■ REVIEW ARTICLE ■ Litigation in

LITIGATION IN OBSTETRICS: A LESSON LEARNT AND A LESSON TO SHARE

Min Min Chou* Department of Obstetrics and Gynecology, Taichung Veterans General Hospital, Chung Shan Medical University, Hung Kuang University and School of Medicine, National Yang Ming University, Taichung, Taiwan.

SUMMARY A perfect baby is the expectation of all parents, and a perfect outcome is the mission of obstetrics. Every obstetrician dreads to hear that there is an unexpected maternal mortality and/or severe fetal injury at the hospital. The role of a perceived public expectation of perfection in obstetric medicine reflects a belief that bad outcomes in obstetrics should not be tolerated and that every maternal–fetal injury merits financial compensation and punishment. What has brought these troubling times to obstetric medicine? The drivers behind malpractice crises are the four leading interest groups in the medical-legal debate: pregnant patients and their environment (husband, parents, relatives, friends, legislators, and the media), health-care providers, insurance companies, and trial attorneys. Litigation in obstetrics is the result of a complex of events when malpractice (presumed or real) impacts on the attitude of pregnant women and their environment. In such complexity, information is mandatory but may often be misinterpreted. If messages are not tailored to the receiver’s capacity, communicating well with the pregnant patient becomes crucial. Therefore, to reduce medical- legal issues in obstetrics, increasing attention and an applicable standard of obstetric care to avoid negligence and medical errors should go along with better communication with pregnant women. Communication should be clear, targeted, effective, flexible, and empathic to share a common language and decisions. This review briefly presents and discusses some of the most frequently encountered medical-legal claim cases in obstetric practice. In-depth review of -related deaths and major morbidities can help determine strategies needed to continue making pregnancy safer. [Taiwanese J Obstet Gynecol 2006;45(1):1–9]

Key Words: litigation, malpractice, maternal mortality, obstetrics, pregnancy

Maternal Mortality: We Must do Better been made toward this goal worldwide. The maternal mortality ratio has remained stable at 7.5/100,000 The leading causes of maternal death are thrombo- since 1982 (US Centers for Disease Control), but embolism and pregnancy-induced hypertension, fol- maternal mortality is underreported in 30–60% of lowed by early pregnancy loss, hemorrhage, amniotic investigated cases [3]. To reduce pregnancy-related fluid embolism (AFE), and genital tract sepsis [1,2]. The mortality, we must understand which deaths are goal of Healthy People 2000 was 3.3 maternal deaths/ potentially preventable and the changes needed to 100,000 live births [1,3]. However, little progress has prevent them. Most deaths due to hemorrhage and complications of chronic diseases are believed to be potentially preventable, whereas deaths due to AFE, *Correspondence to: Dr. Min Min Chou, Department of Obstetrics microangiopathic hemolytic syndrome, and cerebro- and Gynecology, Taichung Veterans General Hospital, 160, vascular accident are not considered preventable Section 3, Taichungkang Road, Taichung 407, Taiwan. [3,4]. Improved quality of medical care is considered E-mail: [email protected] the most important factor in preventing these deaths. Received: December 26, 2005 Revised: January 22, 2006 However, approximately half of malpractice suits in Accepted: January 23, 2006 obstetrics appear to function more like a lottery than

Taiwanese J Obstet Gynecol • March 2006 • Vol 45 • No 1 1 M.M. Chou

as a mechanism for either quality assurance or just Inherited thrombophilia is a genetic condition that retribution. increases the risk of thromboembolic disease, recurrent fetal loss, fetal growth restriction, stillbirth, severe preeclampsia, and abruptio placentae. Clinical fea- Pulmonary Embolism tures are venous thrombosis occurring at a young age and significant family history. The first venous throm- Case boembolic event (VTE) is frequently precipitated by A 31-year-old woman, gravida 3, para 2, went into a pregnancy. The main causes are deficiencies of anti- local clinic to deliver her third child. She had had two thrombin (incidence 1:600–1,500), protein C (1:500), previous cesarean deliveries. A repeated cesarean sec- and protein S, activated protein C resistance (factor V tion was carried out smoothly. The infant did well, as Leiden mutation, 5–15% of Europeans but rare in Asian did the mother for the first 24 hours. Unfortunately, populations), factor II gene (prothrombin G20210A) the woman suddenly collapsed after standing up at mutation (2–3% of white Europeans), and hyperhomo- 08:30 the next morning, and was hypotensive and cysteinemia (mutant of ethylenetetrahydrofolate hypoxic. She quickly became asystolic and was reductase occurring in 11% of white Europeans) [9–14]. immediately transferred to a tertiary medical center for In contrast, acquired thrombophilia occurs in patients further intensive care. On admission, laboratory data who have antiphospholipid syndrome and lupus showed that acute myocardial infarction troponin-T anticoagulant and also among patients with essential test was negative, creatine kinase-myocardial bound thrombocythemia. (CK-MB) was 17 U/L (normal, < 16 U/L), and ammonia In light of the rapid advances in this area, it is certain level was 214 μg/dl (normal 0–70). Blood gas analysis that new causes for genetic susceptibility to venous reported a pH of 6.829, base excess (BE) of –22.4 thrombosis in pregnancy will be identified in the near mmol/L, and oxygen saturation of 59.2%. Autopsy future. suggested a possible pathophysiologic mechanism of sudden death due to pulmonary embolism (PE). HELLP Syndrome: the Great Masquerader However, inherited or acquired thrombophilia disorders or Imitator had not been investigated very thoroughly in this victim. Case Comment On November 17, 2003, a 34-year-old nulliparous PE is a leading cause of maternal death, surpassing woman was admitted to a regional hospital to deliver preeclampsia–eclampsia, hemorrhage, and infection her first child. Antenatal care including blood pressure in recent years [1,2]. Undiagnosed PE has a mortality measurements was unremarkable except a 1+ protein- rate as high as 30%, which falls to near 0.7% if the uria on urine dipstick test at 36 weeks. The labor course condition is diagnosed and treated appropriately. Spiral was smooth, and she delivered a healthy 2,780-g infant computed tomography offers the most cost-effective at 0713. After delivery, the patient experienced nausea method for diagnosing this potentially fatal condition and epigastric pain at 09:00. Medications were provided [5]. Since the key to surviving PE is early diagnosis and but in vain. Sudden onset of nausea with vomiting and prompt treatment, in the pregnant or postpartum patient disorientation of consciousness occurred at noon. She with sudden onset of cardiorespiratory distress in whom was immediately transferred to a tertiary medical center a clear diagnosis is not afforded by examination of the for further intensive care. On admission, laboratory basic parameters, therapy for the most serious diagnos- data were: alanine aminotransferase (ALT) 1,020 IU/L, tic possibilities, such as PE, should generally be instituted lactate dehydrogenase (LDH) 3,032 IU/L, and plate- while investigations progress. Obstetricians must be lets 47,000/mm3. The patient’s condition deteriorated alert to the presence of an incipient PE and aggressive in rapidly despite intensive medical care. On the following the use of diagnostic tests and treatment. This may day, the glutamate-pyruvate transaminase level had mean beginning heparin or tissue plasminogen activator increased to 5,065 IU/L and LDH had climbed to 9,182 therapy when conservative management with heparin IU/L. The patient’s family requested discharge because fails. There may be a place for other treatment options of her near-death status due to multisystemic organ including embolectomy and localized catheter-directed failure. This case is medicolegally closed because the thrombolytic therapy in the treatment of severe, life- lawyers of both parties have reached settlement by threatening PE [6–8]. Inherited or acquired thrombo- mediation outside the law court. The victim’s heirs were philia should be assessed to determine the exact cause awarded approximately NT $2,000,000 no-fault com- of PE. pensation on moral grounds.

2 Taiwanese J Obstet Gynecol • March 2006 • Vol 45 • No 1 Litigation in Obstetrics

Comment this disease so that proper medical treatment can be Hemolysis, elevated liver enzymes, and low platelet offered, thus avoiding morbidity and death. count (HELLP) syndrome is a unique syndrome of preg- nancy that can be a challenge to diagnose because the signs and symptoms of preeclampsia may appear Amniotic Fluid Embolism late in its disease course or not at all, and because the presentation can resemble other disorders. Many pa- In AFE, the obstetrician is a scapegoat, i.e. “one who tients do not, in fact, have hypertension [15,16]. The bears the blame for unexpected, unavoidable, un- lowest platelet count and the peak ALT and LDH values predictable, unpreventable and unacceptable tragic occur within 72–96 hours after delivery. The presence of misfortunes of (placenta–fetus) foreign antigen– nausea with or without vomiting, right upper-quadrant antibody-mediated anaphylactic reaction”. In some or epigastric pain, or both in association with exces- cases, unfortunately, despite our best intentions, pa- sive laboratory values including LDH (> 1,400 U/L), tients die, and the clinicians involved often become the ALT (> 150 IU/L), uric acid (> 7.8 mg/dL), serum creati- “secondary victims”. nine (> 1.0 mg/dL), and 4+ urinary protein can be used to identify the patient at high risk for significant ma- Case ternal morbidity [17]. An ALT of more than 2,000 IU/L, Hsieh et al reported the case of a 34-year-old woman, LDH above 3,000 IU/L, or presence of hypoglycemia gravida 7, para 3, who went to hospital to deliver her is the major marker for maternal death [15,16]. The fourth child [25]. Four hours after admission, the pa- normal time course for resolution of laboratory ab- tient experienced a sudden intrapartum episode of normalities is about 4 days, and delayed recovery needs respiratory distress, cyanosis, and disorientation of to be evaluated for additional causes. consciousness. An emergency cesarean delivery was A recent Cochrane review on the subject concluded performed. Cardiorespiratory resuscitation was per- that there is insufficient evidence to determine whether formed because of acute circulatory and respiratory adjunctive corticosteroid use in women with HELLP collapse. Arteriovenous extracorporeal membrane syndrome decreases major morbidity [18,19]. However, oxygenation (ECMO) and intra-aortic balloon counter- experimental antepartum and/or postpartum high-dose pulsation procedures were instituted after the cesarean corticosteroid therapy (> 24 mg/day; 10 mg dexa- delivery and maintained 40 hours postpartum. The pa- methasone iv q6h for 2 doses followed by 6 mg q6h tient was discharged after 24 days of hospitalization. for 2–4 doses) and/or plasma exchange may be life- Amniotic fluid elements were observed in the blood saving approaches in selected cases without single or from the central vein. In addition, another case has multiple organ injuries [20–23]. Although HELLP syn- recently come to light with sudden onset of AFE during drome takes an indolent course, in some patients, the preoperative preparation for elective cesarean delivery disease may progress with astonishing and frighten- because of placenta previa at term. At the time of ing rapidity and culminate in maternal death. In re- writing this article, the patient is still in a vegetative viewing all the records of patients who died from HELLP state despite aggressive ECMO support after this no- syndrome, Weinstein found that the most common fault tragic event. family of drugs that all patients had received in the days before death was not an antihypertensive or magnesium Comment sulfate but an antacid for “heartburn” [15]. HELLP AFE appears to be initiated after maternal intravascular syndrome could be considered “gastrointestinal exposure to fetal and/or placental materials contained preeclampsia.” [15,24]. Therefore, if a patient complains in amniotic fluid (which itself appears to be nontoxic), of heartburn, epigastric or right upper-quadrant pain, and anaphylactic reaction plays a role in the pathophysio- clinical and laboratory evaluations are warranted. logy of AFE [25–28]. Mediators including amniotic fluid Currently, much research is evaluating the immuno- contents, plasma components, pulmonary endothe- logic basis for the disease, but little progress in clinical lium, and leukocytes (mast cells) are responsible for the management has ensued. Delay in diagnosis is associ- anaphylactic reaction of AFE. The chronology of symp- ated with mortal consequences. The cure for preeclamp- toms and signs of AFE can be divided into three phases. sia will remain elusive but we need to improve early The first phase has three predominant manifestations: recognition, close medical supervision, safe prolonga- respiratory distress and cyanosis; hypotension, pulmo- tion of pregnancy, and timely induction of labor when nary edema, and shock; and seizures or coma. The necessary. We hope this case presentation can alert second phase includes coagulopathy and hemorrhage, other health-care providers to the serious nature of while the third phase includes multisystemic organ

Taiwanese J Obstet Gynecol • March 2006 • Vol 45 • No 1 3 M.M. Chou

injury. All these occur during labor or cesarean delivery from urinary system sequelae (2004, lawbank website). or 30 minutes after with no other Additionally, we are aware of a case of maternal explanation for the findings. For confirmation of AFE, death due to massive hemorrhage after vaginal delivery fetal constituents, for example, fetal squames, lanugo, in a local clinic. In-depth review of the cause of death by vernix caseosa, mucin, or bile granules, must be found autopsy revealed that placenta accreta was responsible. in the maternal lung. It is important to identify the However, in an astonishing case, a patient with pla- responsible antigen and mediators, if possible. Mast centa previa percreta and bladder involvement survived cell degranulation products include histamine, which after two major surgical procedures and massive blood peaks in 10 minutes and has a half-life of 2 minutes, and transfusions at 39 weeks of gestation despite total esti- tryptase, which, in contrast, peaks 1–2 hours after the mated blood loss of 47,000 mL [29]. These case reports inciting event and has a half-life of 2 hours. Thus, tryp- raise concern that the danger of bleeding from placenta tase has proved to be a better laboratory marker for previa accreta is very obvious. Therefore, “fire drills” are anaphylaxis than histamine [27]. The more difficult necessary to practice a protocol for the management of task of searching for specific immunoglobulin E to fetal massive intrapartum and postpartum hemorrhage in proteins could rely on polyacrylamide gel electrophore- every maternity unit, and it is recommended that the sis and Western blot techniques [27]. The obstetrician most experienced obstetrician must be available under must send blood samples from a wedged catheter to these circumstances. the laboratory to confirm the clinical diagnosis when An allegation of failure to diagnose placenta accreta unexpected AFE occurs during labor or delivery. This is a relatively common cause of medical negligence procedure is important in a setting where the manifesta- claims against general practitioners. Therefore, every tions of AFE closely resemble that of PE. obstetrician should be aware of the efficacy, limitation, Pharmacologic treatment includes inotropics and and pitfalls of current imaging techniques in diagnosing hydrocortisone 500 mg intravenously (iv) every 6 hours. placenta accreta [29–31]. Early antepartum identifi- In addition, rapid and aggressive plasma therapy re- cation of placenta accreta/increta/percreta offers oppor- places consumed coagulation factors and inhibitors, tunity for termination of life-threatening pregnancy or and also presumably supplied angiotensin-converting alteration in the surgical approach to achieve a planned enzyme, aminopeptidase P, and carboxypeptidase N, delivery with the use of pelvic artery balloon occlusion which quickly inactivate bradykinin. The rapid clearance techniques in viable late pregnancy and/or internal iliac of vasodilatory kinins probably contributes to preventing artery embolization for non-viable pregnancy in the the extravascular leakage that is usually observed in first and early second trimesters, thus diminishing blood AFE [28]. The last resort of life-saving treatment is to loss, morbidity, and mortality [32]. First-trimester use ECMO and intra-aortic balloon counterpulsation. diagnosis should therefore be pursued aggressively in Anaphylactic reactions induced by AFE may be transient any woman with placenta previa and an at-risk history and resolve within a few hours. Cardiopulmonary by- of previous uterine surgery [29,33,34]. pass plays a major role in saving the lives of patients by providing a few critical hours to maintain the cardio- vascular status of AFE victims [25]. Uterine Rupture

Case Placenta Percreta A 39-year-old woman, gravida 6, para 5, was transferred to a tertiary medical center because of postpartum he- An in vitro fertilization case with antenatal undiag- morrhage. She had a significant history of three success- nosed placenta percreta causing uterine rupture with ful vaginal births after cesarean deliveries (VBAC). On fatal hemoperitoneum at 23 weeks of gestation has December 29, 2003, the patient experienced shoulder been reported [29]. An approximately NT$ 750,000 dystocia (head–body interval delivery time > 30 minutes, settlement was reached through mediation in 1986. infant weight 4,450 g), and neonatal death occurred In contrast, a victim of placenta percreta with bladder during her fourth attempt at VBAC at 40 weeks of gesta- involvement underwent emergency total abdominal tion in a local hospital. The patient underwent emer- and partial cystectomy because of pre- gency laparotomy that revealed hemoperitoneum and mature rupture of membranes and antepartum lower uterine segment rupture. Total abdominal hyste- hemorrhage at 19 weeks of gestation. The patient was rectomy was performed, followed by postoperative pel- awarded no-fault compensation with the hospital vic angiographic embolization due to intraoperative paying NT$ 2,180,000 for her psychosocial suffering difficulty in achieving satisfactory deep pelvic hemosta-

4 Taiwanese J Obstet Gynecol • March 2006 • Vol 45 • No 1 Litigation in Obstetrics

sis. The patient was admitted to the intensive care unit heart rate pattern. Ultrasonographic measurement of (ICU) for further care, and discharged in a stable con- a lower uterine segment thickness of more than 2.0– dition. The attending obstetrician was not sued for mal- 3.5 mm has an excellent negative predictive value for the practice negligence. risk of uterine defect [41]. A standardized cesarean technique should be used Comment to avoid the potential severe consequences of uterine A trial of labor after prior cesarean delivery is associ- rupture, cesarean scar pregnancy, and invasive placent- ated with a greater perinatal risk than elective repeated ation in subsequent . Opening the cesarean delivery without labor, although absolute risks with a knife, the operator makes a small transverse are low. A rate of uterine rupture of 6.2/1,000 trials of incision in the appropriate mid-portion of the uterine labor (880 uterine ruptures in 142,075 trials of labor) muscle (the level of the incision depends upon the has been reported [35]. Risk factors for uterine rupture level of the head and should be as near as practicable to are increased maternal age, history of two previous the widest diameter) and cuts down to the amniotic cesarean deliveries, postpartum fever after the previ- membrane, extending the incision laterally by means of ous cesarean delivery, short interdelivery interval, curved scissors. A final enlargement can be made by and a history of classical incision [36–38]. Therefore, blunt expansion using index fingers into the uterine clinicians should be particularly careful in patients with incision and stretching it from side to side. We do not prior cesarean section or a weakened uterus. The rate recommend making the whole width of the uterine of uterine rupture related to VBAC is 2.4% in pro- aperture by crude tearing, which tends to leave ragged staglandin (PG)-induced labor, in comparison with margins, rather than a regular square edge that allows 0.4–0.52% in nonaugmented, spontaneous labor and accurate repair. When stitching the uterus, care is taken 0.77–1.0% in patients receiving oxytocin augmentation to avoid the uterine decidua. It is most important to use [39]. Therefore, the risk of uterine rupture during VBAC continuous running sutures and maintain good tension attempts is substantially increased with the use of with appropriate insertion and accurate placing of the various prostaglandins, and they should be used with inner row of sutures (ensuring that the shaggy margins caution. Aslan et al reported a higher rate of uterine of the decidua are not interposed between the muscle rupture (9.7%) in patients undergoing a trial of labor tissues), followed by a second layer with wide mus- with a history of previous cesarean delivery in which cular bite closure. If the decidua has not been carefully labor was induced with misoprostol [40]. In contrast, excluded, a gutter runs along the scar on its inner mu- accidental uterine rupture of an unscarred uterus fol- cosal surface. This constitutes a permanent weakness lowing with misoprostol occurs in 0.1– and, in a subsequent pregnancy, favors the occurrence 1% and only rare cases have been attributed to the use of a hernia of the membrane, which gradually protrudes of misoprostol. There is a concern about the use of through the wall. That is the usual manner in which the approved drugs for unlabeled indications. Prescribing a uterus ruptures in a subsequent pregnancy when labor Food and Drug Administration (FDA)-approved drug occurs. In addition, if in a subsequent pregnancy the for an indication that does not have FDA sanction often placenta implants over the scar, the destructive action causes concern and confusion among practitioners. of the chorionic villi on the fibrous tissue becomes The FDA Drug Bulletin 1982;12(1) states: ‘Once a product pronounced and very decidedly predisposes to cesarean has been approved for marketing, a physician may scar pregnancy and/or placenta accreta. prescribe it for use or in treatment regimens or patient populations that are not included in approved labeling. Such “unapproved” (unlabeled) uses may be appropri- Maternal Death Due to Septic Shock after ate and rational in certain circumstances, and may, in Second-Trimester Genetic fact, reflect approaches to drug therapy that have been extensively reported in medical literature.’ We are aware of at least three maternal deaths caused is of limited or no value for silent by septic shock after genetic amniocentesis in Taiwan. uterine rupture, which is difficult to diagnose. Warning Accordingly, we report a recent case of ultimately fatal signals are often absent and the rupture is clinically maternal sepsis caused by chorioamnionitis-associated suspected only after it has occurred. The classic des- septic shock in a twin pregnancy after mid-trimester cription of a decrease in uterine tone or the cessation of genetic amniocentesis. labor at the time of uterine rupture has been refuted by intrauterine pressure monitoring data. The most Case common single warning signal is an abnormal fetal A patient, who had undergone amniocentesis 2 days

Taiwanese J Obstet Gynecol • March 2006 • Vol 45 • No 1 5 M.M. Chou

before presentation at a local hospital because of twin tion of intestinal germs due to inadvertent intestinal pregnancy with an abnormal maternal serum test at 18 penetration, and, rarely, after use of deficiently sterilized weeks of gestation, was referred to a tertiary medical equipment. The principal pathogens responsible for center due to febrile morbidity (39°C). At that time, she septic shock in obstetric patients are Escherichia coli, had a white blood cell (WBC) count of 17,500/mm3, Clostridium spp, Bacteroides spp, Klebsiella–Enterobacter– a C-reactive protein (CRP) of 3.4, and a urine routine Serratia group, Proteus spp, and Pseudomonas aeruginosa. WBC 3+ with bacteria. She was empirically given iv Other pathogens are atypical organisms such as Gram- cefazolin 1 g 6 hourly, gentamycin 80 mg 8 hourly, and positive aerobic bacteria and Gram-negative anaerobic metronidazole 250 mg 8 hourly after consultation with bacteria. the infectious diseases physician. Most obstetric patients with bacteremia respond The patient was transferred to the ICU because of promptly to iv antibiotic therapy. In isolated instances, worsening respiratory distress caused by pulmonary however, the patient may develop the life-threatening edema. She delivered two stillborn fetuses weighing complication of septic shock, which has an alarmingly 230 and 200 g, respectively. Unfortunately, postpartum high mortality rate. Mabie and Barton reported 18 hemorrhage developed after uterine curettage for re- patients with septic shock in pregnancy. The causes of tained placenta. There was a rapid decompensation shock were pyelonephritis (6), chorioamnionitis (3), in both the clinical and laboratory status (CRP 20.5, postpartum endometritis (2), toxic shock (2), and one WBC 24,100/mm3) consistent with septic shock and each of septic abortion, ruptured appendix, ruptured disseminated intravascular coagulopathy. Sudden onset ovarian abscess, necrotizing fascitis, and bacterial of ventricular tachycardia was followed by asystole, and endocarditis [46]. the patient died despite aggressive cardiorespiratory The management of the gravid patient with septic resuscitation. Biopsy of the placenta revealed acute shock due to chorioamnionitis presents a particularly chorioamnionitis. Bacterial culture from the placenta difficult problem for the obstetrician [47]. Sepsis after specimen grew coagulase-negative staphylococci, and prenatal diagnosis can be devastating. Postamnio- no bacteria grew from all other blood cultures, urine centesis chorioamnionitis is usually managed with culture, and cervix culture. The exact etiology of the antibiotics and termination of pregnancy, if necessary sepsis in this case is unclear since autopsy was not per- to prevent maternal sepsis and related morbidity and formed. The temporal relationship between the amnio- mortality. Management strategy suggests that antibiotic centesis procedure and the maternal/fetal death sug- treatment and expectancy may be an option in selected gests the possibility of an association. This case was cases of postamniocentesis chorioamnionitis in multiple medicolegally closed because the lawyers of the three pregnancies. Surgery in the presence of deteriorating parties reached a settlement by mediation outside the maternal vascular function is clearly hazardous. It may law court. The victim’s heirs were awarded approximately be necessary, however, precisely because surgical re- NT$ 2,800,000 no-fault compensation on moral moval of the focus of infection is the only way to grounds. stabilize the critically ill patient. Although immediate evacuation of the uterus by dilatation or by hystero- Comment tomy may be considered, these aggressive efforts did It is generally accepted that severe complications are not prevent maternal death in three reported cases in underreported [42]. Obstetric care providers and genetic the literature [43–45]. In rare cases, simple evacuation counselors should be aware of potential serious mater- of the uterine content is insufficient. However, the nal morbidity and mortality that may occur after a decision to perform a hysterectomy is difficult, especially needle is inserted into the gravid uterus, and any patient in the case of a young nulliparous woman. Moreover, complaints thereafter should be regarded seriously the condition develops so rapidly that it does not always [43]. The incidence of chorioamnionitis is 3.7/1,000 allow time for a proper diagnostic work-up and pre- cases after amniocentesis. Infection is usually mild to paration of the patient. moderate. There are, however, rare sporadic reports in The most effective treatment for maternal sepsis is the literature describing cases of postamniocentesis prevention. Preventive measures that may be employed sepsis with devastating results [43–45]. Review of the in the pre-amniocentesis period include stabilization literature revealed 10 more cases of sepsis after trans- of the patient’s preexisting medical disease, sterile abdominal genetic amniocentesis. Contamination of amniocentesis pack, aseptic skin care, aseptic placement the uterine cavity after prenatal genetic diagnostic pro- of the amniocentesis needle, and prophylactic antibio- cedures may occur through ascending infection from tic treatment in selected patients. The obstetrician’s the patient’s endogenous vaginal flora, direct inocula- attention should be directed primarily towards early

6 Taiwanese J Obstet Gynecol • March 2006 • Vol 45 • No 1 Litigation in Obstetrics

identification of patients with infection-related com- with an increased risk of [49]. This is plications. Appropriate informed consent for genetic especially important given the number of lawsuits amniocentesis should include an additional statement involving shoulder dystocia cases. We are aware of a such as: “rarely there may be severe maternal compli- victim of shoulder dystocia with brachial plexus injury cations from amniocentesis related to infection and/or awarded no-fault compensation of NT$ 2,800,000 sepsis, even if the proper precautions are taken” [43]. because of suspected deviation from the standard of care due to underestimation of EFW (EFW 3,100 g, act- ual birth weight 4,136 g) and inappropriate maneuvers Shoulder Dystocia to deliver this macrosomic infant (March 11, 2004, lawbank website). Case Shoulder dystocia followed by permanent brachial A 34-year-old woman, gravida 2, para 1, had gestational plexus injury or mental impairment is one of the leading diabetes mellitus controlled by diet. Her membranes causes of malpractice allegations. Erb’s palsy (C5–C7) ruptured at 06:40, and she was admitted to the hospital is a paralysis of shoulder and arm muscles resulting in at 38 weeks of gestation. Unfortunately, a shoulder a hanging upper arm that may be extended at the el- dystocia occurred (head–body delivery time about 2 bow. Duchenne palsy (C7–T1) is a palsy including the minutes) after the fetal head was delivered at 13:38. hand (clawhand deformity). Most brachial plexus in- The first stage of labor was 1 hour 30 minutes, and the juries are transient and recover completely by 13 months, second stage lasted only 8 minutes. The shoulder dystocia although 5–22% become permanent. may have been caused by failure to allow the natural In the typical brachial plexus case, plaintiffs always molding of the baby’s shoulders to the contours of claim that the birth attendant must have exerted exces- the birth canal because of the short second stage and sive lateral traction on the infant’s neck during delivery chest-to-head disproportion in the diabetic woman. for a permanent injury to have occurred. However, there The Apgar scores were 4, 7, and 8 at 1, 5, and 10 is solid evidence for the possibility of an in utero injury minutes, respectively. The baby weighed 3,824 g. The and that maternal endogenous factors, namely uterine infant had suffered an evulsive brachial plexus injury expulsive force and maternal pushing, can cause injury. due to shoulder dystocia, resulting in persistent Erb’s Furthermore, Dr. Raymond Jennet observed that bra- palsy in the right arm at the age of 12 months. A lawsuit chial plexus injuries do occur in the absence of shoulder was filed on the child’s behalf, and the case had not dystocia during vaginal deliveries [51,52]. been settled at the time of writing. The obstetrician should call for help and follow all management rules when shoulder dystocia occurs. The Comment standard obstetrical maneuvers used for its relief Shoulder dystocia is one of the obstetrician’s greatest must be well rehearsed. The most popular techniques fears, and the most dreadful complication because of include: McRobert’s maneuver and suprapubic pres- five “U” characteristics: unexpected, unpredictable, sure, followed by Woods corkscrew maneuver or deli- unpreventable, unavoidable, and unacceptable [48]. very of the posterior shoulder, then, in order, Rubin Most cases cannot be predicted or prevented because maneuver, , abdominal rescue (after there are no accurate methods for identifying which failed cephalic replacement), , and frac- fetuses will develop this complication. Ultrasound mea- ture of the clavicle or cleidotomy. Prompt assessment surements for estimating macrosomia have limited and management of shoulder dystocia and preparation accuracy. The absolute error of estimated fetal weight to maximize the efficiency of shoulder dystocia maneu- (EFW) formulas ranges from 7% to 10% within 1 week vers are critical. Documentation of the appropriate use of delivery, and may be as high as 15% [49]. Thus, when of maneuvers to relieve shoulder dystocia demonstrates calculated on a 4,000-g infant, the margin of error standard of care practice, thereby decreasing the poten- is 600 g. Planned cesarean deliveries on the basis of tial for successful malpractice allegations [51]. suspected macrosomia are not a reasonable strategy In conclusion, most obstetricians will potentially be [50]. However, planned cesarean section may be rea- sued if they practice long enough. Similarly, exhortations sonable for the nondiabetic woman with an estimated for birth attendants to become either more educated fetal body weight of more than 4,500 g or for the or more sensitive are unlikely to reduce almost half of diabetic woman at more than 4,000 g. these pregnancy-related deaths, which are potentially Recently, Mehta et al reported their findings in a unpreventable. Therefore, the obstetrician has become unique article demonstrating that errors in EFW, and an endangered species in the era of medical-legal claims, specifically underestimation of EFW, are not associated fault and no-fault financial compensation, unaffordable

Taiwanese J Obstet Gynecol • March 2006 • Vol 45 • No 1 7 M.M. Chou

medical liability awards and increasing professional 9. Bonnar J. Can more be done in obstetric and gynecologic liability premiums, and injustice of National Health practice to reduce morbidity and mortality associated with Insurance. We as a society must get on with the hard, venous thromboembolism. Am J Obstet Gynecol 1999;180: 784–91. time-consuming task of medical injury compensation 10. Lockwood CJ. Maternal mortality: we must do better. Contemp act legislation and develop a more rational and fair Ob Gyn 2002;1:10–4. system of medical jurisprudence [53]. Otherwise, the 11. Seligsohn U, Lubetsky A. Genetic susceptibility to venous pattern of obstetricians reducing their obstetric practice thrombosis. N Engl J Med 2001:344:1222–31. or discontinuing obstetric practice entirely will continue 12. Gerhardt A, Scharf RE, Beckmann MW, et al. Prothrombin and medical students will no longer be interested in and factor V mutations in women with a history of thrombosis pursuing the profession that we all love. Until then, who during pregnancy and the puerperium. N Engl J Med 2000; will deliver our grandchildren [53,54]. 342:374–80. 13. Lockwood CJ. Preventing VTE: part 3-the pregnant patient. Contemp Ob Gyn 2005;May 1. 14. Lockwood CJ. Inherited thrombophilias in pregnant patients: Addendum detection and treatment paradigm. Obstet Gynecol 2002;99: 333–41. Often, the factual information available about the 15. Weinstein L. Syndrome of hemolysis, elevated liver enzymes, medical-legal closed cases presented here is incomplete. and low platelet count: a severe consequence of hyperten- Thus, it may not always be possible to discuss all of the sion in pregnancy. Am J Obstet Gynecol 2005;193:859–63. elements of negligence or nuances involved in a given 16. Sibai BM, Ramadan MK, Usta I, Salama M, Mercer BM, Friedman SA. Maternal morbidity and mortality in 442 preg- situation. The outcomes described may not reflect the nancies with hemlolysis, elevated liver enzymes and low current standard of care or the best practice in obstet- platelets. Am J Obstet Gynecol 1993;169:1000–6. rics. What these cases do represent are the types of 17. Martin JN Jr, May WL, Magann EF Terrone DA, Rinehart BK, clinical situations in maternal–fetal medicine that typi- Blake PG. Early risk assessment of severe preeclampsia: cally result in litigation. We hope that some of the cases admission battery of symptoms and laboratory tests to described have merit. predict likelihood of subsequent significant maternal morbidity. Am J Obstet Gynecol 1999;180:1407–14. 18. Matchaba P, Moodley J. Corticosteroids for HELLP syndrome in pregnancy. Cochrane Database Syst Rev 2004;CD002076. References 19. Fonseca JE, Mendez F, Catano C, Arias F. Dexametha- sone treatment does not improve the outcome of women 1. Chang J, Elam-Evans LD, Berg CJ, Herndon J, Flowers L, Seed with HELLP syndrome: a double-blind, placebo-controlled, KA, Syverson CJ. Pregnancy-related mortality surveillance- randomized clinical trial. Am J Obstet Gynecol 2005;193: United States, 1991–1999. MMWR Surveill Summ 2003; 1591–8. 52:1–8. 20. O’Brien JM, Milligan DA, Barton JR. Impact of high-dose 2. de Swiet M. Maternal mortality: confidential enquiries into corticosteroid therapy for patients with HELLP syndrome. maternal deaths in the United Kingdom. Am J Obstet Gynecol Am J Obstet Gynecol 2000;183:921–4. 2000;182:760–6. 21. Martin JN, Thigpen BD, Rose CH, Cushman J, Moore A, May 3. Panting-Kemp A, Geller SE, Nguyen T, Simpson L, Nuwayhid WL. Maternal benefit of high-dose intravenous corticosteroid B, Castro L. Maternal deaths in an urban perinatal network, therapy for HELLP syndrome. Am J Obstet Gynecol 2003;189: 1992–1998. Am J Obstet Gynecol 2000;183:1207–12. 830–4. 4. Berg CJ, Harper MA, Atkinson SM, et al. Preventability of 22. Sibai BM, Barton JR. Dexamethasone to improve maternal pregnancy-related deaths: results of a state-wide reviews. outcome in women with hemolysis, elevated liver enzymes, Obstet Gynecol 2005;106:1228–34. and low platelets syndrome. Am J Obstet Gynecol 2005;193: 5. Doyle NM, Ramirez MM, Mastrobattista JM, Monga M, 1587–90. Wagner LK, Gardner MO. Diagnosis of pulmonary embolism. 23. Martin JN, Files JC, Blake PG, Perry KG, Morrison JC, Norman A cost-effectiveness analysis. Am J Obstet Gynecol 2004;191: PH. Postpartum plasma exchange for atypical preeclampsia- 1019–23. eclampsia as HELLP syndrome. Am J Obstet Gynecol 1995;172: 6. Bechtel JJ, Mountford MC, Ellinwood WE. Massive pulmonary 1107–27. embolisim in pregnancy treated with tissue plasminogen 24. Isler CM, Rinehart BK, Terrone DA, Martin RW, Magann EF, activator. Obstet Gynecol 2005;106:1156–8. Martin JN Jr. Maternal mortality associated with HELLP 7. Marty AT, Hilton FL, Spear RK, Greyson B. Postcesarean syndrome. Am J Obstet Gynecol 1999;181:924–8. pulmonary embolism, sustained cardiopulmonary resusci- 25. Hsieh YY, Chang CC, Li PC, Tsai HD, Tsai CH. Successful tation, embolectomy, and near-death experience. Obstet application of extracorporeal membrane oxygenation and Gynecol 2005;106:1153–5. intra-aortic balloon counterpulsation as lifesaving therapy 8. Bechtel, JJ, Mountford MC, Ellinwood WE. Massive pulmonary for a patient with amniotic fluid embolism. Am J Obstet embolism in pregnancy treated with catheter fragmentation Gynecol 2000;183:496–7. and local thrombolysis. Obstet Gynecol 2005;106:1158–60. 26. Clark SL, Hankins GDV, Dudley DA, Dildy GA, Porter TF.

8 Taiwanese J Obstet Gynecol • March 2006 • Vol 45 • No 1 Litigation in Obstetrics

Amniotic fluid embolism: analysis of the national registry. 39. Zelop CM, Shipp TD, Repke JT, Cohen A, Caughey AB, Am J Obstet Gynecol 1995;172:1158–69. Lieberman E. Uterine rupture during induced or augmented 27. Benson MD, Lindberg RE. Amniotic fluid embolism, anaphy- labor in gravid women with one prior cesarean delivery. Am laxis and tryptase. Am J Obstet Gynecol 1996;175:737–8. J Obstet Gynecol 1999;181:882–6. 28. Robillard J, Gauvin F, Molinaro G, Leduc L, Adam A, Rivard 40. Aslan H, Unlu E, Agar M, Ceylan Y. Uterine rupture associated GE. The syndrome of amniotic fluid embolism: a potential with misoprostol labor induction in women with previous contribution of bradykinin. Am J Obstet Gynecol 2005;193: cesarean delivery. Eur J Obstet Gynecol Reprod Biol 2004;113: 1508–12. 45–8. 29. Chou MM. Prenatal diagnosis and perinatal management of 41. Cheung VY. Sonographic measurement of the lower ute- placenta previa accrea: past, present and future. Taiwanese J rine segment thickness in women with previous caesarean Obstet Gynecol 2004;43:64–71. section. J Obstet Gynaecol Can 2005;27:674–81. 30. Chou MM, Ho ESC, Lee YH. Prenatal diagnosis of placenta 42. Thorp JA. Maternal death after second-trimester genetic previa accreta by transabdominal color Doppler ultrasound. amniocentesis. Obstet Gynecol 2005;106:409. Ultrasound Obstet Gynecol 2000;15:28–35. 43. Thorp JA, Helfgott AW, King EA, Aaron A, Minyard AN. 31. Chou MM, Tseng JJ, Ho ESC, Hwang JI. Three-dimensional Maternal death after second-trimester genetic amniocente- color power Doppler imaging in the assessment of utero- sis. Obstet Gynecol 2005;105:1213–5. placental neovascularization in placenta previa increta- 44. Elchalal U, Shachar IB, Peleg D, Schenker JG. Maternal percreta. Am J Obstet Gynecol 2001;185:1257–61. mortality following diagnostic second-trimester amnio- 32. Chou MM, Hwang JI, Tseng JJ, Ho ESC. Internal iliac artery centesis. Fetal Diagn Ther 2004;19:195–8. embolization before hysterectomy for placenta accreta. J Vasc 45. Ayadi S, Carbillon L, Varlet C, Uzan M, Pourrist JL. Fatal Interv Radiol 2003;14:1195–9. sepsis due to Escherichia coli after second-trimester amnio- 33. Chou MM, Tseng JJ, Ho ESC Usefulness of gray-scale centesis. Fetal Diagn Ther 1998;13:98–9. ultrasound and complementary color Doppler ultrasound in 46. Mabie WC, Barton JR. Septic shock in pregnancy. Obstet the prenatal diagnosis of placenta previa accrete. Am J Obstet Gynecol 1997;90:553–61. Gynecol 2005;192:1349–50. 47. Harry W, Foley MR. Treating septic shock in ob/gyn patients. 34. Tseng SW, Lin CH, Hwang JI, Chen WH, Ho ESC, Chou MM. Contemp Ob Gyn 2005;May 1. Experience with conservative strategy of uterine artery 48. Quzounian JG, Gherman RB. Shoulder dystocia: are historic embolization in the treatment of placenta percreta in the first risk factors reliable predictors? Am J Obstet Gynecol 2005;192: trimester of pregnancy. Taiwanese J Obstet Gynecol 2006 (in 1933–8. press). 49. Mehta SH, Blackwell SC, Hendler I, et al. Accuracy of esti- 35. Chauhan SP, Martin JN, Henrichs CE, Morrison JC, Magann mated fetal weight in shoulder and neonatal birth injury. Am EF. Maternal and perinatal complications with uterine J Obstet Gynecol 2005;192:1877–81. rupture in 142,075 patients who attempted vaginal birth 50. Gonen R, Bader D, Ajami M. Effects of a policy of elective after cesarean delivery: a review of the literature. Am J Obstet cesarean delivery in cases of suspected fetal macrosomia on Gynecol 2003;189:408–17. the incidence of brachial plexus injury and the rate of cesarean 36. Ofir K, Sheiner E, Levy A, Katz M, Mazor M. Uterine rupture: delivery. Am J Obstet Gynecol 2000;183:1296–300. risk factors and pregnancy outcome. Am J Obstet Gynecol 51. Minkin MJ. A no-fault approach to shoulder dystocia. Contemp 2003;189:1042–6. Ob Gyn 2004;Dec. 1. 37. Landon MB, Hauth JC, Leveno KJ, et al. Maternal and 52. Jennett RJ, Tarby TJ, Kreinick CJ. Brachial plexus palsy: an old perinatal outcomes associated with a trial of labor after prior problem revisited. Am J Obstel Gynecol 1992;166:1673–7. cesarean delivery. N Engl J Med 2004;351:2581–9. 53. MacLennan A, Nelson KB, Hankins G, Speer M. Who will 38. Macones GA, Cahill A, Stamilio DM, Ratcliffe S, Stevens E, deliver our grandchildren? Implications of cerebral palsy Sammel M, Peipert J. Obstetric outcomes in women with litigation. JAMA 2005;294:1688–90. two prior cesarean deliveries: is vaginal birth after cesarean 54. Laros RK. Presidential address: Medical-legal issues in delivery a viable option? Am J Obstet Gynecol 2005;192: obstetrics and gynecology. Am J Obstet Gynecol 2005;192: 1223–8. 1883–9.

Taiwanese J Obstet Gynecol • March 2006 • Vol 45 • No 1 9