Out-of-hospital cardiac arrest is a leading cause of death in the United States. Pregnant women are not immune to cardiac arrest, and the treatment of such patients can be difficult. Pregnancy is a relative contraindication to the use of therapeutic hypothermia after cardiac arrest.
A 20-year-old woman who was 18 weeks pregnant had an out-of-hospital cardiac arrest. Upon her arrival at the emergency department, she was resuscitated and her circulation returned spontaneously, but her score on the Glasgow Coma Scale was 3. After adequate family discussion of the risks and benefits of therapeutic hypothermia, a decision was made to initiate therapeutic hypothermia per established protocol for 24 hours. The patient was successfully cooled and rewarmed. By the time she was discharged, she had experienced complete neurologic recovery, apart from some short-term memory loss. Subsequently, at 40 weeks, she delivered vaginally a 7-lb 3-oz girl whose Apgar scores were 8 and 9, at 1 and 5 minutes respectively.
To our knowledge, this is only the 3rd reported case of a successful outcome following the initiation of therapeutic hypothermia for out-of-hospital cardiac arrest in a pregnant woman. On the basis of this and previous reports of successful outcomes, we recommend that therapeutic hypothermia be considered an option in the management of out-of-hospital cardiac arrest in the pregnant population. To facilitate a successful outcome, a multidisciplinary approach involving cardiology, emergency medicine, obstetrics, and neurology should be used.
Out-of-hospital cardiac arrest (OHCA) is a leading cause of death in the United States. Approximately 424,000 such deaths—as evaluated by emergency medical services (EMS)—occur annually.1 Survival to hospital admission or discharge is dismal: less than 30% of patients survived to hospital admission and less than 10% survived to discharge, according to data from the Cardiac Arrest Registry (CARE).2 Survival to discharge is driven mainly by neurologic injury, which accounted for two thirds of OHCA deaths in one study.3 Two studies from 20024,5 reported significant improvements in the rates of neurologic recovery and subsequent survival following the induction of mild hypothermia in patients who remained comatose upon resuscitation. However, a more recent and larger study suggests that targeted temperature management, rather than the induction of hypothermia, is the fundamental element for neurologic recovery.6
As of 2013, there were 2 reported cases of pregnant patients who underwent therapeutic hypothermia (TH) with a favorable outcome for both mother and fetus.7,8 A 3rd case study reported mixed results, with fetal demise balanced by the survival of the mother after prolonged hospitalization.9
We report the use of TH in a pregnant patient, with favorable recovery for both mother and fetus. Our case, along with the other published cases, highlights the fact that pregnancy should not be considered an absolute contraindication for TH.
A 20-year-old previously healthy black woman, 18 weeks pregnant, experienced OHCA while running on a treadmill at a gymnasium. She was down for approximately 5 minutes before cardiopulmonary resuscitation (CPR) was begun upon EMS arrival. The initial rhythm check revealed that the patient was in ventricular fibrillation cardiac arrest. She promptly received 2 defibrillator shocks and experienced return of spontaneous circulation (ROSC) after approximately 8 minutes of resuscitation. Upon arrival at the nearest emergency room, the patient received a low dose of epinephrine for mild hypotension and was intubated. Before intubation, she displayed minimal spontaneous movement and was subsequently transferred to our facility for a higher level of care.
Approximately 100 minutes after the event, the patient arrived at our emergency department, intubated, with vital signs as follows: blood pressure, 143/90 mmHg; pulse rate, 110 beats/min; and temperature, 36.1 °C. A 12-lead electrocardiogram showed sinus tachycardia with incomplete right bundle branch block. A bedside echocardiogram revealed normal global systolic function, with an estimated left ventricular ejection fraction (LVEF) of 0.65. The patient was emergently evaluated by our obstetric service via bedside ultrasonography, which confirmed a fetus of approximately 17 to 18 weeks' gestational age. No fetal abnormalities were identified. Initial laboratory results included a troponin I level of 0.21 (>0.10 suggests cardiac injury); creatine kinase, 311 U/L; and lactic acid, 2.8 mg/dL. Our patient's drug screen was negative, with normal thyroid-stimulating hormone and electrolyte levels. A noncontrast computed tomogram of the head and cervical spine revealed no acute pathologic condition.
We discussed the risks and benefits of TH at length with the patient's family, and they agreed to its initiation. The TH protocol (Fig. 1) was initiated approximately 180 minutes after cardiac arrest. She was given midazolam and fentanyl for sedation and analgesia, respectively. To reduce her risk of shivering, we administered the skeletal muscle relaxant cisatracurium besylate. No vasopressor support was needed, and her hypothermic phase was continued for 24 hours with no major event.
She was rewarmed over the next 24 hours, and the neuromuscular blockade was discontinued after titration. Upon reaching normothermia, the patient became tachycardic and periodically made some hand gestures toward the endotracheal tube, despite being maintained on low doses of midazolam and fentanyl. Approximately 72 hours after the event, the patient was extubated, after which she was able to answer questions, follow commands, and interpret abstract phrases; she exhibited some short-term memory loss. A follow-up magnetic resonance image of the brain displayed no abnormalities.
On day 4, the patient was stable and showed continued improvement in cognitive function; a cardiac computed tomogram was not indicative of coronary artery disease. The electrophysiology team was consulted and a single-chamber permanent implantable cardioverter-defibrillator (ICD) was placed. After successful interrogation of the ICD, the patient was discharged from the hospital.
She continued outpatient follow-up with electrophysiology and high-risk maternal fetal medicine specialists. During a subsequent hospitalization for one episode of chest pain, interrogation of the patient's ICD revealed multiple episodes of supraventricular tachycardia, occurring in the ventricular tachycardia and ventricular fibrillation zones. She was offered atrioventricular nodal blocking agents, which she refused because of the risks of birth defects. To evaluate fetal growth and anatomy, we monitored her via serial ultrasonography. The remainder of her pregnancy was rather uneventful. The fetus grew at a normal rate, and no abnormalities were seen. At 40 weeks, she delivered vaginally a 7-lb, 3-oz girl whose Apgar scores were 8 and 9, at 1 and 5 minutes respectively. The patient has since delivered another child with no complications.
Therapeutic hypothermia has shown effectiveness in aiding both neurologic recovery and (ultimately) mortality rates after cardiac arrest.4,5 When cardiac arrest occurs in a pregnant woman, it is a challenge for clinicians who are unaccustomed to treating such patients. This challenge is compounded if the patient would have been a candidate for TH after ROSC, had she not been pregnant.
Society for Obstetric Anesthesia and Perinatology. A draft consensus statement by the Society for Obstetric Anesthesia and Perinatology in 2013 recommends that TH be strongly considered after cardiac arrest in obstetric or postpartum patients whose indications are similar to those of the general population. However, the statement acknowledges that the available data are inadequate and recommends continuous fetal monitoring in response to reported fetal bradycardia during TH.10
American Heart Association. Guidelines published by the American Heart Association (AHA) recommend TH for comatose patients with OHCA, provided that they have initial shockable rhythm.11 The qualification criteria for TH are derived mainly from the inclusion and exclusion criteria of clinical trials that have shown benefit.11 A key exclusion criterion in the studies is pregnancy, and that exclusion is reflected in the guidelines.4,5,11
Pregnancy as a contraindication is standard in most clinical studies, because special provisions must be made to ensure the safety of both the mother and the fetus. Pregnant women are not immune to cardiac arrest: the reported rate is 1 in every 20,000 pregnancies, and the survival rate is similarly poor when compared with that of the general population.12 The lack of clinical trial evidence and real-world experience makes the treatment of a pregnant woman suffering from cardiac arrest quite complex. The current AHA guidelines (contrary to the AHA's exclusion criterion mentioned above11 ) recommend considering TH on an individual basis and applying to some pregnant patients the criteria for non-pregnant patients.12
Cases Reported to Date
In 2008, Rittenberger and colleagues7 published the first case of a pregnant patient who was successfully treated with use of TH, after a witnessed OHCA. This 35-year-old woman, 13 weeks pregnant, experienced approximately 21 minutes of greatly diminished or no flow, before ROSC and after several rounds of CPR and 5 defibrillator shocks. She was rapidly cooled via cold saline infusion and cutaneous ice packs. Sedation and shivering cessation were achieved with the use of propofol and pancuronium. The patient was slowly rewarmed and awoke on hospital day 2 with a mild neurologic deficit. Her drug screen was negative; she manifested mild global hypokinesia on emergency bedside echocardiography. The infant was delivered via caesarian section at 39 weeks' gestation, and the baby's neurodevelopmental testing was appropriate both at birth and at follow-up 2 months later.7
Wible and associates9 reported another case: a 20-week-pregnant, 44-year-old patient who underwent TH for an OHCA that necessitated multiple rounds of CPR, including 10 rounds of defibrillator shocks. The time to ROSC was not documented, but a core temperature of 33 °C was achieved approximately 6 hours after cardiac arrest. Approximately 28 hours after the OHCA, a stillborn fetus was delivered spontaneously. It is worth noting that on admission the patient had a positive urine drug screen for cocaine and an estimated LVEF of less than 0.10. After a prolonged hospital course, the patient was discharged, alert and oriented to person, place, and time but with no recollection of the events leading to the OHCA. At her one-year follow-up, the patient was still able to perform her activities of daily living. Given her poor cardiac function, positive drug screen, and prolonged resuscitation, it is difficult to evaluate the role that TH played in the death of this patient's fetus.9
In 2012, Chauhan and co-authors8 reported another case of successful TH—in a 33-year-old pregnant patient who had a witnessed OHCA at 20 weeks' gestation, followed by ROSC after 25 minutes. Sedation and paralysis were achieved with use of propofol, fentanyl, and vencuronium at standard doses. The patient was discharged on day 10 in stable condition and subsequently underwent an uncomplicated vaginal delivery at term. Follow-up data through 3 years after cardiac arrest showed the infant's reaching all normal neurodevelopmental milestones.8
Maternal cardiac arrest leads to cessation of uteroplacental perfusion; however, the extent of the fetal damage varies with the duration of absent perfusion. In our patient, the duration of absent or decreased uteroplacental perfusion was 5 to 8 minutes, yet no appreciable fetal damage was noted. This observation is consistent with a case series published by Baghirzada and Balki,13 in which neonatal survival was evident up to 14 minutes after maternal arrest. Our patient had a fairly rapid return of spontaneous circulation, so our experience might not be helpful in treating patients with a prolonged course of ROSC; however, we must remember that, in the case reported by Chauhan and colleagues,8 ROSC was achieved after 25 minutes.
It is worth noting that the cases reported to date have involved patients who were less than 21 weeks pregnant, as can be seen in Table I.7–9 The impact of TH in patients who are more than 20 weeks' pregnant or whose ROSC is substantially delayed is not known. Current guidelines should be followed in the treatment of such patients, including the consideration of emergency caesarean section.10,12 Fetal bradycardia is a potential adverse effect related to TH and should be continuously monitored in pregnant patients who are undergoing hypothermia.
Medication use in pregnancy is always of concern, given the risk of fetal impairment. However, the medication requirements during TH are not extensive, and are limited primarily to sedatives and analgesics. There is no preferred sedative agent recommended for use in pregnancy. Agents commonly used for prolonged sedation—such as midazolam, propofol, lorazepam, and dexmedetomidine—fall into U.S. Food and Drug Administration pregnancy risk categories D, B, D, and C, respectively.14 Although propofol might appear to be the preferred agent on the basis of its risk categorization, there are no controlled human studies of propofol, and its rating arose primarily from animal data.14 Our patient received midazolam, which has not manifested teratogenicity in animal studies, but has been linked to respiratory depression in neonates who were exposed primarily during parturition.14 There are no data on the long-term effects of midazolam on fetal growth. Similarly, fentanyl (an analgesic commonly used in intensive care) is classified as pregnancy risk category C, with most use reported around labor or obstetric delivery.15
Despite the limitation imposed by the overall small number of reported cases, our case and the others reported to date provide some reassurance that induced hypothermia is not directly harmful to the fetus and that pregnancy should not be considered an absolute contraindication to the use of hypothermia in post-cardiac arrest care. A comprehensive, team-based approach involving cardiologists, emergency physicians, obstetricians, and neurologists should be used in evaluating and selecting pregnant patients who might benefit from TH.
From: Division of Cardiology (Dr. Oguayo), Baylor Heart and Vascular Institute, Baylor University Medical Center at Dallas, Dallas, Texas; Division of Internal Medicine (Dr. Oguayo) and Obstetrics and Department of Gynecology (Dr. Jones), Pharmacy Department (Dr. Oyetayo), and Division of Cardiology (Drs. Costa and Stewart), Baylor Scott & White Health, Temple, Texas 76508; and Colleges of Medicine (Drs. Costa, Jones, Oguayo, and Stewart) and Pharmacy (Dr. Oyetayo), Texas A&M University Health Science Center; Temple, Texas 76508