interpreting sex chromosome mosaicism and aneuploidy, and diagnosing sex-linked conditions presenting after 14 weeks. Errors in fetal gender determination can have potentially significant effect on care and pregnancy outcome. Misdiagnosing the side of affectation of fetal anomalies is not often of major significance in utero , as the correct diagnosis can usually be made postnatally. Right-sided congenital diaphragmatic hernias carry a worse prognosis than left-sided lesions, and an error in correctly situating such lesions can significantly affect the parents’ decision to continue with or terminate the pregnancy.
| Error type | Examples |
|---|---|
| Diagnostic | Error or delay in diagnosis of fetal malformation or disease. Failure to use indicated tests (e.g. amniocentesis and chorionic villus sampling). Use of outmoded tests or treatments (e.g. inferior prenatal screening tests for aneuploidy with poor sensitivities and high false-positive values). Failure to act on results of monitoring or testing (e.g. suggestions of fetal compromise and risk of in-utero demise on antenatal surveillance). |
| Treatment | Error in the performance of an operation, procedure, or test (e.g. procedure-related fetal trauma or miscarriage from invasive testing or treatment, system failures in batch prenatal screening for aneuploidy, or technical errors in sample retrieval for genetic investigations). Error in administering the treatment (e.g. treating the ‘wrong’ unaffected fetus in multiple pregnancy). Error in the dose or method of using a drug. Avoidable delay in treatment or in responding to an abnormal test. Inappropriate (not indicated) care. |
| Preventive | Failure to provide prophylactic treatment (e.g. maternal rhesus sensitisation from failure to administer prophylactic anti-D immunoglobulin after potential feto–maternal haemorrhage). Inadequate monitoring or follow up of treatment. |
| Other | Failure of communication. Equipment failure (e.g. resuscitation equipment, cytogenetic and molecular genetics equipment failures). Other system failure. |
Mistakes can occur during prenatal fetal screening, diagnosis, or treatment. Although emphasis is placed in this review on those errors relating to prenatal diagnosis, errors further upstream in the fetal care pathway, such as screening errors, are often the bases of diagnostic errors, and may then result in erroneous treatment plans and decisions, sometimes with catastrophic consequences.
The nature of clinical errors in prenatal diagnosis
Most medical errors do not result from individual recklessness or the actions of a particular group. It is increasingly recognised that individual or personal clinical errors are often a result of system failures or so called ‘latent errors’— errors waiting to happen. Such errors are caused by faulty systems, processes, and conditions that lead people to make mistakes or fail to prevent them. Such a systems-based approach to error diagnosis and mitigation has led to improved safety in many high risk industries such as the aerospace and nuclear industries where systems redesigns have reduced the occurrence of human error. Application of these concepts to managing clinical risk, including situations in prenatal diagnosis and therapy, is leading to improvements in patient safety.
The types, timing, and domains of medical errors in prenatal diagnosis are summarised in Table 1 . Such errors may have deleterious effects on the fetus, the mother, or both. They may be minor or major, and therefore material or immaterial. Their severity and impact may depend on the gestational age at which they occur, and may be related to birth order, with several errors arising from dealing with multiple pregnancies. Accurate determination of fetal gender is crucial to care in several clinical situations: determining chorionicity in twins, interpreting sex chromosome mosaicism and aneuploidy, and diagnosing sex-linked conditions presenting after 14 weeks. Errors in fetal gender determination can have potentially significant effect on care and pregnancy outcome. Misdiagnosing the side of affectation of fetal anomalies is not often of major significance in utero , as the correct diagnosis can usually be made postnatally. Right-sided congenital diaphragmatic hernias carry a worse prognosis than left-sided lesions, and an error in correctly situating such lesions can significantly affect the parents’ decision to continue with or terminate the pregnancy.
| Error type | Examples |
|---|---|
| Diagnostic | Error or delay in diagnosis of fetal malformation or disease. Failure to use indicated tests (e.g. amniocentesis and chorionic villus sampling). Use of outmoded tests or treatments (e.g. inferior prenatal screening tests for aneuploidy with poor sensitivities and high false-positive values). Failure to act on results of monitoring or testing (e.g. suggestions of fetal compromise and risk of in-utero demise on antenatal surveillance). |
| Treatment | Error in the performance of an operation, procedure, or test (e.g. procedure-related fetal trauma or miscarriage from invasive testing or treatment, system failures in batch prenatal screening for aneuploidy, or technical errors in sample retrieval for genetic investigations). Error in administering the treatment (e.g. treating the ‘wrong’ unaffected fetus in multiple pregnancy). Error in the dose or method of using a drug. Avoidable delay in treatment or in responding to an abnormal test. Inappropriate (not indicated) care. |
| Preventive | Failure to provide prophylactic treatment (e.g. maternal rhesus sensitisation from failure to administer prophylactic anti-D immunoglobulin after potential feto–maternal haemorrhage). Inadequate monitoring or follow up of treatment. |
| Other | Failure of communication. Equipment failure (e.g. resuscitation equipment, cytogenetic and molecular genetics equipment failures). Other system failure. |
Mistakes can occur during prenatal fetal screening, diagnosis, or treatment. Although emphasis is placed in this review on those errors relating to prenatal diagnosis, errors further upstream in the fetal care pathway, such as screening errors, are often the bases of diagnostic errors, and may then result in erroneous treatment plans and decisions, sometimes with catastrophic consequences.
Errors arising during prenatal screening
Marked variation exists in the national governing policy for prenatal screening for aneuploidy and fetal structural abnormalities. In England, Public Health England co-ordinates the National Health Service Fetal Anomaly Screening Programme, which has responsibility for developing, implementing, and maintaining a high-quality, uniform screening programme for all pregnant women. National guidance prescribes that all women should be offered a screening test for Down’s syndrome that meets agreed standards, and an ultrasound scan between 18 and 21 weeks’ gestation to check for physical abnormalities in the baby. It also mandates that all women are provided with information to help them decide if they want screening or not. Given that this care standard largely applies to the rest of the UK as well, and is similar to guidance in most other developed countries, failure to offer women these options during antenatal care is often a source of dissatisfaction with a service, and may lead to litigation should an adverse fetal outcome thereby result.
Even when prenatal screening is carried out without any errors of omission or commission, an inherent margin of ‘error’ or inaccuracy can occur whereby tests are associated with limited predictive ability to identify or exclude aneuploidy. Diagnostic accuracy is usually based on the test’s ability to detect positive (abnormal) and negative (normal) results correctly, the sensitivity of the screening test being its ability to detect fetal aneuploidy when it is present (the probability of a true-positive result set against the probability of the disease being present). The specificity of the test is its ability to discriminate a true-negative result in the absence of aneuploidy (the probability of a true negative result set against the probability of the disease not being present).
Screening and diagnosis of chromosomal and genetic abnormalities
Errors and mistakes may occur during the processes involved in screening and diagnosis of chromosomal and genetic abnormalities during pregnancy.
Errors and adverse outcomes relating to screening tests
Data on the quantum of significant errors in prenatal fetal screening (i.e. Downs’ syndrome) are sparse. These mistakes may occur in relation to counselling, tissue sampling, laboratory testing and quantification, or pregnancy dating. Healthcare providers may fail to offer recommended screening tests to a pregnant woman or to manage the uptake of such tests, an error for which a maternity service could be liable if, for example, a baby affected by Downs’ syndrome is born to parents who would have opted to discontinue the pregnancy had they been offered screening and a diagnostic test if screened positive. Counselling during prenatal screening needs to be non-directive: failure to follow this approach could inform claims of misdirected care should an adverse outcome result. Errors also arise when pregnancy duration is wrongly calculated rendering calculated risks of aneuploidy inaccurate. The laboratory request forms for biochemical testing may be incorrectly or incompletely filled in, or the assessment of nuchal translucency by ultrasound incorrectly performed. Any of these errors may lead to falsely positive or negative results, which would in turn affect parental uptake of diagnostic invasive testing.
Although ultrasound screening for aneuploidy is carried out during the first and second trimesters, only the first trimester sonographic markers of aneuploidy are able to achieve sensitivities that exceed 50%. In a recent review, the optimal predictive accuracies reported in different studies for the different types of markers of Down’s syndrome were summarised: of all the studies, only the assessment of first-trimester ultrasound markers (e.g. nuchal translucency, nasal bone, ductus venosus, tricuspid regurgitation, and frontomaxillary facial angle) attained sensitivities that exceeded 60% for a false–positive rate of about 5%. Despite the predictive accuracy of these first-trimester markers, several pitfalls in their clinical application can lead to errors in screening and diagnosis. Such pitfalls may result from inability to examine the markers, or incorrect assessment or interpretation of the findings when assessed. Adequate marker assessment may be precluded by poor image resolution owing to maternal reasons such as body habitus and history of previous surgery, or by unfavourable fetal position or movements.
Observations may be incorrectly measured or interpreted if the image magnification is insufficient, the calliper placement for nuchal translucency assessment incorrect, or the angle of insonation of structures such as the fetal nasal bone incorrect. The effect of image magnification on screening marker measurements is exemplified by the observation that, at a magnification of 200%, the median nuchal translucency measurement is up to 29% smaller than at 60%, a finding that is independent of gestational age. Increasing image magnification for nuchal translucency screening could lead to a dramatic reduction in screen-positives rates. Several ultrasound marker assessments require that a sagittal section of the fetus is obtained (e.g. nuchal translucency and nasal bone assessments), and that the fetus is imaged in a neutral position without hyperextension or hyperflexion (e.g. nuchal translucency assessment). Failure to obtain these required planes will result in erroneous measurements and risk calculations. Images that are not crisp, or fail to demonstrate clear but thin echogenic measurement margins or lines, prevent accurate nuchal translucency and nasal bone assessments from being obtained. Furthermore, data may become erroneous from venous contamination (when measuring the ductus venosus), or arterial contamination (when assessing for tricuspid regurgitation), among others. A markedly raised nuchal translucency reading may suggest a dire prognosis or may be misinterpreted as a cystic hygroma, resulting in an altered prognosis for the patient.
Errors and adverse outcomes relating to diagnostic tests
A screening test that results in a risk estimate higher than the threshold recommended for invasive testing should inform a non-directive discussion with the couple regarding diagnostic test options and the various decision scenarios that may result from confirmation of an affected fetus. Errors may arise during the management of the decision to have a test, the conduct of the procedure for the test, the relay of test results, and the management of care resulting from the entire screening and diagnostic pathways.
Even when diagnostic tests have been carried out to a reasonably good standard, the procedure may result in undesirable adverse outcomes or complications, including miscarriage. It is often difficult to distinguish fetal loss as a result of an invasive test from spontaneous loss not attributable to the test. One study of 3000 consecutive women undergoing amniocenteses reported failure to obtain amniotic fluid for analysis on first attempt in 0.7% of cases, and to establish a culture in 0.3% of women. Procedure-related loss rate in this study was reported as 1.5%; only 14 (less than 0.01%) diagnostic errors occurred, less than one-half of which were serious enough to affect the outcome of the pregnancy. The karyotyping error rate was 0.07%. Another estimate of miscarriage risk associated with amniocentesis comes from a randomised trial from Denmark, involving 4606 women: loss rate in the amniocentesis group exceeded the control group by 1%.
Similar difficulties are experienced when attempting to estimate procedure-related fetal loss associated with chorionic villus sampling (CVS), as no published studies comparing CVS with ‘no testing’ have been published. Meta-analyses of randomised trials comparing CVS, however, with second-trimester amniocentesis showed an excess pregnancy loss rate after CVS, but similar loss rates in the only direct randomised comparison of trans-abdominal CVS with second-trimester amniocentesis.
Invasive testing is often assessed in terms of the diagnostic performance of laboratory testing measured as the laboratory failure rate, karyotype quality (e.g. G-band score, rate of follow-up samples, rate of wrong diagnoses), and karyotype representativity (defined as rate of follow-up samples and rate of wrong diagnoses). In the final analysis, the gold standard for estimating wrong diagnosis on prenatal karyotype testing is the ultimate fetal karyotype determined postnatally on fetal tissue (usually blood). The incidence of discordance between these results is reported as 0.2% of cases undergoing amniocentesis. The United Kingdom National External Quality Assessment Service provides a comprehensive external quality assessment service in laboratory medicine. The United Kingdom National External Quality Assessment Service sets standards for the performance of cytogenetic and molecular genetic laboratories involved in the provision of such services for regional and national fetal screening and diagnosis programmes. These standards to which prenatal screening and diagnosis laboratories ought to adhere aim to minimise systematic laboratory errors in relation to prenatal diagnosis.
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
