Abstract
Abstract
Introduction
Case
A 33-year-old woman, G3P2-0-0-2 parturient at 33 weeks' gestation, American Society of Anesthesiologists physical status II, was admitted to an obstetrical ward for elective cesarean section. She was evaluated at an outlying facility, was found to be at high risk for mortality and morbidity, and was referred to the current facility for evaluation and delivery of twins. The patient's past medical history was significant for hyperthyroidism and two prior cesarean sections. During prenatal evaluations, ultrasound revealed that the patient had placenta increta; therefore, consent was obtained for the possibility of a hysterectomy after cesarean section.
On the day of her admission, the patient underwent placement of bilateral hypogastric balloons by interventional radiology to aid intraoperative hemostasis.2,3 The patient was prepared and draped in the normal sterile fashion. General endotracheal anesthesia was induced with sodium thiopental and succinylcholine, and anesthesia was maintained with sevoflurane. The patient was monitored with an indwelling 20-gauge radial arterial catheter, and venous access was obtained via a right internal jugular Shiley catheter, all of which had been placed preoperatively. After the cesarean delivery, the patient underwent a cesarean section, hysterectomy, and cystoscopy with a temporary bilateral stent placement. The patient gave birth to healthy diamniotic, dichorionic twins. Twin A was a male neonate in the vertex position with an Apgar score of 8 and 9 at 1 and 5 minutes, respectively, with a weight of 2224 g. Twin B was a male neonate in the vertex position with an Apgar score of 9 and 9 at 1 and 5 minutes, respectively, with a weight of 2827 g.
Intraoperatively, the patient had an estimated blood loss of 5 L, despite the intermittent occlusion with iliac artery balloons, which were inflated for 10–20-minute intervals at times of maximal intraoperative bleeding and deflated repeatedly to allow for reperfusion. Ten (10) units of packed red blood cells (pRBC), 10 units of fresh frozen plasma (FFP), 20 units of cryoprecipitate, 20 units of platelets, and 1 L of albumin were administered intraoperatively to achieve hemostasis and maintain hemodynamics.
The patient was transferred from the operating room to the postanesthesia care unit and intubated with stable hemodynamics. Her postoperative laboratory values were: hemoglobin, 10.8 g/dL; hematocrit, 21.9%; platelets, 116,000/mm3; sodium, 141 mEq/L; potassium, 4.2 mEq/L; chloride, 106 mEq/L; carbon dioxide, 23 mmol/L; blood–urea–nitrogen, 9 mg/dL; creatinine, 0.8 mg/dL; and glucose, 147 mg/dL, respectively. Her arterial blood gases showed the following: pH, 7.30; partial pressure of carbon dioxide, 40 mm Hg; partial pressure of oxygen, 117 mm Hg; and bicarbonate, 20 mEq/L. Coagulation laboratory values were: prothrombin time, 14.0 seconds; international normalized ratio, 1.1 seconds; partial thromboplastin time, 28.7 seconds; and fibrinogen level, 524 mg/dL. The patient remained intubated and sedated, and was transferred to the intensive care unit (ICU).
On postoperative day 1, the patient developed hemodynamic instability. An ultrasound scan was obtained secondary to the concern of postoperative intra-abdominal hemorrhage, which revealed a developing hemoperitoneum. The patient was returned to the operating room and underwent an exploratory laparotomy, evacuation of the hemoperitoneum, ligation of bleeding sites, and placement of drains. Because of the presence of hematuria, an intraoperative consultation with the hospital's urology team was obtained, and the patient underwent an intraoperative cystoscopy. The patient lost >2 L of blood loss from the hemoperitoneum. Intraoperatively, the patient received 12 units of pRBC, 6 units of FFP, 40 units of platelets, 40 units of cryoglobulin, and 3 L of crystalloid. In the operating room, the patient also empirically received 40 μg/kg of rFVIIa, after all blood factors were infused in order to help facilitate coagulation caused by continual profuse bleeding. The bleeding decreased significantly following the administration of rFVIIa.
Following the operation, the patient was transferred back to the ICU, intubated, febrile (38.8°C), and tachycardic. She continued to be hemodynamically unstable, with continuous bleeding, despite receiving large volumes of blood components. Postoperative blood gas testing revealed mild acidosis (pH 7.22) and hypoxemia (Pa
Following the second dose of rFVIIa, the patient's prothrombin time was <10 seconds with a partial thromboplastin time of 24.7 seconds, a fibrinogen level of 580 mg/dL, and a platelet count of 119,000/mm3. The patient remained intubated in the ICU to achieve optimal hemodynamic control after these extensive blood transfusions. Her status continued to improve slowly, and she was weaned off the ventilator and extubated on postoperative day 3. On postoperative day 4, she was transferred to a ward and discharged to go home on postoperative day 9.
One (1) week after discharge, the patient returned for a visit to her obstetrician. This patient was febrile and presented with localized pinpoint pain to the right part of her chest. An initial computed tomography (CT) angiogram from an outside clinic, obtained to rule out pulmonary embolism, yielded negative results. The patient was admitted to the current authors' hospital again for a clinical follow-up and received a second CT angiogram because of her continuing pain, dyspnea, and symptoms of deep-vein thrombosis. The second CT angiogram showed pulmonary emboli, but no occlusion of the large pulmonary vasculature was noted. There was also moderate, right-sided pleural fluid collection, with an adjacent compression atelectasis. The patient underwent an ultrasound-guided right thoracentesis, which resulted in the removal of 1290 mL of serosanguineous fluid and a reduction in the size of the pleural effusion. In addition, she was started on a heparin infusion and an inferior vena cava filter was placed. An echocardiogram revealed a pericardial effusion without evidence of pulmonary hypertension.
Her general status improved, but her radiologic and clinical symptoms were persistent fluid with a right, wedge-shaped pulmonary infarction. Two (2) days later after the thoracentesis, the patient underwent a right, video-assisted, diagnostic thoracoscopy, pneumolysis, and a wedge resection of the right lower-lobe pulmonary infarct.
Results
The patient had an uncomplicated course and was discharged to go home, with complete recovery from the infarcted right lower lobe.
Discussion
The question raised in this case is if the nonspecific administration of rFVIIa contributed to the postoperative peripheral thrombosis pulmonary infarction. Given the patient's high-risk pregnancy, preventive and precautionary steps were implemented predelivery. These included placement of bilateral hypogastric balloons to aid with hemostasis and obtaining consent from the patient for a possible hysterectomy in case of massive hemorrhage. Despite all the preventive and precautionary steps taken, the patient had a total of 7 L of blood loss; required a hysterectomy and a right lung lobectomy; and received a total of 22 units of pRBC, 14 units of FFP, 50 units of platelets, 20 units of cryoprecipitate, 40 units of cryoglobulin, and 2 doses of rFVIIa to help achieve hemostasis.
The empirical use of off-label rFVIIa is controversial. Logan and colleagues discovered, after researching respective databases across the United States, that 96% of all in-hospital cases of rFVIIa use from 2000 to 2008 and 97% of cases in the year 2008 were off-label. 4 These researchers demonstrated that hospital mortality for patients who received rFVIIa was high—27% overall and as high as 40%–50% for several indications. In-hospital mortality was noted to be higher when reported in retrospective studies of rFVIIa use outside of clinical trials, compared to selective clinical trial populations. Thus, the use of rFVIIa in the hospital setting far exceeds the use for approved indications. These patterns raise concern about the use of rFVIIa for conditions in which strong supporting evidence is lacking. 4
Levi and colleagues conducted a large (4468 patients), comprehensive, cohort placebo-controlled trial of rFVIIa. The results showed that treatment with high doses of rFVIIa on an off-label basis significantly increased the risk of arterial but not venous thromboembolic events, especially among the elderly. 5
O'Connell and colleagues also showed that deep venous thrombosis, ischemic cerebrovascular accidents, and myocardial infarction were the most common adverse events associated with the use of rFVIIa. These researchers reported that most thromboembolic adverse events that followed the use of rFVIIa for unlabeled indications occurred in arterial and venous systems, often resulting in serious morbidity and mortality. 6
Leighton and colleagues analyzed retrospective case reports of amniotic-fluid embolism (AFE) from 2003 to 2009, and counted 16 patients who received rFVIIa during their hospital courses and compared these data with that of 28 patients who did not receive rFVIIa. Death, permanent disability, and full recovery occurred in 8, 6, and 2 rFVIIa patients, respectively, and 7, 4, and 17 patients who did not receive rFVIIa, respectively (risk ratio 2.2; 95% confidence interval, 1.4–3.7 for death or permanent disability versus full recovery). Leighton and colleagues concluded that rFVIIa utilization in these cohorts had significantly worse outcomes than similar patients who did not receive rFVIIa. The researchers recommended that rFVIIa be used in AFE only in patients in whom hemorrhage cannot be stopped by massive blood-component replacements. 7 Although AFE was not observed in the current patient, she did have continued hemorrhage despite the use of many factors (as mentioned above) to achieve hemostasis. It is to be noted that the initial administration of rFVIIa was only used to assist in hemostasis after all conventional factors were exhausted. A second dose of factor rFVIIa was repeated empirically when the patient did not have rapid progress while she was in the intensive care unit. 7
Lin et al. conducted a meta-analysis of 26 randomized controlled trials that analyzed the use of rFVIIa in patients without hemophilia. 8 A total of 1137 patients were given rFVIIa as an off-label prophylactic dose, and 2538 patients were given rFVIIA as an off-label therapeutic dose. Doses ranged from 5 μg/kg to many doses that totaled 360 μg/kg. The researchers found modest benefits for patients receiving rFVIIa prophylactically and no benefit for patients receiving therapeutic doses. The researchers were unable to determine what should be a recommended dose for off-label use of rFVIIa. 8
Despite this, the precise etiology of the current patient's considerable postoperative complications is elusive. The use of rFVIIa may be the primary cause; however, rFVIIa has been given many times in past cases, such as for patients with penetrating or blunt trauma, and these patients had had no postoperative complications. In addition, there have been situations in which 1 dose of rFVIIa led to many postoperative complications. It is possible that the second dose of rFVIIa caused the current patient's pulmonary emboli and infarction, requiring a lobectomy, or that this second dose actually helped saved her life. In addition, the patient's mild postoperative acidosis may have impaired coagulation 9 or reduced the effectiveness of the initial dose of the rFVIIa.10,11 Clinical trials have found that acidosis-induced coagulation impairment is often exacerbated in patients with hypothermia 12 ; however, the current patient was febrile for most of her hospital stay.
Conclusions
After reviewing past experiences and the present case, the authors recommend that rFVIIa be used almost only for the direct indications that are stated by the FDA, and that off-label use be confined to extenuating emergent circumstances in which the benefits may well outweigh the risks. It is surmised that the second dose of 90 μg/kg of rFVIIa may have contributed to right lower lobe pulmonary infarction in this obstetric patient.
Footnotes
Disclosure Statement
The authors state they have no conflicts of interest.