Background
This is the first article in a monthly series devoted to malpractice litigation and diagnostic ultrasound. Malpractice litigation has been an interest of mine since 1981 when I was the target of lawsuit related to a missed spina bifida and since that time I have tracked malpractice litigation by surveying practitioners, getting details of cases from expert witnesses and from reviews of settled cases.Information has been obtained about well over 450 cases and I have personally reviewed 170 cases myself (about 50% for the plaintiff and 50% for the defendant; in many of these cases however, ultrasound had been performed but was not the focus of the litigation). In the ensuing months I will discuss the various types of ultrasonic litigation and give some tips on how to avoid this expensive and psychologically debilitating experience. In this first article I will describe what usually occurs in a malpractice case in the USA.

Setting up the lawsuit
Typically a patient is dissatisfied with the results of medical care either because of a poor result (if it is an unrecognized fetal anomaly) or, more often, because another member of the medical community comments that they cannot understand why a medical decision was made or why a poor result occurred. The patient visits a lawyer. As a rule the initial lawyer is not a malpractice specialist; this lawyer sends the patient to a trial lawyer who specializes in plaintiff medico-legal cases. Medical malpractice lawyers usually concentrate only on medical suits and some may be as subspecialized as physicians. A trial lawyer may concentrate on a field as small as missed chromosomal anomalies or litigation involving the liver.

Creating the "complaint"
Each state has it’s own legal system so the pattern varies by state but the initial process is the same throughout the country. The medical records are obtained by court order. Trial lawyers have a stable of doctor friends who review cases. Often these are local doctors who are reluctant to antagonize other members of the local community by testifying against them but are prepared to say whether they think the case has any merit. The case is then sent to an expert witness to see if they concur that substandard care was given. If the expert agrees a “complaint” is developed by the lawyer and signed by the expert witness. The “complaint” details the patient’s history, as seen by the plaintiff lawyer, and states why care was below the standard of care. (Standard of care is defined as what a reasonable and prudent physician would do under the same clinical circumstances). A series of questions are asked of the defendants as part of this exercise such as background qualifications, training, hospital or group policies and procedure manuals, any other malpractice cases etc.

The defendants deposition
Sometime later (usually two or three years after the case took place) in most states the plaintiff lawyer obtains a deposition from the defendant doctor or doctors and the staff involved with the case. (In some states such as New Jersey the depositions of the plaintiff expert witnesses are taken first and in other states such as Pennsylvania no deposition is taken but the expert witnesses prepare statements). Depositions are not to be taken lightly. Deposition material is given under oath and can be quoted in court at a subsequent trial. If a defendant doctor is found to have lied during a deposition when other witnesses testify or other records are obtained it is almost certain the case will be lost. One should prepare well for a deposition and be quite clear about the sequence of events and think about a plausible explanation for the poor result that is on the one hand truthful and accords with standard practice and on the other hand brings out the difficulties of the case. It is wise to give as little information as possible and make ones answers as short as possible. A chance remark can lead the plaintiff lawyer down dangerous byways. Any type of negligence even though quite unconnected with the damage in the case can make a defendant look unsympathetic. For example details of a kickback scheme whereby an obstetrician charged for interpretation of a case even though he never looked at the images and the films were being interpreted as a “second read” by a radiologist for a small fee tainted the defense of one defensible case. Make sure that the memories of others involved with the case such as sonographers, nurses or office staff agrees with your memory. Often nobody can remember the case in which case the best answer is “I don’t remember”, Indeed it is usually better to say, “I don’t remember” if you are uncertain what happened and have only a vague remembrance. Since it usually takes at least two years for a case to get to the deposition stage it is not surprising that often you and your staff have forgotten the patient and what actually happened even though the patients impression of what happened is embedded in their brain.

The expert witness deposition
Next the depositions of the plaintiff expert witnesses are obtained. The expert witness will be asked about his or her qualifications and will then state his or her criticisms of the case and why he or she considers the care given was below the standard of care. (There are no clearly established qualifications for expert witnesses – they do not have to be board certified or practicing in the same specialty however previous testimony in depositions or trials can be used to trap those “professional” expert witnesses who tailor their testimony to the needs of the plaintiff lawyer.)

Settlement talks
At this point settlement talks will probably start. Sometimes the malpractice is so obvious that a settlement is inevitable. Haggling between defendants then begins as to which defendant bears the major responsibility. This haggling may be complicated by the extent of the malpractice insurance coverage with some defendants having coverage of more than a million dollars but others having only a $100000 even though they may be more at fault. Disagreement between defendants about culpability may result in indefensible cases going to trial. Sometimes the case is so trivial and so unlikely to succeed that the question then occurs of whether it is better to settle for a small sum or to go to trial and have the case dismissed. Trials are very expensive since expert witnesses have to be flown in and paid several hundred dollars an hour and the defense Iawyer’s fees are large. (See table 1) The insurance company may well conclude that it is cheaper to settle for a few thousand dollars than to defend a case that is certain to be won in court. About 98% of cases are dropped or settled and it is a tiny minority that proceed to court. (See table 2) Settlement is delayed until the last possible moment so that the money can continue to earn interest. It is common for the settlement to take place days or even hours before the trial occurs.

The trial
At this point settlement talks will probably start. Sometimes the malpractice is so obvious that a settlement is inevitable. Haggling between defendants then begins as to which defendant bears the major responsibility. This haggling may be complicated by the extent of the malpractice insurance coverage with some defendants having coverage of more than a million dollars but others having only a $100000 even though they may be more at fault. Disagreement between defendants about culpability may result in indefensible cases going to trial. Sometimes the case is so trivial and so unlikely to succeed that the question then occurs of whether it is better to settle for a small sum or to go to trial and have the case dismissed. Trials are very expensive since expert witnesses have to be flown in and paid several hundred dollars an hour and the defense Iawyer’s fees are large. (See table 1) The insurance company may well conclude that it is cheaper to settle for a few thousand dollars than to defend a case that is certain to be won in court. About 98% of cases are dropped or settled and it is a tiny minority that proceed to court. (See table 2) Settlement is delayed until the last possible moment so that the money can continue to earn interest. It is common for the settlement to take place days or even hours before the trial occurs.

If the case goes to court the plaintiff presents first. Elaborate artistic renderings of the injuries sustained or the surgical or ultrasonic technique used are produced. Images are blown up to a huge size. The patient is shown in the worst light to elicit jury sympathy; a cute child with a striking deformity perhaps unrelated to the consequence of the alleged malpractice is a very desirable witness for the plaintiff. Expert witnesses battle credentials and opinions although often, since imaging is highly specialized, expert witnesses tend to cancel each other out in the eyes of the jury.

The damages
An important feature of this phase is the contribution of the medical economist who has to calculate damages. He or she calculates the future cost of medical care, loss of earnings, amount to allow for inflation using an estimated life expectancy related to the average life expectancy of individuals of the plaintiff patient’s age. Computed in this fashion the total damages may sound huge – as large as $25-30,000,000 in total for an infant. Typically an annual allowance is made to the patient and insurance companies are prepared to sell a policy covering pay outs of 2-300,000 dollars per year to age 72 for 2 or 3 million dollars in the expectation that the patient will not survive that long.

What the plaintiff lawyer gets
Pay out to the plaintiff lawyer is dependent on whether settlement occurs before the trial or after the trial. Plaintiff lawyers usually get 30% of the award at settlement and 40% of any trial award.

Until about 15 years ago, ectopic pregnancy was by far the most common cause of litigation associated with ultrasound. Today, lawsuits related to missed or invented ectopic pregnancy are less frequent but remain a significant problem for ultrasound practitioners. A forthcoming article in the Journal of Ultrasound in Medicine will review experiences with ultrasound-related litigation since 1981. Twelve lawsuits relating to ectopic pregnancy were reported prior to 1983, nine between 1983 and 1986, 22 between 1986 and 1996, and two additional suits since 1996. In all, a total of 45 lawsuits relating to ectopic pregnancy have been filed (Sanders 2003 in press JUM).

Why is the diagnosis of ectopic pregnancy so tricky? Before the use of the vaginal probe, small ectopic pregnancies often could not be directly visualized and only secondary signs of ectopic, such as a cul-de-sac bleed and/or adnexal mass, were detectable. Non-specific findings led to non-specific reports that sometimes failed to mention ectopic in the list of possibilities. In several instances, even when the patient had a previous history of ectopic with almost identical symptoms, the condition was not sought nor was a sonogram performed. In other instances, the sonogram report failed to mention ectopic pregnancy in the differential diagnosis. Today, with use of the vaginal probe, ectopic pregnancy is more easily detected and definitively diagnosed and the number of lawsuits has diminished significantly. However, a number of pitfalls still exist.

In a landmark paper from the University of San Francisco, Marks, et al., described a “decidual cast” as a sign of ectopic pregnancy (The decidual cast of ectopic pregnancy: a confusing ultrasonographic appearance Marks WM, Filly RA, Callen PW & Laing FC Radiology (1979) 133: 451-4). The term “decidual cast” is misleading because a cast does not form in the decidua. The entity is better described as a pseudogestational sac. Although the most common sonographic indication of ectopic pregnancy consists of blood pooled in the cul-de-sac, in some cases blood from the leaking ectopic pregnancy moves through the fallopian tube and accumulates in the endometrial cavity. The endometrial cavity walls are thickened by a decidual reaction to the ectopic pregnancy and appear somewhat like the walls of a gestational sac. The internal echoes within the blood in the endometrial cavity may be faint or not detected with a lower cost system. Often, the shape of the pseudo sac is surprisingly circular. A pseudogestational sac can be distinguished from a true intrauterine pregnancy because the “double wall” seen in a normal gestational sac is absent. Several of the aforementioned lawsuits were generated when a pseudogestational sac was termed an intrauterine pregnancy (Figure 1).

“Decidual cast” or pseudosac wrongly reported as normal early pregnancy in an ectopic pregnancy that subsequently ruptured with loss of the tube on the involved side.

Since the pseudo sac of ectopic pregnancy was first described in radiology literature, it took some time for knowledge of this confusing appearance to percolate to the obstetrical literature. During this period, obstetricians misdiagnosed a number of these pseudogestational sacs as early intrauterine pregnancies.

Most ectopic pregnancies present at about 5 to 6 weeks when, often, there is no visible gestational sac in a normal intrauterine pregnancy. Some lawsuits have been based upon a failure to mention ectopic in the differential when nothing is seen in the cavity. If the patient has had vaginal bleeding, three possibilities exist: 1) the patient has had a complete spontaneous abortion;  2) there is an ectopic pregnancy; or less likely,  3) there is a developing intrauterine pregnancy.  On the other hand, if no bleeding has occurred, a late-developing intrauterine pregnancy is likely. When ectopic pregnancy is not mentioned in the differential diagnosis, legal problems arise if a rupture with bleeding from an ectopic occurs later on.

In the absence of free blood in the cul-de-sac, we now know that ectopic pregnancy can be treated in a conservative fashion without surgery; however, the new alternative therapy, i.e., oral administration of methotrexate, has led to a number of new lawsuits. In several instances, obstetricians have acted prematurely by administering methotrexate before the diagnosis of ectopic pregnancy was ascertained. In two cases, an obstetrician reviewed the sonogram in the emergency room, thought an ectopic pregnancy was present, and immediately administered methotrexate. Subsequent review by a radiologist the following morning correctly reported a viable intrauterine pregnancy. In another case, when a radiologist definitively diagnosed a cornual pregnancy, methotrexate was administered. Two days later, a repeat sonogram showed a viable intrauterine pregnancy in a more normal location. A later laparoscopy showed that a bicornuate uterus had created this confusing appearance. An eccentric location to the gestational sac suggestive of a cornual pregnancy is a notoriously confusing situation. Sometimes, an off-center location with a thin myometrial wall is due to fibroids, or a bicornuate or subseptate uterus. There are times when this genuinely represents a cornual pregnancy, but more often, it is related to the degree of bladder filling or a myometrial contraction. If there is no evidence of symptoms of rupture and no sonographic signs such as blood in the cul-de-sac, a cautious approach to the treatment of a possible cornual pregnancy seems advisable. A later sonogram may be more reassuring.

In summary, litigation related to ectopic pregnancy falls into four main categories:

1. missed ectopic pregnancy in which a patient has typical symptoms, there is a non-specific appearance, or ectopic is not mentioned in the differential diagnosis
2. confusion between a decidual cast and an intrauterine pregnancy
3. early pregnancy dilemmas when the evidence of an intrauterine pregnancy is subtle and methotrexate is administered to a normal pregnancy
4. issues relating to the misdiagnosis of cornual pregnancy.

Free fluid in the adnexa or cul-de-sac is a crucial sonographic sign since it suggests active bleeding or rupture and rapid surgical treatment is required. For the most part, ectopic pregnancy is no longer a surgical emergency and decisions on management can await expert sonographic opinion.

Defective diagnosis or management of multiple pregnancy is a relatively common form of malpractice related to ultrasound. There are 26 examples in my series (Sanders R, J. Ultrasound Med. In press).

Year of series showing number of multiple pregnancy cases

Fifteen of these cases occurred before real-time ultrasound was routinely used. Additional pregnancies could be easily missed when static scanning was the only form of ultrasound available because the scanning process took 20-30 seconds for each cross section and the fetus often moved while the scan was taking place. An additional fetus might be confused with a fetus that had moved. This was especially common if there was polyhydramnios. Despite the introduction of real-time an occasional missed twin or triplet continues to be overlooked with today’s equipment if a casual scan is performed without a thorough look at the entire uterine cavity. This type of miss is almost indefensible and falls into the category of “Res ipsa loquitor” – literally “the thing speaks for itself”. In a normal legal case the plaintiff lawyer must show 1) that the defendant owed a duty to act reasonable 2) breached that duty by acting unreasonably 3) caused an injury and 4) that the injury resulted in real and tangible harm. In the case of a missed twin the injury is so obvious that it would be difficult to contend that malpractice has not occurred.

Failure to perform ultrasound in a pregnancy that is “large for dates” is another common form of litigation related to multiple pregnancy. “Large for dates” is a recognized indication for an ultrasound study (1984 NIH ultrasound consensus conference). The usual malpractice issue with a “large for dates” pregnancy is an exceptionally large fetus with subsequent complications at delivery related to the large fetal size, such as Erbs palsy. Another important reason for an ultrasound study and cause of malpractice litigation with “large for dates” is the unexpected twin. When the presence of a twin or triplet has not been recognized prior to delivery, the fetus that remains in the uterus is often damaged because of problems such as a breech delivery with an intracranial bleed or intrauterine asphyxia.

There have been several cases related to the “stuck twin syndrome” or “twin-twin transfusion syndrome”. These two related entities both result from a shared placenta with one twin monopolizing the blood supply. They are probably the most common cause of multiple pregnancy litigation today. The crucial distinction is between a monochorionic diamniotic pregnancy which requires close follow up to avoid the complications related to twin-twin transfusion syndrome or stuck twin syndrome, versus the relatively benign course of a diamniotic dichorionic pregnancy where the main hazards are prematurity and IUGR. When multiple pregnancy ultrasound is first found it is therefore very important that one distinguishes between monochorionic diamniotic pregnancy and dichorionic diamniotic pregnancy. . The easiest time to diagnose a dichorionic diamniotic pregnancy is during the first trimester when there is a separate thick-walled gestational sac for each of the pregnancies. This is however a difficult time to distinguish between a monochorionic diamniotic pregnancy and a monochorionic monoamniotic pregnancy because the intervening thin membrane in monochorionic diamniotic twin pregnancy may be very subtle until the second trimester.

In the second and third trimester the distinction between a monochorionic diamniotic pregnancy and a dichorionic diamniotic pregnancy can be made in four ways. The most definitive and easiest technique is finding that the twins are of different gender. If one of the fetuses is a boy and the other a girl a diamniotic dichorionic pregnancy must be present. A second very helpful sign of a dichorionic diamniotic pregnancy is the presence of the “lambda” or “twin peak” sign. In these cases a triangular portion of placenta enters the base of the membrane separating the two sacs at the point where the membrane lies in contact with the placenta. Although this sign is not infallible its presence almost always means that there is a dichorionic diamniotic pregnancy. The presence of two placentas usually means that the pregnancy is dichorionic diamniotic, however, beware of a succenturiate lobe with a monochorionic diamniotic pregnancy. Often the two placentas lie in contact with each other so the absence of two placentas does not automatically mean that the pregnancy is monochorionic diamniotic. The last and most difficult sign relates to the intervening membrane. In a dichorionic diamniotic pregnancy there are four components to the “membrane” since there will be a membrane from each amniotic sac and from each chorion. In a monochorionic diamniotic pregnancy the intervening membrane will only have two components - one from each amniotic sac.

When a monochorionic diamniotic pregnancy is discovered follow up should be relatively frequent between 16 and 32 weeks. Complications related to twin-twin transfusion syndrome or stuck twin syndrome occur in up to one third of monochorionic diamniotic pregnancies. A three to four weekly interval between follow up sonograms is recommended.

The standard of care requires that this type of pregnancy be sent to a specialist in prenatal ultrasound. If there is markedly increased fluid in one cavity and little or none in the other an acute obstetrical emergency is present. Recognition that one of the twins is fixed in position, sometimes on the anterior aspect of the amniotic cavity, is important and easily missed. The amniotic membrane around the stuck twin is tightly wrapped around it.

A stuck twin (ST) is adherent to the anterior aspect of the amniotic cavity. The membrane around the stuck twin cannot be seen because it is “shrink-wrapped” around the twin because there is so little fluid. Polyhydramnios is present around the larger twin, which has some fetal ascites (arrow). Recognition of the presence of a stuck twin in this case was delayed and led to complications.

Immediate drainage of large quantities of amniotic fluid (1.5-2.5 liters) from the larger cavity is required if the fetuses are to be salvaged. There is an 80% mortality rate without therapy and even with therapy the survival rate is only 50-80%. Litigation has revolved around failure to refer to a perinatologist, failure to recognize the danger of the situation, or failure to initiate immediate therapy.

Complications related to multiple pregnancy are a common source of malpractice litigation. In the past missing an additional pregnancy at a routine ultrasound examination was the most frequent issue; this problem is unusual today. Some of the management issues that result in litigation in multiple pregnancy such as prematurity cannot be diagnosed with ultrasound but sonography is the prime diagnostic tool for complications related to monochorionic diamniotic pregnancy. The distinction between monochorionic diamniotic pregnancy, a situation often complicated by fetal ill health, and diamniotic dichorionic pregnancy is often not reported and can result in serous long-term consequences.

Guidelines for the performance of a basic level obstetrical sonogram were first developed in 1984 by the American Institute of ultrasound in medicine (AIUM) at the urging of its obstetrical section (SOGU). The American College of Radiology (ACR) adopted identical guidelines in the same year. Guidelines for first trimester sonograms were part of the document and have never been controversial. The second and third trimester guidelines have been the subject of much debate and litigation. These required photographic or videotape documentation of the number of fetuses, a series of views of the fetus including the standard measurement data views (the biparietal diameter (BPD), head circumference (HC), abdominal circumference (AC) and femoral length (FL) and views of the stomach, cord insertion, bladder and lateral ventricles. Images of the kidney and spine “areas” were suggested (At that time the resolution on some ultrasound systems was insufficient to see the kidneys in the early second trimester and seeing the spine adequately was considered impractical when the fetal position was suboptimal). Documentation of the uterine configuration, fetal and placental position and amount of amniotic fluid was also part of the required study. The guidelines were subsequently updated and in the latest version issued in 1994 requirements to document the kidneys, spine, the cerebellar/cisterna magna view of the skull and of the four-chamber view of the heart were added. Notably absent from any of the guidelines is a requirement to examine the facial structures or the outflow tracts of the heart. In addition, examination of all limbs and hands and feet are not required.

At about the same time the American College of Obstetrics and Gynecology (ACOG) issued a technical report (which despite denials from ACOG carries the same legal weight as a guideline from other organizations) recommending examination of the same structures. There is a crucial difference however; no photographic or videotape documentation is required.

What has been the impact of the guidelines? Since the bulk of obstetrical ultrasound in the USA is performed by sonographers but reported by physicians these guidelines have been helpful in establishing a standard ultrasonic series and a standard reporting framework. For example in a lawsuit related to a little performed area of medicine such as labial ultrasound where standards do not exist the opinions of opposing expert witnesses battle to establish the accepted standard of care in that case. With formalized obstetrical ultrasound guidelines the vagaries of expert witnesses as to what is standard practice are supplanted by the document.

Let’s look at the benefits of guidelines when litigation occurs. In a case from my practice an infant was discovered to have no cerebellum at birth. A suit was initiated. Expert review of the sonogram that I reported at 18 weeks showed a normal cerebellum so the case was dropped because it had no merit. Presumably a cerebellar infarct occurred between 18 weeks and term but standard obstetrical ultrasound documentation had occurred proving there was a normal cerebellum earlier in the pregnancy. Several cases have been sent to me for review where hypoplastic left heart syndrome has been found at birth but was not detected in utero at the 18-20 week sonogram. Review of these sonograms showed a normal four-chamber view. Since the development of hypoplastic left heart syndrome may be progressive over the course of pregnancy rather than congenital, a normal four-chamber view early in the pregnancy is very helpful when defending a missed hypoplastic left heart syndrome case. (Other examples of conditions that progress over the course of pregnancy that also may not be detected at the 18-20 weeks exam include hydronephrosis, achondroplasia and duodenal atresia.)

The disadvantage of relatively rigid guidelines stems from those times when obtaining the required views is impractical. The fetal position may make it impossible to obtain a satisfactory abdominal circumference for instance. A suboptimal examination may occur even with the most skilled examiner when there is gross obesity or a tantalum mesh over a ventral hernia. In these situations the onus is on the reporting physician; he or she needs to comment on the poor quality of the study and why it is suboptimal. In one case, a missed spina bifida, the patient was overweight and limited views of the spine were taken; the study was performed on an emergency basis for abdominal pain in pregnancy. An abnormal lemon shape to the head was present and a banana sign was subtly visible on one view. The radiologist settled the case for a substantial sum because he reported the case as a normal 18 weeks study. If he had commented on the difficulty performing the study with a clarifying explanation of why he could not rule out abnormalities the case would have been defensible.

A major discrepancy between the AIUM and ACR guidelines, which require documentation, and the ACOG guidelines, which do not, means there are different accepted standards for obstetricians than for radiologists. How does this difference play out during litigation? Judges and juries tend to presume that in the absence of documentation the observation was not made. It is hard to defend a case in which you state in your report that a four-chamber view of the heart is normal (without a photograph) when at delivery there is hypoplastic left heart disease. Equally indefensible is a case where you state without images that all four limbs are present yet at birth both arms are missing; this in fact occurred in a recent case reported and performed by an obstetrician. Absence of a documentation requirement makes for a speedy examination and reporting, however it is difficult to justify a doctor’s signature on the report of a sonographer on an examination with limited documentation if the doctor never saw the images.

On the other hand there can be an upside to the absence of documentation. A retrospective review of an 18-week obstetrical case did show evidence of posterior urethral valves and a suit was successfully brought against the defending radiologist. Since the condition was not clinically detected until age 2 and might theoretically have worsened over time and not been visible at 18 weeks documented evidence worked against the defendant. It is unlikely that the case would have been initiated without the abnormalities visible on the prenatal sonogram. However it is unusual that a retrospective review of a sonogram shows an anomaly that was missed at the time of the report. More commonly the anomaly was missed at the time of the study, and cannot be seen in retrospect. This is often the case because images were below the standard of care. In an ongoing series of 478 litigation cases related to ultrasound there were 194 missed diagnoses of which 83 were fetal abnormalities; out of 66 misreported cases there were only 12 instances in which fetal abnormalities could be recognized in retrospect. The details of this survey are to be published in the JUM in 2003.

Guidelines for obstetrical ultrasound have been invaluable in standardizing an examination that could be unrealistically lengthy. Without these guidelines differing experts could claim different levels of detail as being within the standard of care. For instance one expert might claim that the details of the lips and orbits should always be seen when we know in practice that facial views may not be practical due to fetal position. In no other imaging technique is the selection of images and the level of detail so much at the discretion of the operator. Documenting the standard obstetrical ultrasound study, with images that conform to the guidelines, protects the examiner against possible litigation. The skill to recognize the presence of abnormalities is the other essential ingredient. If, through lack of training or experience, the examiner has not attained full competence then hard-copy pictures can be damaging. People in this category should not perform unsupervised ultrasound examinations.

The standard obstetrical measurements have been in use since the early 1980’s. The biparietal diameter was described in 1970 (Campbell 1970), the abdominal circumference in 1975 (Campbell and Wilkin 1975), the head circumference in 1982 (Hadlock et al 1982) and the femur length in 1981 (O’Brien and Queenan 1981). Measurements of the head and abdominal circumferences have become technically easier as automated ellipse calculation has been incorporated into equipment. Initially there was some disagreement about the appropriate gestational age or estimated weight for a given ultrasonic measurement but since 1990 and the work of Hadlock there is now general agreement about which chart to use and charts based on large series show relatively minor differences. All systems and charts have standardized on an assumed speed of sound through tissue of 1540 meters per second. Slight population differences have been found but the differences are so trivial that they are usually of no clinical importance; it is now known that Asian fetuses are slightly smaller and that measurement for males and females are minimally different. Measurements for twins differ so little from singletons that the same charts can be used.

Although obstetrical measurements are measured every day in every pregnancy in the same way that they have been used for years litigation related to the misuse or misestimation of dates or weight remains commonplace. In the series that I have examined dating from 1981 there have been 24 cases in which the litigation related to measurements (about 8% of obstetrical ultrasound litigation) (Sanders, in press, JUM). The frequency with which this type of case is being seen is essentially unchanged over the years. Many cases hinge on technical errors in the way in which the measurements are obtained. The abdominal circumference measurement is technically the most difficult measurement and is the measurement that most often results in litigation because estimation of fetal weight is so heavily dependant on the abdominal circumference. In some instances the abdominal circumference is underestimated because the measurement outline does not include the soft tissues around the trunk. In other cases a more or less round abdominal circumference was not or could not be obtained and the ellipse includes amniotic fluid or other tissues around the fetus. In other instances the measurement was taken at the wrong level or in an oblique fashion.

Legend. Poorly performed abdominal circumference, which is oblique and at too low a level since it includes the kidneys (arrow).

Another unsatisfactory abdominal circumference. A portion of tissue that is not part of the fetal abdomen was included in the abdominal circumference outline presumably because of the fetal lie. This suboptimal circumference may be the best that can be done. If so commenting on the suboptimal level of the view in the report would prevent litigation.

Legend.Technically satisfactory abdominal circumference, which was mismeasured. The markers should outline the outer edge of the soft tissues around the fetal trunk rather than the outer edge of the fetal liver, as was done here.

Poorly performed abdominal circumference measurements may result in underestimation of weight and missing macrosomia or in overestimation of weight and missed IUGR. It is much less common to see the other measurements – biparietal diameter, head circumference and femur length technically poorly performed but in one recent case in which three different examiners were involved the femur length was overestimated by one examiner. Another examiner underestimated the abdominal and head circumferences and took an oblique biparietal diameter and the third examiner obtained a squashed unusually shaped abdominal circumference.

It has been known since obstetrical measurement data was first described that the reliability of dating a pregnancy with ultrasound becomes progressively less accurate as the pregnancy proceeds and by about 26 – 28 weeks is of about the same accuracy as the exceedingly inaccurate technique of measuring the fundal height and is clinically useless as a dating technique. Providing that earlier ultrasonic measurements have been made or that accurate conception dates are known ultrasound later in pregnancy becomes a valuable method estimating fetal weight especially in smaller fetuses. In larger fetuses the margin of error is wide but this is of much less clinical importance. (Even so suits for missed macrosomia are commonplace; certainly a properly performed obstetrical ultrasound study in such cases should indicate that the fetus is large although weight estimation in the best of hands can be off in a 5000-gram fetus by as much as an 800-1000 grams).

Although it is well known that late dating with ultrasound is not possible, it is still commonplace to see reports that baldly state a gestational age at 36 weeks with no caveat about the inaccuracy of dating at this gestational age. Delivery of an immature fetus that is actually of 32 weeks gestational age, although measuring larger, has taken place as a result of such a crude report.

Even more common is failure to reference an earlier obstetrical study and show that growth has been less than expected since the earlier study even when it was performed in the same institution. The standard of care requires that reference be made to an earlier study if one exists. If the study cannot be obtained a statement to this effect in the report safely covers this problem.

If the first study takes place at a relatively late date and an accurate LMP is known it is very important that this is entered into the measurement computer accurately. Several lawsuits relate to putting in a date that was wrong, usually a month off, making the fetus seem a month older than it actually was. Eventual elective early delivery led to lung or cranial injuries.

Although there is a difference in the suggested amount of documentation of an obstetrical ultrasound study, depending on whether you are using the AIUM/ACR guidelines or the ACOG technical bulletin documentation of obstetrical measurement, data should always be obtained. In one instance where a recently performed outpatient sonogram had shown the fetus to be 23 weeks a subsequent undocumented inpatient study suggested a 25 week gestation and extensive efforts wee made to maintain a very immature fetus that subsequently had severe brain injury and would probably not have been the subject of resuscitative measures had the correct gestational age been used.

In sum, misuse of or misestimation of obstetrical measurement data is a common source of litigation. Problems relate to:

1. technically poorly performed measurements resulting in over or underestimation of fetal weight
2. giving a gestational age by ultrasound when the fetus is greater than 28 weeks with no earlier study or caveat that gestational age estimation at a late date is inaccurate
3. failure to compare a set of ultrasound measurement data to an earlier study resulting in failure to recognize poor or excessive growth
4. using the wrong LMP for estimating fetal growth
5. absence of obstetrical measurement documentation in cases with a poor outcome.

Surprisingly few lawsuits have occurred related to the use of ultrasound as a guidance technique for invasive procedures. Almost all the suits that have occurred relate to the use of ultrasound for amniocentesis with subsequent loss of the pregnancy. (However two suits related to funipuncture performed related to transfusions for Rh. disease). The following table shows the number of cases related to amniocentesis in my series of litigation cases (Sanders 2003)

Amniocentesis typically occurs for two reasons. Genetic amniocentesis to determine chromosomal status is commonly performed between 14 and 20 weeks and is usually a relatively simple procedure. Amniocentesis to determine fetal lung maturity by measuring the L/S ratio is performed in the third trimester. Amniocentesis late in pregnancy is often more challenging because the amount of amniotic fluid normally diminishes in the third trimester however since amniocentesis for fetal lung maturity is less frequent than genetic amniocentesis fewer suits have occurred related to third trimester amniocentesis.

There has been a gradual change in the way in which amniocentesis is performed over the last 25 years. Initially amniocentesis was performed with the bladder empty low in the midline without any ultrasound guidance. Later a site with plenty of fluid but no fetal parts was marked using ultrasound and the amniocentesis was performed in another room. Next the amniocentesis was performed in the same room as the ultrasound examination after choosing a site with ultrasound and ultrasound was used to check the site if the amniocentesis failed. Finally the modern technique in which the needle path is chosen with ultrasound and the path of the needle is followed as the needle is inserted with ultrasound was developed. In one variant on the this technique the operator holds the transducer in one hand and the needle in the other; in a second version a sonographer holds the transducer at right angles to the needle while the needle is inserted vertically by the physician. Both techniques are now standard of care but the older techniques have become obsolete. For a time amniocentesis was performed in the radiology department by the obstetrician but a radiologist or a sonographer working for a radiologist provided the guidance. Nowadays most amniocenteses are performed in the obstetrical area by obstetricians sometimes with the aid of sonographers.

About 40 % of the suits in my series relate to amniocenteses that failed in which there was no ultrasound guidance. Suits of this type are most unusual these days. The other 60% of suits are cases in which ultrasound was used for guidance but in which it was alleged that guidance was improperly used or the amniocentesis was technically flawed. Complications related to amniocentesis occur in about 1 in 300 amniocenteses depending on the series. Fluid leakage per vagina is common but usually of no importance. If fluid leakage results in anhydramnios pulmonary hypoplasia may result. Direct needle puncture of the fetus is usually of little consequence although loss of an eye or a limb has been reported. Premature onset of labor is a relatively common complication, which is of limited importance if the amniocentesis is being performed in the third trimester but is fatal with a genetic amniocentesis. . Finally, infection complicates a small minority of amniocenteses. This is a lethal complication if it occurs following a genetic amniocentesis.

In most instances the husband is present when an amniocentesis is performed and if both husband and patient see unsuccessful passes made with only blood aspirated and the pregnancy is subsequently lost with premature delivery a suit may result. In many instances no image of the needle within the amniotic fluid is available or there is a contradiction between the physician notes on the amniocentesis and the patient’s and her husband’s testimony. Most textbooks recommend that only two amniocentesis passes be made. The definition of a pass is unclear and some physicians leave the needle within the patient but reinsert in a different direction. Sometimes the radiologist notes on the procedure differ from the obstetrical records.

Informed consent for the amniocentesis is often performed in a perfunctory fashion and the patient is barely aware that fatal complications may occur. Suits related to failure to obtain proper informed consent have occurred.

In one suit a patient was sent in for genetic amniocentesis for twins. Two amniocenteses were performed one in each amniotic cavity. The fluid was sent to a cytogenetic laboratory. The courier froze the fluid so it did not grow any cells and the amniocentesis had to be repeated. When the amniocentesis was performed again a few days later the fluid in one of the twin sacs was blood stained related to the earlier amniocentesis. Chorioamnionitis ensued and both pregnancies had to be terminated. Damages were awarded against the cytogenetics company.

In another suit an amniocentesis was performed to determine lung function at 39 weeks. Very little amniotic fluid was present and the fetus was damaged and was stillborn. There was discrepancy between the testimony of the obstetrician and the radiologist’s notes. In any event it was alleged that in a well-dated pregnancy (as this was) amniocentesis was unnecessary for fetal lung maturity at 39 weeks.

Amniocentesis is the one guidance procedure performed with the aid of ultrasound that has resulted in significant litigation. Litigation has occurred following genetic amniocentesis and third trimester amniocentesis for fetal lung maturity. Suits have occurred related to failure to obtain informed consent and to loss of the pregnancy either by premature labor or infection. Physicians who perform amniocentesis would be well advised to obtain proper informed consent, to document an image showing the needle in amniotic fluid when the fluid is removed and to write detailed notes on the procedure. Amniocentesis in the third trimester can be technically challenging and should be performed by experienced operators.

Sanders R Changing patterns of ultrasound-related litigation 2003 J Ultrasound Med 22: 1009-1015