Failure to diagnose postpartum stroke
A 27-year-old woman came to Ob-gyn A on October 19. An ultrasound revealed that she was pregnant with twins. Her due date was estimated as May 2.
At each prenatal visit, the patient’s blood pressure was normal, urinalysis revealed a trace of protein, and she was not edematous. Her pregnancy was uncomplicated with the exception of one visit to the Labor and Delivery triage at 30 weeks gestation. She reported pelvic pressure, back pain, and difficulty urinating. She was placed on an electronic fetal monitor for three hours. She was found not to be in labor, and was discharged home in stable condition.
On March 28, Ob-gyn A admitted her for an elective cesarean delivery at 35 weeks gestation. She was extremely uncomfortable and one of the twins was breech. The patient’s blood pressure upon admission was 135/77 mm Hg. By delivery, both babies moved into the vertex position. Two healthy babies were delivered: a 4-pound, 13-ounce boy with Apgars of 7 and 9; and a 5-pound, 6-ounce boy with Apgars of 8 and 9.
On March 29, the patient’s temperature became elevated, ranging from 99.2 to 101.3 degrees. Her blood pressure was 137/74 mm Hg. The ob-gyn prescribed antibiotics “due to numerous risk factors patient has for post-op febrile morbidity.” He suspected that she might be suffering from endometritis. The patient’s temperature decreased and she was discharged on March 31. Her blood pressure was 123/76 mm Hg. She had no edema, proteinuria, or other evidence of preeclampsia during her hospital stay.
The patient called Ob-gyn A’s office the evening of April 1. She spoke with Ob-gyn B who was on call, and reported abdominal pain and bright red bleeding. Ob-gyn B told the patient that bleeding after a cesarean delivery was not unusual. He advised her to continue taking her pain medication and to call back if the pain worsened or did not diminish.
On April 2, the patient spoke with the Ob-gyn B again. She reported pain and a stiff neck. Ob-gyn B recalled that the patient was crying because of the pain. He advised her to go to the emergency department (ED). Ob-gyn B called the ED to let them know she was coming and requested that the ob-gyn resident check on her in the ED. During this call, he spoke to an ED nurse and ordered several tests including a urinalysis and CBC. He did not order the administration of IV fluids.
The first documented assessment of the patient in the ED took place at 9:55 a.m. Her temperature was 97.9 degrees; pulse 53; respirations 20; blood pressure 147/77 mm Hg; oxygen saturation 97%. At 10:25 a.m., a nurse recorded that the patient was passing clots and complained of chest pain when she moved. She also described a “heavy feeling.”
The nurse noted that the patient’s lungs were clear bilaterally and her abdomen was soft and tender with bowel sounds present. The patient was noted to be in moderate distress. She was placed on supplemental oxygen.
At 10:48 a.m., a nurse started an IV of 1000 cc normal saline. At 11:50 a.m., the nurse noted that he re-paged the ob-gyn resident to see the patient. (It was not documented when the nurse first paged the ob-gyn resident.) The nurse hung a second bag of normal saline at 11:55 a.m.
The resident, Ob-gyn C, examined the patient between 12 and 12:39 p.m. The patient’s vital signs at noon were: blood pressure 172/89 mm Hg; pulse 50; and respirations 24. The patient reported dizziness, substernal chest pressure radiating to the right side and worse when supine, nausea, headache, and occasional scotomata. Ob-gyn C’s impression was of a patient in moderate distress who was bradycardic and dyspneic with shallow breaths.
She found the patient’s abdomen was moderately tender and noted vaginal bleeding. She did not document an assessment of reflexes or a neurological exam. Ob-gyn C’s provisional diagnosis was rule out postpartum preeclampsia and possible endometritis. Her note specifically stated that she discussed her findings with Ob-gyn B. The plan was to admit the patient, start IV piperacillin/tazobactam, and obtain stat labs.
The patient’s vital signs were again documented at 12:39 p.m.: pulse 122; respirations 24; blood pressure 188/121 mm Hg. At 12:50 p.m., a nurse documented the administration of 0.1 mg of clonidine by mouth. There is no record of which physician ordered this medication. (Since this is not a medication typically ordered by ob-gyns, most witnesses presumed an ED physician ordered it.) Additionally, although it was unknown who ordered them, additional tests were ordered including an ABG, basic metabolic panel, uric acid, LDH, chest x-ray, and cardiac enzymes.
At 1:50 p.m., a nurse noted that the patient’s blood pressure was 170/96 mm Hg and her pulse was 79. The patient also continued to report a headache. (It was unknown when the patient first complained of a headache. There was no documented triage assessment and headache was not mentioned in the initial nurse’s notes.) A third bag of saline was hung at 3:05 p.m.
The floor nurse received the patient at 3:30 p.m. She noted that the patient was incontinent of a large amount of urine, agitated with slurred speech, and incoherent at times. The patient was oriented to self only, her mouth was asymmetrical, and was unable to move her left arm or leg. The nurse also noted 2+ edema in the legs and audible crackles when listening to her lungs. The patient’s blood pressure was 187/120 mm Hg and her pulse was 89. The nurse called the chief resident.
At 3:30 p.m., the chief resident — Ob-gyn D — arrived. The patient’s blood pressure was 181/120 mg Hg and then 199/105 mm Hg. Urine protein by catheterization was 2+. Ob-gyn D felt that patient had clinical signs of pulmonary edema, hypertension, and proteinuria, all of which represented a change in her initial ED evaluation and postpartum course. She suspected postpartum preeclampsia with hypoxia, and wanted to rule out pulmonary embolus and CVA.
She noted the plan to transfer the patient to the labor and delivery floor, start intravenous seizure prophylaxis with magnesium sulfate, administer furosemide intravenously, order a chest CT, and start an arterial line. She also noted that Ob-gyn B was notified and on the way.
Ob-gyn D further stated that she would not start antihypertensive medication until a CT had been performed to evaluate whether the patient had a hemorrhagic stroke, hypertensive encephalopathy, or sagittal sinus thrombosis.
Ob-gyn B’s un-timed note thereafter states that he agreed with the Ob-gyn D’s note and plan. He saw the patient, but the rest of his note was illegible. A CT of the chest obtained at 6:27 p.m. revealed pulmonary vascular congestions with interstitial and alveolar edema and bilateral pleural effusion. A CT of the head revealed a right capsular/basal ganglionic intracerebral hematoma, probably hypertensive in etiology. The neuroradiologist stated there was no evidence of a global increase in intracranial pressure. He described a halo surrounding vasogenic edema suggestive for early subacute presentation.
The patient was transferred to the ICU where she was intubated, hyperventilated, and given mannitol and antihypertensive therapy. A neurologist who evaluated the patient after the stroke said her type of brain bleed was usually hypertensive in nature. The patient’s condition improved but then deteriorated. A CT scan revealed extension of the bleeding in her brain with expansion of the previously identified hematoma.
A neurosurgeon recommended emergent craniotomy for evacuation of the hematoma. On April 3, the neurosurgeon performed a right frontotemporal parietal craniotomy along with placement of an intracranial pressure-monitoring device. A clot measuring 7.0 x 5.0 x 3.0 cm was found during surgery. Both her pre- and postoperative diagnoses were right temporal and basal ganglia intracerebral hemorrhage.
The patient underwent an additional surgery on April 4 due to elevated intracranial pressure. A repeat head CT revealed a 2-cm hematoma anterior to the previously evacuated hematoma. The neurosurgeon performed a re-exploration of the right craniotomy and evacuation of the new hematoma.
The neurosurgeon discussed the patient’s grave condition with her family. They indicated that all measures possible should be taken to keep the patient alive. It was the neurosurgeon’s opinion that the patient would likely have permanent neurological damage and/or remain in a persistent vegetative state.
On April 16, the patient began to open her eyes and follow commands. By April 21, she began weaning to a trach tube. She was discharged to a rehab facility on May 2.
According to more recent medical records, the patient has left spastic hemiparesis due to intracranial hemorrhage. She is seen by a brain injury clinic every three months and is seen by a neurologist every six months for possible seizure disorder.
A lawsuit was filed against Ob-gyn B and his group practice. The main allegation against Ob-gyn B was that he failed to elicit sufficient information from Ob-gyn C at the hospital concerning the status of his patient in the ED. As a result, the diagnosis of postpartum preeclampsia was not made or treated, causing the patient to suffer a debilitating stroke. Other allegations included failure to supervise Ob-gyn C and other hospital personnel. The hospital, Ob-gyn C, the ED physicians, and two ED nurses were also sued.
The plaintiffs retained credible experts who supported their allegations. Their ob-gyn expert testified that the patient suffered from postpartum preeclampsia and the resulting hypertension caused the stroke. He further stated that aggressive antihypertensive therapy and magnesium sulfate could have prevented the stroke.
This expert was critical of the defendant for failing to obtain information from Ob-gyn C about the patient’s reflexes, neurological status, and whether she had epigastric or right upperq uadrant pain. This information, he argued, would have alerted the defendant to the seriousness of the patient’s condition.
Defense experts argued that the patient came to the hospital with an evolving stroke that caused the hypertension. According to one expert, not all hemorrhagic strokes associated with elevated blood pressures several days after delivery are due to preeclampsia. The various symptoms the patient exhibited could not be explained as preeclampsia, but could be explained by the diagnosis of evolving stroke.
Further, serious strokes such as this one are rarely seen in the postpartum period and “almost for certain when those events happen, they are under circumstances where the stroke occurred suddenly and could not have been predicted or prevented.”
One defense ob-gyn expert stated that — based on Ob-gyn C’s history and physical — there were no signs and symptoms of neurological abnormality, impending stroke, or preeclampsia. (Ob-gyn C was not told of the patient’s blood pressure reading of 188/121 mm Hg documented at 12:39 p.m., 30 minutes after the exam.) In a patient with no history of preeclampsia five days after delivery, postpartum preeclampsia would be low on the differential diagnosis.
In retrospect, even if the blood pressures were interpreted as suspicious for preeclampsia, blood pressures alone would still not have led one to believe the patient was in a critical condition. Even with a suspected diagnosis of postpartum preeclampsia, transfer to the floor in a non-stat manner would have been appropriate, as would observation, magnesium sulfate, and mild antihypertensive therapy.
It was also pointed out that, because the aggressive antihypertensive management prescribed by Ob-gyn D did not prevent further hemorrhage, this treatment may not have prevented the initial stroke/hemorrhage from occurring when the patient first came to the ED.
An obstacle in the defense of this case was the discrepancy between Ob-gyn B and Ob-gyn C regarding their telephone conversation. Ob-gyn B did not recall what Ob-gyn C told him during the phone call. However, he testified he is certain that she did not tell him of the blood pressure of 172/89 mm Hg. Ob-gyn C was adamant that she told Ob-gyn B everything in her note, including the blood pressure reading.
This case was taken to trial and the jury returned a verdict in favor of all the defendants except the hospital.
Risk management considerations
In this case, communication among those caring for the patient was less than ideal. Important information about the patient’s blood pressure was not relayed by nurses to Ob-gyn C. Would the outcome have been different if the nurses had informed Ob-gyn C of the patient’s blood pressure reading of 188/121 mm Hg recorded at 12:39 p.m.? A disconnect also occurred between Ob-gyn B and Ob-gyn C regarding the patient’s blood pressure.
Events occurred that were not documented in a timely manner or not at all. A triage assessment was not recorded in the medical record so there was no way to confirm when the patient began reporting a headache. Additionally, medications and tests were ordered in the ED and there was no way to know who ordered them. Failure to document this information made it difficult to verify the actions taken by ED personnel.
Lack of documentation also made it difficult to determine what was said during the conversation between Ob-gyn C and Ob-gyn B. When physicians take calls from their colleagues at hospitals, the office-based physicians do not always document the conversation. The reason for this is often that the office-based physician believes that the hospital physician is documenting the conversation. This documentation is adequate except when a discrepancy about what was said occurs. Office-based physicians can document conversations about patients in their office chart or, if done timely, in the hospital chart