Sepsis remains an important cause of maternal mortality. Deaths due to maternal sepsis increased in the United States from 1998 to 2008.1 The Mothers and Babies: Reducing Risk through Audits and Confidential Enquiries across the UK report for 2009–2012 indicated almost 25% of all maternal deaths were the result of sepsis.2 The diagnosis of sepsis requires a source of infection and that the criteria for the systemic inflammatory response syndrome be met.3 The diagnosis of sepsis in pregnancy is challenging because many healthy pregnant women will manifest some component of the systemic inflammatory response syndrome criteria (except temperature) in the absence of infection.4
Little is known about pregnant and postpartum women who died of maternal sepsis in the United States. Although maternal deaths due to sepsis are rare, they are potentially preventable with early recognition and treatment: mortality in severe sepsis and septic shock has been found to increase 7.6% for each hour delay in appropriate antibiotic treatment.5 The aims of this study are to identify maternal deaths due to sepsis in the state of Michigan from 1999 to 2006, review the events leading to diagnosis, and evaluate treatment to identify areas for improvement.
MATERIALS AND METHODS
The institutional review board for the Michigan Department of Community Health approved this study. The institutional review boards of William Beaumont Health System and the University of Michigan Health System declared this study not regulated because all participants are deceased.
In this case series from a cohort of maternal deaths, all maternal deaths reported in the state of Michigan in 1999–2006 during pregnancy and up to 1 year postpartum were identified with Maternal Mortality Surveillance records from the Michigan Department of Community Health. This study period was chosen based on availability of records from the Michigan Department of Community Health. The definition of maternal death due to sepsis during pregnancy and up to 42 days postpartum was the presence of any of the following: 1) sepsis listed on the death certificate as the cause of death, 2) determination that sepsis was the cause of death by consensus of the Maternal Mortality Medical Surveillance Committee, or 3) identification of a specific source of infection leading to organ failure, consistent with the accepted definition of severe sepsis by the American College of Chest Physicians and Society of Critical Care Medicine.3 “Maternal sepsis contributing deaths” were defined as deaths from another cause, but with sepsis as a contributing factor. After identification, data were collected by an obstetric anesthesiologist (M.E.B.) and double checked by an obstetrician (F.W.J.A.). Data collected for each chart included demographic information, details of presentation on admission, unresolved issues from delivery hospitalization if applicable, antibiotic selection throughout hospitalization, culture results, vital signs at presentation, vital signs at time of escalation of care, laboratory results, case narrative, contributing causes leading to death, cause(s) of death, and findings from the Michigan Maternal Mortality Surveillance review process. All cases were reviewed by all authors and disagreements were resolved by discussion. Deidentified record summaries were provided to an infectious disease specialist (K.R.) for determination of whether initial and eventual antibiotic choices were appropriate based on the clinical scenario and recommended guidelines the year the case occurred. For example, ampicillin–sulbactam was an appropriate choice for an intraabdominal process in the early 2000s; however, with the emergence of resistance among organisms, especially Escherichia coli, it is currently not recommended.6 Delay of care was defined as 1) not providing timely appropriate antibiotics (within 1 hour of evidence of sepsis) or 2) failure to escalate care to an intensive care unit or tertiary care center after it appeared by chart review that the patient was critically ill based on meeting objective parameters for admission to an intensive care unit as defined by the American College of Obstetricians and Gynecologists' (the College) Practice Bulletin on critical care in pregnancy.7
Michigan Maternal Mortality Surveillance identifies maternal deaths in the following manner: voluntary reporting of maternal deaths (since inception in 1950), screening death certificates for causes of death specific to pregnancy (starting in 1950), matching death certificates of women of reproductive age (10–45 years old) with live birth certificates (1990–present) and fetal death certificates (2004–present) of the current year and previous year, and checking a box on the death certificate indicating the patient was either pregnant or pregnant within 1 year at the time of death (2004–present).8 The new changes in 2004 brought an increase in reporting of maternal deaths. The Michigan Department of Community Health uses the College–Centers for Disease Control and Prevention definitions to classify maternal deaths. Pregnancy-associated deaths are defined as the death of a woman while pregnant or within 1 year of pregnancy, irrespective of cause. Pregnancy-related deaths are defined as the death of a woman while pregnant or within 1 year of termination of pregnancy, irrespective of the duration and site of the pregnancy, from any cause related to or aggravated by her pregnancy or its management, but not from accidental or incidental causes.9 Every pregnancy-associated death in the state of Michigan is reviewed by an expert review committee, and a final cause of death and a determination about pregnancy relatedness are made for each case.
All results were reported as proportions. If data for a particular element were missing, it was reported as indeterminate when evaluating a clinical situation and was excluded from the numerator and denominator for categorical data.
From 1999 to 2006 there were 1,047,857 live births in the state of Michigan and 558 pregnancy-associated deaths (53.3 deaths/100,000 live births). Of these, 151 (14.4 deaths/100,000 live births) were pregnancy-related deaths.10 (In e-mail personal communication to Sarah Lyon-Callo, MA, MS, from the Michigan Department of Community Health on June 11, 2015, the previously unpublished maternal mortality ratios for 2005 and 2006 were confirmed.) Maternal sepsis was the cause of death in 4% (22/558) of pregnancy-associated deaths reviewed during the time period and 15% (22/151) of all pregnancy-related deaths. Sepsis contributing deaths comprised another 1% (8/558) of pregnancy-associated deaths for a total of 5% (30/558) of pregnancy-associated deaths and 20% (30/151) pregnancy-related deaths. No cases were excluded as a result of differences in interpretation. The rate of maternal deaths due to sepsis was 2.1 per 100,000 live births, and the rate of maternal death due to sepsis and sepsis-contributing deaths was 2.9 per 100,000 live births. There was an increase in maternal deaths due to sepsis at home starting in 2004, with the addition of the pregnancy check box on the death certificate.
The mean maternal age of the 22 patients who died from sepsis was 27 years (range 13–40 years) (Table 1). The clinical details of each case are presented in the Appendix (available online at http://links.lww.com/AOG/A684).
Of the women presenting to the hospital with sepsis in Table 2, 75% (9/12) demonstrated one or more of the following vital sign derangements at presentation: heart rate higher than 120 beats per minute, respiratory rate higher than 30 breaths per minute, systolic blood pressure lower than 90 mm Hg, and SpO2 less than 95% on room air (Maternal Early Warning Criteria).11 Only 18% (2/11) of septic women were febrile on presentation and 25% (3/12) never developed a fever during hospitalization. The clinical details of each case are presented in the Appendix (http://links.lww.com/AOG/A684).
Delays of care were found in the majority of patients. Initial antibiotics with inadequate coverage based on the clinical situation were identified in 11 of 15 (73%) patients who received hospital care for sepsis; in 2 of 15 (13%) patients whether appropriate antibiotics were used was indeterminate. Only 2 of 15 (13%) patients received appropriate initial antibiotics. Once infectious disease consultation or critical care consultation was obtained and antibiotics were changed, in 67% (10/15) subsequent antibiotic care was appropriate for the clinical situation, 20% (3/15) of women did not live long enough for subsequent therapy, and in 13% (2/15) of patients the appropriateness of therapy could not be determined. Of patients who presented to the hospital for care, delay of escalation of care was found in 53% (8/15), no delay was found in 33% (5/15), and whether timing was delayed could not be determined in 13% (2/15). Seven patients died at home or were dead on arrival, and it could not be determined whether they sought medical care before death. The most common organism was Streptococcus pyogenes in 29% (4/14) of women with an identified organism.
Common elements in the maternal deaths due to sepsis illustrate three key delays that may have contributed to the deaths: in recognition of sepsis, in administration of appropriate antibiotics, and in escalation of care.
The majority of women who died of maternal sepsis were afebrile on presentation and did not develop a fever for hours or days; 25% (3/12) did not develop a fever at all. Although a temperature of lower than 36°C or higher than 38°C is one of the criteria for systemic inflammatory response syndrome, it is not required for the diagnosis of sepsis. Sepsis should be considered in critically ill women regardless of temperature. Although Maternal Early Warning Criteria was not a recommendation at the time period for which the data were collected, it is reassuring that 75% of women had one or more criteria for Maternal Early Warning Criteria on presentation and would have triggered the system had it been in place. The National Partnership for Maternal Safety recommends that all women who meet any of the Maternal Early Warning Criteria receive prompt bedside evaluation by a physician or other clinician with the ability to activate resources to initiate emergency diagnostic and therapeutic interventions as needed. Maternal Early Warning Criteria screens for multiple types of possible maternal morbidity through a simple bedside assessment.11
Delay in diagnosis may have been the result of incomplete data available for decision-making. Many records demonstrated incomplete assessments. Pulse oximetry was the most common missing assessment. The majority of women with a pulse oximetry reading documented at admission would have triggered Maternal Early Warning Criteria. Record reviews suggest that respiratory rate measurements may have been inaccurate. Of 11 women with respiratory rate recorded, eight had measurements of 18, 20, or 24 breaths per minute. On review, the pH and lactate levels taken shortly after decompensation and respiratory rate documentation depicted a state of metabolic acidosis so severe that a normal respiratory rate would not likely have been possible. Although respiratory rate is poorly documented in clinical settings, it has been shown to correlate with severity of sepsis in the general population.12
For cases with sufficient records for a determination, the majority were found to have delays in care. Mothers and Babies: Reducing Risk through Audits and Confidential Enquiries across the UK recommends early consultation with an infectious disease specialist in cases of maternal sepsis.2 Each hour delay of appropriate antibiotics in patients with severe sepsis or septic shock is associated with an increase in mortality of 7.6%.5 For treatment of postpartum endometritis, a combination of clindamycin and gentamicin with the addition of ampicillin in refractory cases is recommended by the College and is the most common regimen used.13 A recent meta-analysis showed more treatment failures occurred with the use of a penicillin and gentamicin compared with gentamicin and clindamycin.14 Clindamycin was underused in this population and was not given to any patient who died of Group A Streptococcal infection. Clindamycin, a protein synthesis inhibitor shown to inhibit toxin production, is beneficial because the efficacy is not decreased by the stage of growth and has been shown to decrease mortality in Group A Streptococcal infection.15–17 Although our study highlights delay of appropriate antibiotics during 1999–2006, we were unable to determine whether there was improvement in the timing and antibiotic selection after the publication of Surviving Sepsis Campaign Guidelines over time.18 However, in this case series, antibiotic selection was not improved in 2004, 2005, or 2006 after publication of the first set of guidelines. It is unlikely that more current data would have shown changes in clinical management; delay in recognition led to delay in treatment for the majority of patients.
The limitations of this study include that it is a case series of women rather than a study including women in a control group. Although our cases are limited to 22 women who died of sepsis, given the rarity of maternal deaths due to sepsis, this study represents a large number of maternal deaths due to sepsis in the United States. It is possible that there are cases that may have not been identified during the years of 1999–2006 because the state of Michigan does not have mandatory reporting of maternal deaths. Our records were also limited to the records where the death occurred. Prenatal records, delivery records, and records from transferring hospitals were unavailable in some cases. Owing to inadequate granularity of the data available, we were unable to fully assess whether early goal-directed therapy or Surviving Sepsis Campaign Guidelines were followed to provide more specific areas for improvement.19,20
Although maternal deaths due to sepsis are rare, early diagnosis and treatment could potentially make them preventable. Observations from this study lead to the following suggestions: 1) consider maternal sepsis in critically ill women even in the absence of fever; 2) administer early appropriate antibiotic therapy and use consultation (maternal–fetal medicine, infectious disease, critical care) early if there is marked derangement in vital signs or an inadequate response; 3) vital sign derangement should be investigated, closely monitored, and, if severe, should prompt escalation of care; and 4) vital signs should be accurately taken at appropriate intervals consistent with the maternal condition. If further studies reveal a similar rate of maternal deaths at home due to sepsis and postpartum deaths after discharge, there may be a role for improving patient education concerning when to seek medical attention, timing of postpartum visits, and use of home visits.
1. Bauer ME, Bateman BT, Bauer ST, Shanks AM, Mhyre JM. Maternal sepsis mortality and morbidity during hospitalization for delivery: temporal trends and independent associations for severe sepsis. Anesth Analg 2013;117:944–50.
2. Knight M, Kenyon S, Brocklehurst P, Neilson J, Shakespeare J, Kurinczuk JJ. Saving lives, improving mothers' care—lessons learned to inform future maternity care from the UK and Ireland confidential enquiries into maternal deaths and morbidity 2009–2012. Oxford (UK): National Perinatal Epidemiology Unit, University of Oxford; 2014.
3. Bone RC, Balk RA, Cerra FB, Dellinger RP, Fein AM, Knaus WA, et al.. Definitions for sepsis and organ failure and guidelines for the use of innovative therapies in sepsis. The ACCP/SCCM Consensus Conference Committee. American College of Chest Physicians/Society of Critical Care Medicine. Chest 1992;101:1644–55.
4. Bauer ME, Bauer ST, Rajala B, MacEachern MP, Polley LS, Childers D, et al.. Maternal physiologic parameters in relationship to systemic inflammatory response syndrome criteria: a systematic review and meta-analysis. Obstet Gynecol 2014;124:535–41.
5. Kumar A, Roberts D, Wood KE, Light B, Parrillo JE, Sharma S, et al.. Duration of hypotension before initiation of effective antimicrobial therapy is the critical determinant of survival in human septic shock. Crit Care Med 2006;34:1589–96.
6. Paterson DL, Rossi F, Baquero F, Hsueh PR, Woods GL, Satishchandran V, et al.. In vitro susceptibilities of aerobic and facultative Gram-negative bacilli isolated from patients with intra-abdominal infections worldwide: the 2003 Study for Monitoring Antimicrobial Resistance Trends (SMART). J Antimicrob Chemother 2005;55:965–73.
7. Critical care in pregnancy. ACOG Practice Bulletin No. 100. American College of Obstetricians and Gynecologists. Obstet Gynecol 2009;113:443–50.
8. Mhyre JM, Riesner MN, Polley LS, Naughton NN. A series of anesthesia-related maternal deaths in Michigan, 1985–2003. Anesthesiology 2007;106:1096–104.
9. Berg C, Danel I, Atrash H, Zane S, Bartlett L, editors. Strategies to reduce pregnancy-related deaths: from identification and review to action. Atlanta (GA): Centers for Disease Control and Prevention; 2001.
10. Michigan Maternal Mortality Surveillance (MMMS) 1999–1994 report. 2006. Available at: http://www.michigan.gov/documents/mdch/MDCH_MCH_Epi_VG_MMMSReport2006_220171_7.pdf
. Retrieved June 3, 2015.
11. Mhyre JM, D'Oria R, Hameed AB, Lappen JR, Holley SL, Hunter SK, et al.. The maternal early warning criteria: a proposal from the national partnership for maternal safety. Obstet Gynecol 2014;124:782–6.
12. Kenzaka T, Okayama M, Kuroki S, Fukui M, Yahata S, Hayashi H, et al.. Importance of vital signs to the early diagnosis and severity of sepsis: association between vital signs and sequential organ failure assessment score in patients with sepsis. Intern Med 2012;51:871–6.
13. American Academy of Pediatrics, The American College of Obstetricians and Gynecologists. Guidelines for perinatal care. 7th ed. Washington, DC: The American College of Obstetricians and Gynecologists; 2012.
14. Mackeen AD, Packard RE, Ota E, Speer L. Antibiotic regimens for postpartum endometritis. The Cochrane Database of Systematic Reviews 2015, Issue 2. Art. No.: CD001067. DOI: 10.1002/14651858.CD001067.pub3.
15. Mulla ZD, Leaverton PE, Wiersma ST. Invasive group A streptococcal infections in Florida. South Med J 2003;96:968–73.
16. Carapetis JR, Jacoby P, Carville K, Ang SJ, Curtis N, Andrews R. Effectiveness of clindamycin and intravenous immunoglobulin, and risk of disease in contacts, in invasive group a streptococcal infections. Clin Infect Dis 2014;59:358–65.
17. Stevens DL, Bisno AL, Chambers HF, Dellinger EP, Goldstein EJ, Gorbach SL, et al.. Practice guidelines for the diagnosis and management of skin and soft tissue infections: 2014 update by the Infectious Diseases Society of America. Clin Infect Dis 2014;59:e10–52.
18. Dellinger RP, Carlet JM, Masur H, Gerlach H, Calandra T, Cohen J, et al.. Surviving sepsis campaign guidelines for management of severe sepsis and septic shock. Crit Care Med 2004;32:858–73.
19. Rivers E, Nguyen B, Havstad S, Ressler J, Muzzin A, Knoblich B, et al.. Early goal-directed therapy in the treatment of severe sepsis and septic shock. N Engl J Med 2001;345:1368–77.
20. Dellinger RP, Levy MM, Rhodes A, Annane D, Gerlach H, Opal SM, et al.. Surviving sepsis campaign: international guidelines for management of severe sepsis and septic shock: 2012. Crit Care Med 2013;41:580–637.