changes of pregnancy and the challenges associated with the care of 2 patients (mother and unborn baby) require the coordination

Transcription

1 Research EDUCATION Deficits in the provision of cardiopulmonary resuscitation during simulated obstetric crises Steven S. Lipman, MD; Kay I. Daniels, MD; Brendan Carvalho, MBBCh, FRCA; Julie Arafeh, RN, MSN; Kimberly Harney, MD; Andrea Puck, RN, MSN; Sheila E. Cohen, MBChB, FRCA; Maurice Druzin, MD OBJECTIVE: Previous work suggests the potential for suboptimal cardiopulmonary resuscitation (CPR) in the parturient but did not directly assess actual performance. STUDY DESIGN: We evaluated 18 videotaped simulations of maternal amniotic fluid embolus and resultant cardiac arrest. A checklist containing 10 current American Heart Association recommendations for advanced cardiac life support (ACLS) in obstetric patients was utilized. We evaluated which tasks were completed correctly and the time required to perform key actions. RESULTS: Proper compressions were delivered by our teams 56% of the time and ventilations 50% of the time. Critical interventions such as left uterine displacement and placing a firm back support prior to compressions were frequently neglected (in 44% and 22% of cases, respectively). The mean SD overall composite score for the tasks was 45 12% (range, 20 60%). The neonatal team was called in a median (interquartile range) of 1:42 (0:44-2:18) minutes:seconds; 15 of 18 (83%) teams called only after the patient was completely unresponsive. Fifty percent of teams did not provide basic information to the neonatal teams as required by neonatal resuscitation provider guidelines. CONCLUSION: Multiple deficits were noted in the provision of CPR to parturients during simulated arrests, despite current ACLS certification for all participants. Current requirements for ACLS certification and training for obstetric staff may require revision. Key words: advanced cardiac life support, amniotic fluid embolus, obstetric crises, simulation Cite this article as: Lipman SS, Daniels KI, Carvalho B, et al. Deficits in the provision of cardiopulmonary resuscitation during simulated obstetric crises. Am J Obstet Gynecol 2010;203:179.e1-5. Maternal mortality is likely underreported and continues to occur at an unacceptably high rate. 1,2 In maternal deaths involving acute cardiac arrest, advanced cardiac life support (ACLS) must be rapidly administered. 1,3,4 Recent studies at our institution and elsewhere demonstrate inadequate theoretic knowledge of cardiopulmonary resuscitation (CPR) in pregnant women. 5,6 However, there is sparse literature on the quality of CPR in actual or simulated obstetric cardiac arrest scenarios. Obstetric resuscitation is challenging when compared with resuscitation of nonpregnant adults. 7 The anatomic and physiologic changes of pregnancy and the challenges associated with the care of 2 patients (mother and unborn baby) require the coordination of multiple teams and aggressive interventions (ie, perimortem cesarean section for both maternal and fetal benefit if an instrumented vaginal delivery is not possible). 7,8 For obstetric cardiac arrests, there are additional tasks to perform, and more personnel are required to perform them than in a nonpregnant adult resuscitation. 7 The aim of this study was to evaluate the quality of obstetric ACLS performed during the management of a simulated cardiac arrest in a term gravid patient. It From the Departments of Anesthesiology (Drs Lipman, Carvalho, and Cohen) and Obstetrics and Gynecology (Drs Daniels, Harney, and Druzin), Stanford University School of Medicine, Stanford, CA; the Center for Advanced Pediatric and Perinatal Education (Ms Arafeh); and the Labor and Delivery Ward, Lucile Packard Children s Hospital, Palo Alto (Ms Puck). Presented at the 40th Annual Meeting of the Society for Obstetric Anesthesia and Perinatology, Chicago, IL, April 30-May 4, Received July 23, 2009; revised Nov. 23, 2009; accepted Feb. 10, Reprints: Steven Lipman, MD, Department of Anesthesiology, MC5640, 300 Pasteur Dr., Stanford, CA steve.lipman@stanford.edu. This study was supported in part by the Association of Professors of Gynecology and Obstetrics /$ Mosby, Inc. All rights reserved. doi: /j.ajog is unlikely that a randomized, doubleblinded, prospective trial evaluating the quality of obstetric resuscitation will ever be published because the logistic and ethical issues preclude it. The study prospectively reviewed and analyzed preexisting videotapes of highfidelity, simulated maternal cardiac arrests. The simulated arrests created a setting marked by extreme time pressure but no potential for patient harm. MATERIALS AND METHODS ObSim is the name given to the labor and delivery team training simulation program developed at the Center for Advanced Pediatric and Perinatal Education (CAPE) at Lucile Packard Children s Hospital (LPCH) (Stanford, CA). The use of ObSim training course videotapes does not meet the definition of human research as defined by the Stanford University Institutional Review Board. CAPE is located in a separate building across the street from LPCH. Eighteen ObSim courses conducted there over a 25 month period, from August 2005 to AUGUST 2010 American Journal of Obstetrics & Gynecology 179.e1

2 Research Education TABLE The proportion of teams performing interventions ACLS or pregnancy-appropriate interventions Teams performing task (%) Removes fetal/uterine monitors before 0/18 (0) Defibrillation 1/18 (6) Places firm support under back 4/18 (22) Switches compressor every 2 minutes 6/18 (33) Left uterine displacement 8/18 (44) Correct ventilation rate 9/18 (50) Correct compression rate 10/18 (56) Report given to neonatal team 14/18 (78) Cricoid pressure utilized 14/18 (78) Correct hand position on the sternum 15/18 (83) Delivery of fetus in 5 minutes 15/18 (83) ACLS, advanced cardiac life support. Lipman. Deficits in cardiopulmonary resuscitation in simulated obstetrics. Am J Obstet Gynecol September 2007, served as the basis of our investigation. CAPE contains a high-fidelity labor and delivery suite with working programmable fetal monitors that produce real-time printouts and a continuous audible signal of fetal heart rate (FetalSim Advanced Medical Simulations, Inc, Binghamton, NY), a pelvic model (Simulaids Inc, Saugerties, NY), a patient simulator (SimMan; Laerdal Medical, Wappinger Falls, NY), and digital videotape recording. CAPE also contains video and audio playback capability, and after each scenario a debriefing session took place. During the debriefing (facilitated by trained faculty), the whole team both reviewed and discussed various portions of the scenario. The study population consisted of teams of labor and delivery nurses, anesthesiologists, and obstetricians. The labor and delivery nurses had experience in obstetric care ranging from 1 to 10 years. The obstetric residents, anesthesiology residents and fellows, and obstetric nurses came from 2 major teaching programs. Obstetric residents were always paired with nurses from their own institutions. Anesthesiologists delivered care at both hospitals and were placed randomly with teams from both institutions. Each multidisciplinary team studied was comprised of 1-2 obstetricians, 1-2 labor and delivery nurses (1 designated as primary), and 1 anesthesiologist. Neonatologists and code providers were available to respond if requested by the scenario participants. Actions of neonatologists and other requested providers were not evaluated as part of this study. Care delivered and/or tasks performed by these responders were as directed by the study participants as described in previous text. Participants were not prepared for ObSim with formal lectures prior to the course, nor were they provided with any didactic material. On the day of the course, they were oriented to the simulation room, manikin, and equipment but remained naïve to the scenario. The simulated scenario was an amniotic fluid embolus (AFE) in a term parturient with a singleton intrauterine pregnancy. The patient had a working lumbar epidural catheter in situ for labor analgesia and an 18-gauge intravenous line in the left antecubital fossa. The fetal vertex was placed in position to allow for an outlet operative vaginal delivery. Required interventions in the scenario included the provision of ACLS after maternal cardiopulmonary collapse from an AFE and operative vaginal delivery of the fetus. The simulation room contained fully stocked epidural and code carts with expired medications and intravenous fluids, catheters for intravenous access, blood pump tubing, central line kits, arterial line kits, and a live Zoll M Series Biphasic 200J Max defibrillator (Zoll Medical Corp, Chelmsford, MA). This defibrillator model is currently used on our labor and delivery ward. A simple checklist of 10 basic interventions (Table) was developed by 4 of the authors, including a board-certified maternal fetal medicine specialist (M.D.), a general obstetrician (K.I.D.), a clinical nurse specialist in high-risk obstetrics (J.A.), and a fellowship-trained obstetric anesthesiologist (S.S.L.). The checklist was based on expert recommendations for ACLS and obstetric resuscitation from the following 2 definitive consensus publications: (1) American Heart Association (AHA), part 7.2 (Management of Cardiac Arrest 9 ) and (2) AHA, part 10.8 (Cardiac Arrest Associated With Pregnancy 7 ). Neonatal resuscitation provider (NRP) guidelines were based on AHA and the American Academy of Pediatrics NRP guidelines. 10 The 4 aforementioned individuals reviewed the videotapes using the checklist and determined whether a recommended task was completed. A separate table was used to measure the time required to complete 4 critical actions (delivery, endotracheal intubation, Code Blue activation, and call for a neonatal team). There was complete agreement between the 4 reviewers with respect to whether tasks were completed and the time required to perform them. Data are presented as mean SD, median times (interquartile range), and number of teams (percentage) as appropriate. Time zero for the 4 interventions was taken from the start of a nonperfusing rhythm. Descriptive data analysis was performed using Microsoft Excel (Richmond, CA) and SPSS version 11 (SPSS, Inc, Chicago, IL). Data were assessed for normal distribution of variance using QQ plots and Kolmogorov-Smirov tests. RESULTS A total of 18 teams including 69 individuals (31 labor and delivery nurses; 15 anesthesiology residents, fellows, or attendings; 23 obstetric postgraduate year 179.e2 American Journal of Obstetrics & Gynecology AUGUST 2010

3 Education Research 2, 3, and 4 residents, fellows, or attendings) participated in the simulated scenarios. The Figure summarizes the experience of the participants. The Table summarizes the proportion of all teams performing the basic ACLS interventions on the checklist. No team (0/18) performed all of the 10 specified tasks adequately. The mean overall composite score SD with respect to performance of the 10 AHA recommended tasks was 45 12% (range, 20 60%). The neonatal team was called in a median (interquartile range [IQR]) of 1:42 (0:44 2:18) minutes:seconds; 15 of 18 teams (83%) called only after the patient was completely unresponsive. Nine of 18 teams (50%) did not provide basic information required by the NRP guidelines to the neonatal response teams. Code Blue activation took a median (IQR) of 2:01 (0:57 3:09) minutes:seconds; 3 of 18 teams (17%) did not call a Code Blue. Intubation of the parturient took a median (IQR) of 2:30 (2:07 5:02) minutes:seconds; 2 of 18 teams (11%) did not intubate, and 2 additional teams did not auscultate for breath sounds after intubation. Delivery of the fetus took a median (IQR) of 2:51 (2:23 3:55) minutes: seconds; 3 of 18 teams (17%) did not deliver the fetus within 5 minutes. FIGURE Demographics of team members A, Physician s level of training (obstetric and anesthesia residents, fellows, and attendings). Values are presented as percentage distribution. B, Labor and delivery nurse years of practice on labor and delivery. Values are presented as percentage distribution. PGY, postgraduate year. Lipman. Deficits in cardiopulmonary resuscitation in simulated obstetrics. Am J Obstet Gynecol COMMENT This study demonstrated striking deficiencies in the performance of key ACLS tasks critical to the resuscitation of pregnant women during a maternal cardiac arrest. This performance/behavioral study during simulated maternal cardiac arrests confirms recent reports that demonstrated inadequate theoretic/cognitive knowledge (as assessed by a written multiple-choice test) of cardiopulmonary resuscitation in pregnant women. 5,6 Of the 10 recommended interventions evaluated, the ability to correctly deliver chest compressions is among the most critical. 9,11,12 Concerns with regard to the poor quality and number of interruptions in chest compressions resulted in major changes in the 2005 AHA ACLS guidelines. 11 Correctly performed compressions are predicated on proper rate and depth and utilization of a backboard. In pregnant women, left uterine displacement (LUD) and correct hand position (midsternal) are necessary. Based on these 4 criteria, it is notable that 0 of 18 teams (0%) properly administered this most basic intervention. LUD minimizes aortocaval compression by the gravid uterus, improves venous return, and facilitates cardiac output. LUD is essential when gestation is greater than 20 weeks. However, two thirds of our ACLS-certified teams neglected to perform LUD. Notably, LUD was demonstrated on the manikin to all teams during room orientation. The majority of teams (83%) were able to deliver the fetus in 5 minutes or less. However, the scenario was written to allow the obstetrical provider to opt for an operative vaginal delivery. Operative vaginal deliveries are often performed at the bedside, allow the obstetrician to avoid an abdominal procedure, and may be less detrimental to the ongoing provision of effective chest compressions than a perimortem cesarean section (CS). Yet maternal arrests may occur when operative vaginal delivery is not possible. Had a less favorable station and cervical dilation been present in our manikin, a perimortem CS would have been required. Simultaneous performance of a perimortem CS during CPR may degrade the quality of ongoing compressions or require temporary cessation of compressions. Many of our teams indicated they would have moved to the operating room (OR) for CS. Had such transport been initiated, it seems unlikely delivery of the neonate would have occurred within 5 minutes. In addition, expertly performed compressions are thought to generate a cardiac output of only ap- AUGUST 2010 American Journal of Obstetrics & Gynecology 179.e3

4 Research Education proximately 10% of normal in parturients (vs approximately 30% of normal in nonpregnant patients). 2,13-15 It is therefore questionable whether effective maternal CPR could even be administered while en route to the OR. In almost all situations of maternal cardiac arrest, delivery of the neonate should commence when no improvement in maternal condition has occurred by 4 minutes after the arrest. 2,7,14,15 Delivery improves maternal venous return and thus cardiac output, reduces additional metabolic demand (increased oxygen consumption and CO 2 production) caused by the fetus, and may increase the chance for both intact maternal and neonatal survival without neurologic sequelae. Numerous reports have described successful maternal resuscitation in pregnancy only after rapid delivery was performed. Delivery of any fetus with gestational age greater than 20 weeks in 5 minutes or less is now considered among the most critical goals in cardiac arrest during pregnancy. 2,7,9,14 All participants in this study were certified and current in ACLS. However, it is possible that such certification conferred limited advantage. In a recent study at our institution, Cohen et al 5 found individuals who had never previously attended life support courses to have equivalent knowledge to those who had. Of the 75 subjects in that study, only 15% achieved a passing score of greater than 85% (the passing score used in national AHA ACLS courses). 5 These findings agree with other studies suggesting that substantial skill loss occurs over time. 10,16 Labor and delivery ward staff do not often practice ACLS skills because cardiac arrest is rare in the obstetric setting. Because recertification is currently required only every 2 years, ACLS skills are not often refreshed. Moreover, most ACLS courses do not discuss issues specific to cardiac arrest in pregnant women. Nine of 18 teams neglected to provide basic NRP recommended information, or even the specific information that the mother had arrested, to the responding neonatal teams in our scenarios. 17 Although neonatal resuscitation providers may require several minutes to arrive, 83% of our teams waited nearly 2 minutes after the maternal arrest to call for the neonatal team. In the context of a severely depressed neonate, the resultant delay could have deleterious consequences. Although we attempted to simulate real-life ACLS resuscitation, simulated scenarios have a number of limitations. The layout of the simulated labor and delivery room was slightly different from the actual rooms in which our teams render care, and it is possible that the manikin did not provide the same cues or engender the same responses as an actual patient. Whether our teams would have performed better in their native environment with real patients is unclear. Financial and logistic challenges precluded studying larger numbers of subjects, and comparisons of performances among teams and level of training were not possible. However, all participants routinely provided high-risk obstetric care, and the study teams composition was very similar to teams responding to actual maternal arrests in our institutions. We therefore believe our findings are highly relevant to our 2 programs (combined delivery rate of 12,000/year) and may be relevant to other tertiary care academic centers and community hospitals. There are other limitations to our study. Although a team of experienced physicians developed the checklist, it did not undergo validation testing. However, the checklist was derived from current, published AHA guidelines on this subject. We believe that it addresses areas critical to the proper performance of ACLS in the parturient. Lastly, our teams knew the purpose of the ObSim course was to practice obstetric crises. We extensively reviewed the operation of the Zoll defibrillator during orientation, and it is likely that participants anticipated the need to perform CPR. Our teams therefore had certain advantages relative to individual providers or teams in actual events: (1) they expected something bad to happen, and hence, their level of vigilance and reaction to slight physiologic aberrations was likely enhanced; (2) the cervix was completely dilated with the fetal vertex at the perineum, so an operative vaginal operative delivery was possible; (3) the courses were conducted during normal working hours between 8 and 11:30 AM, so decrements in human performance secondary to diurnal variation and fatigue were presumably not factors; 18 (4) our participants had no other patients to care for, were on-site, and were immediately available when called into the simulation; (5) all of our participants were ACLS certified and worked on busy labor and delivery units in large teaching hospitals that serve as referral centers for high-acuity patients; (6) some of the physician participants were fellow or attending level providers; (7) our participants received a hands-on in-service of the Zoll defibrillator during orientation, so they were familiar with the operation of the device, and (8) some of the anesthesiologists and all of the nurses were volunteers. Because of self-selection bias, volunteers may be more confident or skilled than the general population. Despite these multiple advantages, our study teams made multiple errors in the performance of basic physical and cognitive tasks. Yet in actual maternal cardiac arrests occurring in the field, emergency rooms, or small hospitals with few or no in-house obstetricians or anesthesiologists, team performance is likely to be worse. Our findings support previous studies that assessed fund of knowledge only. 5,6 Such data suggest biennial ACLS recertification alone is not adequate for obstetric staff. In the top 10 recommendations from the report, Saving Mothers Lives , The Confidential Enquires into Maternal and Child Health (CEMACH) specifically recommends the improvement of basic, immediate, and advanced life support skills on an annual basis. 3,19 However, whether annual (or even more frequent recertification) would result in better performance is untested. Possible measures to improve performance of maternal CPR include the following: (1) creation of immersive learning environments with simulation training; (2) frequent clinical multidisciplinary obstetric team drills as recommended by the Joint Commission and 179.e4 American Journal of Obstetrics & Gynecology AUGUST 2010

5 Education Research CEMACH; 3,19,20 (3) more frequent recertification and/or the development of an AHA obstetric life support (ObLS) course certification; and (4) inclusion of didactic modules specific to resuscitation of pregnant women in ACLS courses. f ACKNOWLEDGMENTS The development of AFE/cardiac arrest scenario was done by K.I.D., J.A., S.S.L., M.D., and K.H. The faculty and staff for ObSim courses at CAPE were K.I.D., J.A., S.S.L., M.D., K.H., and A.P. The study conception and design was by J.A. and M.D. The acquisition of data was done by K.I.D., J.A., M.D., and S.S.L. Analysis and interpretation of data was done by B.C. and S.S.L. Draft and revisions of manuscript was done by S.S.L. Manuscript edits, tables, and statistics were done by B.C., K.I.D., S.E.C., and M.D. Critical analysis and major revisions of manuscript were done by S.S.L. and S.E.C. REFERENCES 1. World Health Organization. Available at: maternal_mortality_2005/. Accessed March 26, Katz V, Balderston K, DeFreest M. Perimortem cesarean delivery: were our assumptions correct? Am J Obstet Gynecol 2005;192: ; discussion CEMACH. Confidential Enquiries Into Maternal and Child Health (CEMACH), Saving Mothers Lives London, UK: RCOG Press. 4. Banks A. Maternal resuscitation: plenty of room for improvement. Int J Obstet Anesth 2008;17: Cohen SE, Andes LC, Carvalho B. Assessment of knowledge regarding cardiopulmonary resuscitation of pregnant women. Int J Obstet Anesth 2008;17: Einav S, Matot I, Berkenstadt H, Bromiker R, Weiniger CF. A survey of labour ward clinicians knowledge of maternal cardiac arrest and resuscitation. Int J Obstet Anesth 2008;17: American Heart Association. Part 10.8: cardiac arrest associated with pregnancy. Circulation 2005;112: Yeomans ER, Gilstrap LC. Physiologic changes in pregnancy and their impact on critical care. Crit Care Med 2005;33(10 Suppl): S American Heart Association. Part 7.2: management of cardiac arrest. Circulation 2005; 112: Choudhry NK, Fletcher RH, Soumerai SB. Systematic review: the relationship between clinical experience and quality of health care. Ann Intern Med 2005;142: Hazinski MF, Nadkarni VM, Hickey RW, O Connor R, Becker LB, Zaritsky A. Major changes in the 2005 AHA Guidelines for CPR and ECC: reaching the tipping point for change. Circulation 2005;112(24 Suppl):IV-206-IV Ristagno G, Tang W, Chang YT, et al. The quality of chest compressions during cardiopulmonary resuscitation overrides importance of timing of defibrillation. Chest 2007;132: Sanders AB, Meislin HW, Ewy GA. The physiology of cardiopulmonary resuscitation. An update. JAMA 1984;252: Katz VL, Dotters DJ, Droegemueller W. Perimortem cesarean delivery. Obstet Gynecol 1986;68: Mallampalli A, Guy E. Cardiac arrest in pregnancy and somatic support after brain death. Crit Care Med 2005;33(10 Suppl):S Arthur W Jr, Bennett W Jr, Stanush PL, Mc- Nelly TL. Factors that influence skill decay and retention: a quantitative review and analysis. Hum Performance 1998;11: American Heart Association, American Academy of Pediatrics. Neonatal resuscitation guidelines. Pediatrics 2006;117:e Cao CG, Weinger MB, Slagle J, et al., Differences in day and night shift clinical performance in anesthesiology. Hum Factors 2008; 50: Royal College of Obstetrics and Gynaecology. Standards for maternity care. Report of a working party. London: RCOG Press; Joint Commission for Accreditation of Healthcare Organizations (Joint Commission). Sentinel event alert #30: preventing infant death and injury during delivery Available at: SentinelEventAlert/sea_30.htm. Accessed March 26, AUGUST 2010 American Journal of Obstetrics & Gynecology 179.e5