COSMIC = consultation, surveillance, monitoring, and intensive care; EMR = electronic medical record; ICU = intensive care unit; QA = quality assurance
Multidisciplinary care for high-risk obstetric patients represents a challenge for the rational allocation of health care resources. Although the majority of our obstetric patients experience good outcomes, a small group of women develop serious morbidity or even die during pregnancy, labor, or delivery. Unfortunately, it is not always possible in advance to discriminate between these populations. Some women are overlooked when clinicians fail to correctly identify medical risk or when known medical risk factors fail to trigger the antepartum anesthetic, subspecialty, or multidisciplinary consultations needed to optimize peripartum care. Women are also overlooked when we fail to recognize early signs of sepsis, hemorrhage, or other deterioration in time to prevent a potentially avoidable bad outcome. On the other hand, other women are overassessed. For example, the great majority of obstetric patients sent to intensive care units (ICUs) for monitoring do not receive high level intervention.1,2 Such expensive and labor-intensive care is frequently not reserved for those with the greatest potential for benefit.1
In this article, we describe the consultation, surveillance, monitoring, and intensive care (COSMIC) program, a novel, 4-tiered approach to obstetric care that directs acute care resources to the patients who need them most. Using a combination of low-tech and high-tech methods, this model aims to (1) determine risk factors for acute obstetric complications; (2) identify and communicate patient deterioration to caregivers at the bedside; and (3) facilitate multidisciplinary team management between nurses, obstetricians, obstetric anesthesiologists, and critical care physicians. The goal of this model is to more effectively direct resources to high-risk obstetric patients in a targeted and timely fashion, while being mindful of costly ICU resources.
The COSMIC program is particularly relevant today due to the surprising observation that maternal mortality is rising in the United States.3 Although the mechanism(s) driving this trend are incompletely understood, possible contributors include rising rates of preexisting maternal disease (eg, morbid obesity and diabetes)4 and an increasing number of women with severe cardiac disease who survive to reach childbearing age. Any approach designed to address maternal mortality must focus on maternal morbidity as the root cause.
COSMIC is an acronym that describes the following components of our model:
- Consultation: A set of maternal diagnostic criteria that trigger referral by the treating obstetrician to an obstetric anesthesiologist, with the activation of multidisciplinary care pathways where needed.
- Surveillance: An automated system to identify maternal deterioration in the general in-hospital obstetric population and trigger team-specific early warning alerts.
- Monitoring: A remote, high-fidelity system using ICU telemedicine to monitor known high-risk or deteriorating obstetric patients, which enables peripartum management in the labor unit with oversight by the remote ICU telemedicine service.
- Intensive Care: A care pathway including ICU-level monitoring and therapy for obstetric patients who require intubation, mechanical ventilation, or aggressive hemodynamic or other intensive therapy.
We describe these components in greater detail later in the article.
The presence of medical risk factors for adverse maternal outcomes should trigger a referral for antenatal anesthesia assessment and multidisciplinary planning where relevant. The failure of appropriate referral for consultation is a regular finding in case reviews after major adverse maternal outcomes.5,6 Examples of such risk factors are listed in the Table. Accordingly, the first tier of the COSMIC program is consultation, during which patients who meet established criteria are referred by their primary physician or obstetrician for antepartum anesthesia assessment with multidisciplinary referral as needed (Table).
This approach should be applied to all patients in the obstetric referral network and could be based on either clinical data (where available) or clinical experience. The alphabetic classification in the Table was designed for ease of implementation and has now been in use in the Hadassah Hebrew University Medical Center in Jerusalem and at Washington University, St Louis, MO, for >3 years.
The goal of antenatal consultation is to develop a plan for intrapartum and postpartum anesthesia management for both vaginal and cesarean delivery and to communicate that goal to both the obstetric team and the patient. Where appropriate, patients are referred for additional consultation and multidisciplinary planning. There are few data to guide the timing of these antepartum consultations. In our practice, patients are typically referred between 28 and 32 weeks of gestation to allow time for additional consultations or follow-up if appropriate. However, patients with severe cardiopulmonary disease may be referred earlier and may occasionally undergo preconception multidisciplinary assessment. Most sizeable obstetric anesthesia services already provide such a service. In smaller hospitals, a general preoperative anesthesia clinic could potentially provide this service.
The second tier of the COSMIC program involves automated surveillance of the general obstetric population in the hospital, as part of an early warning system.
More than 10 years has elapsed since the 2007 report of the Confidential Enquiry into Maternal and Child Health in the United Kingdom, which placed the immediate institution of an early warning system for the early detection of severe illness in pregnancy as a central recommendation.5 That report states:
The early detection of severe illness in mothers remains a challenge to all involved in their care. The relative rarity of such events combined with the normal changes in physiology associated with pregnancy and childbirth compounds the problem. Modified early warning scoring systems have been successfully introduced into other areas of clinical practice and a system which has been modified for obstetric mothers … should be introduced for all obstetric admissions in all clinical settings.5
In 2019, we remain far from that goal.
The primary problem is no longer the timely acquisition of patient data. Modern health care is accompanied by dizzying quantities of real-time electronic clinical and physiological data. However, it is not clear which physiological signals best identify clinical deterioration, and clinically meaningful signals may easily be missed. In the obstetric population, key confounding factors include the altered maternal physiology in normal pregnancy7 and the range of hemodynamic responses to normal childbirth. Accordingly, several early warning systems have been suggested for pregnancy,6,8 each with different physiological variables to surveil and cutoff parameters for escalating care.
Unfortunately, when patients deteriorate, few institutions have well-developed escalation pathways. Even where protocols exist, subspecialty silos, institutional culture, and human factors frequently prevent timely escalation.9 Such delays in communication, referral for consultation, diagnosis, and the initiation of appropriate therapy all contribute to pregnancy-related deaths.10 The National Partnership for Maternal Safety has recommended that an ideal maternal early warning system should bypass human communication barriers and be based instead on monitors that automatically collect vital signs and alert appropriate teams about deterioration in real time.6 We agree and emphasize that the COSMIC surveillance system we describe does not replace human care; but rather, it adds another level of vigilance that alerts the correct human care team to attend the deteriorating patient in real time.
An effective early warning system would provide clinicians with both an individual patient view (Figure 1A) and a consolidated overview of their entire labor and delivery service, as a visual “patient census view” (Figure 1B). Deviation of selected physiological or laboratory data from preprogrammed limits will trigger an alarm that can be scaled depending on severity. Similarly, such a system can automatically trigger a call to a nurse or a physician or generate a “stat” group page. The use of automatic targeted paging removes cognitive bias that may otherwise delay requests for patient evaluation and escalation of care. The system can be further refined by iterative feedback to reduce unnecessary pages; as an example, tachycardia alerts are suspended during maternal expulsive efforts when it is assumed that providers are already present by the bedside.11 Although many early warning interfaces are possible, visual representations of patient data allow obstetricians and anesthesiologists to keep track of large numbers of obstetric patients and condense a myriad of clinical data streams into a more rapidly interpretable format.12
A primiparous patient developed postpartum endometritis presenting as severe hypotension, tachycardia, and fever 2 days postpartum; she was taken emergently to the operating room for dilation and curettage and then transferred to the ICU for hemodynamic and ventilatory support. Retrospective review of the records revealed that during the course of that day, the patient’s vital signs were gradually deviating from locally agreed normal values. Despite this, a physician evaluation was not requested until many hours after the initial deterioration in vital signs. An electronic surveillance system would have bypassed the human biases that delay escalation of care and flagged this patient as one who required a bedside physician evaluation several hours earlier than occurred in practice. For example, at the University of Michigan, the obstetric surveillance uses AlertWatch:OB (AlertWatch, Inc, Ann Arbor, MI)11 with the following parameters that trigger a page: sustained heart rate >130, sustained systolic pressure <85, and sustained heart rate/systolic blood pressure <1.3 with systolic blood pressure <105.
The third tier of COSMIC involves the use of an ICU telemedicine program to provide ICU-level monitoring in the labor suite, with direct oversight by critical care specialists for selected high-risk obstetric patients. This tier represents an intermediate level of care between the surveillance/alerting system described above and formal admission to the ICU.
In many hospitals, the labor unit and the ICU are 2 disconnected silos, and clinicians managing high-risk obstetric patients have to make a binary choice between these options. Both approaches are associated with compromises in care that take either the obstetric or ICU teams out of their comfort zones.
The labor unit is the optimal site for managing the dynamic changes in maternal and fetal condition associated with labor and cesarean delivery. By comparison, an ICU has a greater capacity for invasive monitoring and interventional therapy.13 However, patients referred to an ICU for labor will deliver in an environment remote from obstetricians and obstetric anesthesiologists and typically far from the operating room and the neonatal ICU. In the postnatal period, mothers in the ICU are also far from their babies.
Frequently, the binary decision between labor unit and ICU is differentiated by intrapartum and postpartum management. The 3 most common causes for ICU admission in pregnancy are obstetric hemorrhage, preeclampsia with severe features, and maternal cardiac disease.14 Each condition has consensus bundles15,16 or clinical practice guidelines17 according to which all may be routinely managed in the labor unit until after delivery. After delivery and initial stabilization, patients with these conditions are often admitted to the ICU, but generally only for monitoring, and they rarely require intubation, ventilation, or aggressive vasopressor therapy.1,2
Telemedicine ICU offers an alternative to this binary approach. We envision that the telemedicine ICU component of COSMIC will allow selected high-risk or deteriorating patients to be cared for by a bedside ICU-trained obstetric nurse in the labor unit throughout labor and the postnatal period, with ICU-level monitoring and vasopressors if needed, and with high-definition 2-way video backup from a remote telemedicine ICU team.
The telemedicine ICU will represent an additional level of care for most patients. Many of these patients are not currently referred to the ICU, so the telemedicine ICU will in practice accommodate a significantly wider frame of patients; for most of them, this will represent an additional level of enhanced care.
Of the patients who are currently referred to the ICU, most are referred postpartum.2 For those patients too, the antepartum or intrapartum components of telemedicine ICU also provide enhanced care because there is currently no option for this service during in labor.
In this approach, all patients requiring intubation, ventilation, or aggressive hemodynamic support will still be referred to the conventional ICU. So the only patients for whom the COSMIC program represents a “downgrade” in the level of care are those who are currently referred to the ICU for monitoring only. However, the ICU is a limited and expensive resource that may not be warranted for these purposes.1
A 2015 review of critical disease in pregnancy stressed that whatever approaches are used, cooperation between “a multidisciplinary team including the intensivist, obstetrician, maternal-fetal medicine specialist, anesthesiologist, neonatologist, nursing specialist, and transfusion medicine expert is key to optimize outcomes.”14 The ICU telemedicine program we describe (Figure 2) provides a framework for such multidisciplinary cooperation.
Continuous, remote monitoring and real-time analysis of data from the bedside and the electronic medical record (EMR) allow for early detection of and timely response to acute events and negative trends. Remote ICU telemedicine programs improve early recognition and intervention for clinically important events in general ICU patients and reduce both mortality and ICU length of stay.18,19 The pregnant physiology requires adjustments to the existing ICU telemedicine alerting algorithms, especially for patients in labor. Fundamental to the program is 2-way high-definition audio and video communication that allows the virtual presence of critical care physicians and nurses at the bedside to respond to alerts or alarms, assess the patient, and assist with care as needed. Immediate access to the remote ICU team can also be initiated at any time from the patient’s bedside.
Implementing the monitoring (telemedicine ICU) tier requires electronic access to real-time critical care monitoring, the EMR, laboratory and pharmacy data, and 2-way high-definition audiovisual connections between the remote ICU telemedicine program center and the patient’s bedside. The need for specialized equipment would either require the transfer of patients to designated hardwired rooms in the obstetrics unit or, alternatively, portable telemedicine ICU and monitoring carts that can be interfaced with the telemedicine hardware and software. This latter approach has the advantage of allowing telemedicine ICU capability in any patient room. Because it may be difficult to predict which patients from the general obstetric population will require telemedicine ICU and transfer to an available wired room, we currently believe that a portable system is more efficient and cost-effective. In addition, the telemedicine ICU monitoring component requires that the hospital either has a functioning telemedicine ICU program that is able to cover the obstetric unit or can outsource this service to another telemedicine ICU program.
This care model also requires that a core group of obstetric nurses receive augmented training in critical care nursing, including management of arterial lines and vasoactive infusions, or that critical care nurses float to the labor floor to comanage these patients when needed. The presence of dual-trained nurses in the obstetric unit can provide peer experts for other labor nurses and can facilitate integration of the bedside and remote ICU teams. This approach also requires a commitment on the part of both the bedside obstetric teams and the remote ICU teams to work cooperatively with clear processes and lines of communication, particularly for emergencies.
In this paradigm, clinical management remains the primary responsibility of the obstetric/obstetric anesthesia team who, together with the specialist nurse, remain the first responders for management of acute issues. The critical care physicians and nurses oversee monitoring, provide clinical recommendations, and function in an advisory or consultative role. Remote monitoring by the telemedicine ICU team provides an additional layer of vigilance that is independent of the number or level of acuity of other patients on the labor floor or the availability of additional obstetric or anesthesiology personnel. In the event of deterioration, they can advise and begin discussion about when patients require formal ICU admission. The specific details of these interactions can be determined at the institutional level within the COSMIC model. The telemedicine ICU service could be used to provide expert backup within a single institution or between different institutions.
A patient with moderate aortic stenosis experienced mild volume overload at 25 weeks gestation, presenting as tachycardia, and shortness of breath on minimal exertion. After cardiology consultation and resulting changes in her medication regimen, she improved clinically. After multidisciplinary discussion, the plan was for a vaginal delivery close to term, with epidural analgesia and invasive arterial pressure monitoring. During vaginal delivery, the patient would be managed in the labor unit with an ICU-trained obstetric nurse, with the additional support of a telemedicine ICU. This additional level of care would be maintained on site in the labor unit into the postpartum period. In such a case, COSMIC offers enhanced maternal monitoring throughout labor and similar postnatal monitoring without activating a formal ICU admission, allowing the mother to remain in close proximity to both her obstetricians and her baby.
The fourth tier of COSMIC is formal admission to an ICU. The decision to admit a patient to an ICU will depend on the level of care available on the floor, but in most cases, it would be reserved for patients requiring cardiorespiratory support such as endotracheal intubation, ventilation, or aggressive hemodynamic support with vasopressors.1
THE ADVANTAGES OF COSMIC TO AN INSTITUTION
Complex medical comorbidities are increasingly common during pregnancy. The COSMIC approach provides an organized process for obtaining antenatal anesthesia assessment of medically challenging obstetric patients and for obtaining other specialist consultation as needed. Such a service addresses the rising need for complex medical management in pregnant patients, improves the management of antenatal risk, and facilitates multidisciplinary communication and planning.
The COSMIC program provides the surveillance framework to facilitate early detection and management of pregnant patients who are clinically deteriorating within the context of a global obstetric inpatient population. Although we have described our implementation of the COSMIC concept, other institutions adopting it would identify which detection thresholds best suit their needs.
The COSMIC model also provides a monitoring framework in the labor unit, using nurses with both ICU and obstetric training, and ICU-level monitoring and telemedicine ICU supervision in selected high-risk obstetric patients. This monitoring would provide an additional level of vigilance for those patients who are not deemed sufficiently unstable for ICU referral. For patients who need ICU-level monitoring but not cardiorespiratory support, COSMIC brings ICU facilities to the bedside in the labor unit. By so doing, it allows care to be given in immediate proximity to obstetricians, obstetric anesthesiologists, and the operating room and avoids removing obstetric teams away from other patients on the labor unit. In a specialty for which minutes of delay can substantially affect maternal and fetal outcomes, these advantages may be crucial.
COSMIC also may provide opportunities for medical outsourcing that benefit both large and small institutions. Smaller institutions that lack high-risk perinatal care or intensivists may link with a tertiary care center that can provide consultation as to whether it is advisable to deliver locally with or without remote monitoring, or whether it is preferable to transfer care to the referral center. This model may also provide service to patients deemed inappropriate for transport. To the tertiary center, COSMIC may provide a source of revenue if it offers outsourcing for consultation, surveillance, and monitoring services over a wider population, and it may also allow the tertiary center to expand its referral base.
A large tertiary center operates its own telemedicine ICU service, which covers multiple ICUs on site and also multiple ICUs in other hospitals. In this hospital, the telemedicine ICU would also cover its own labor unit. A small rural hospital a 4-hour drive away is interested in using an outsourcing model of COSMIC. It identifies high-risk patients using criteria similar to those in the Table and consults with the larger tertiary care center regarding the suitability of the remote hospital for delivery. The telemedicine ICU cover would be provided by the tertiary care center for the occasional high-risk patient or for unscheduled admissions of patients who are not deemed safe for transport.
It is not clear what the net financial impact of COSMIC will be. Expenditures in one area such as automated surveillance, remote telemonitoring, and ICU monitoring carts may be offset by the savings from reducing formal ICU referrals or by the revenue from providing outsourcing services. It is likely that the realities will change from institution to institution, and COSMIC should be flexible enough to accommodate different types of coverage models.
While the COSMIC paradigm has not been assessed objectively, the introduction of telemedicine has improved outcomes in nonobstetric ICU patients.18,19 In addition to improving clinical care, data from these paradigms may advance our understanding of predictors of adverse maternal outcomes. By studying the clinical presentations of sepsis, hemorrhage, preeclampsia, and thromboembolism in pregnant patients, COSMIC may identify more accurate clinical early warning systems to herald impending maternal collapse. A strong quality assurance (QA) process is essential to improvement and discovery, and it would be important to review response times, processes that were not followed, delays in treatment, and appropriateness of system generated alerts.
The implementation of a multitiered, interdisciplinary surveillance program such as COSMIC requires the leadership and clinical support of obstetricians, obstetric anesthesiologists, critical care physicians, nursing staff, and hospital administration. We believe that it will provide a platform for enhanced connection between the obstetric unit and the ICU.
Name: Ellen M. Lockhart, MD.
Contribution: This author helped jointly write the manuscript.
Name: Alexander Hincker, MD.
Contribution: This author helped jointly write the manuscript.
Name: Thomas T. Klumpner, MD.
Contribution: This author helped draft parts of the section on surveillance.
Name: Jennifer Hofer, MD.
Contribution: This author helped draft the section on critical care.
Name: Alison G. Cahill, MD, MSCI.
Contribution: This author helped critically review and edit the manuscript.
Name: Arvind Palanisamy, MD, FRCA.
Contribution: This author helped critically review and edit the manuscript.
Name: Walter Boyle, MD.
Contribution: This author helped draft the section on telemedicine ICU.
Name: Yehuda Ginosar, MBBS.
Contribution: This author conceived the idea of COSMIC and helped jointly write the manuscript.
This manuscript was handled by: Avery Tung, MD, FCCM.
1. American College of Obstetrics and Gynecology. Practice bulletin No. 170: critical care in pregnancy. Obstet Gynecol. 2016;128:e147–e154.
2. ICNARC. Female Admissions (Aged 16–50 years) to Adult, General Critical Care Units in England, Wales and Northern Ireland reported as ‘Currently Pregnant’ or ‘Recently Pregnant’: Report from the Intensive Care National Audit and Research Centre. Available at: https://www.icnarc.org/Our-Audit/Audits/Cmp/Our-National-Analyses/Obstetrics
. Accessed March 27, 2019.
3. MacDorman MF, Declercq E, Cabral H, Morton C. Recent increases in the US maternal mortality rate: disentangling trends from measurement issues. Obstet Gynecol. 2016;128:447–455.
4. Fridman M, Korst LM, Chow J, Lawton E, Mitchell C, Gregory KD. Trends in maternal morbidity before and during pregnancy in California. Am J Public Health. 2014;104suppl 1S49–S57.
5. Lewis G. The Confidential Enquiry into Maternal and Child Health (CEMACH) Saving Mothers’ Lives: Reviewing Maternal Deaths to Make Motherhood Safer 2003–2005: The Seventh Report of the Confidential Enquiries into Maternal Deaths in the United Kingdom. 2007.London, UK: CEMACH.
6. Mhyre JM, D’Oria R, Hameed AB, et al. The maternal early warning criteria: a proposal from the national partnership for maternal safety. Obstet Gynecol. 2014;124:782–786.
7. Gat R, Hadar E, Orbach, -Zinger S, Shochat T, Kushnir S, Einav S. Distribution of extreme vital signs and complete blood count values of healthy parturients: a retrospective database analysis and review of the literature. Anesth Analg. 2018 [Epub ahead of print].
8. Edwards SE, Grobman WA, Lappen JR, et al. Modified obstetric early warning scoring systems (MOEWS): validating the diagnostic performance for severe sepsis in women with chorioamnionitis. Am J Obstet Gynecol. 2015;212:536.e1–536.e8.
9. Shearer B, Marshall S, Buist MD, et al. What stops hospital clinical staff from following protocols? An analysis of the incidence and factors behind the failure of bedside clinical staff to activate the rapid response system in a multi-campus Australian metropolitan healthcare service. BMJ Qual Saf. 2012;21:569–575.
10. Mitchell C, Lawton E, Morton C, McCain C, Holtby S, Main E. California Pregnancy-Associated Mortality Review: mixed methods approach for improved case identification, cause of death analyses and translation of findings. Matern Child Health J. 2014;18:518–526.
11. Klumpner TT, Kountanis JA, Langen ES, Smith RD, Tremper KK. Use of a novel electronic maternal surveillance system to generate automated alerts on the labor and delivery unit. BMC Anesthesiol. 2018;18:78.
12. Kheterpal S, Shanks A, Tremper KK. Impact of a novel multiparameter decision support system on intraoperative processes of care and postoperative outcomes. Anesthesiology. 2018;128:272–282.
13. Quinn A, Berry A, Banerjee A, et al. Care of the Critically Ill Woman in Childbirth; Enhanced Maternal Care. 2018. London, UK: Royal College of Anaesthetists; Available at: https://www.rcoa.ac.uk/system/files/EMC-Guidelines2018.pdf
. Accessed March 27, 2019.
14. Guntupalli KK, Hall N, Karnad DR, Bandi V, Belfort M. Critical illness in pregnancy: part I: an approach to a pregnant patient in the ICU and common obstetric disorders. Chest. 2015;148:1093–1104.
15. Bernstein PS, Martin JN Jr, Barton JR, et al. National partnership for maternal safety: consensus bundle on severe hypertension during pregnancy and the postpartum period. Obstet Gynecol. 2017;130:347–357.
16. Main EK, Goffman D, Scavone BM, et al.; National Partnership for Maternal Safety; Council on Patient Safety in Women’s Health Care. National Partnership for Maternal Safety: consensus bundle on obstetric hemorrhage. Obstet Gynecol. 2015;126:155–162.
17. Canobbio MM, Warnes CA, Aboulhosn J, et al.; American Heart Association Council on Cardiovascular and Stroke Nursing; Council on Clinical Cardiology; Council on Cardiovascular Disease in the Young; Council on Functional Genomics and Translational Biology; and Council on Quality of Care and Outcomes Research. Management of pregnancy in patients with complex congenital heart disease: a scientific statement for healthcare professionals from the American Heart Association. Circulation. 2017;135:e50–e87.
18. Lilly CM, Cody S, Zhao H, et al.; University of Massachusetts Memorial Critical Care Operations Group. Hospital mortality, length of stay, and preventable complications among critically ill patients before and after tele-ICU reengineering of critical care processes. JAMA. 2011;305:2175–2183.
19. Willmitch B, Golembeski S, Kim SS, Nelson LD, Gidel L. Clinical outcomes after telemedicine intensive care unit implementation. Crit Care Med. 2012;40:450–454.