Nationwide study of mortality and survival in pregnancy-related streptococcal toxic shock syndrome

Streptococcal toxic shock syndrome is associated with the highest rates of infection-related maternal mortality. We conducted a comparative analysis of background factors and treatment course between patients who survived and those who did not to improve our understanding of the optimal initial treatment approach for this fulminant disease.


Conclusions
Intensive care, including systemic administration of antibiotics, may contribute to maternal survival when administered immediately (within 1 hour) after the onset of fulminant streptococcal toxic shock syndrome. Eliciting a family history of streptococcal infection and conducting a rapid antigen test can identify patients needing early intervention.

Background
Streptococcal toxic shock syndrome (STSS) is a severe complication of invasive group A streptococcal (GAS) infection that can rapidly progress to shock and multi-organ failure (MOF) and is associated with high mortality [1,2]. The clinical course of STSS is particularly aggressive if it occurs during pregnancy and the puerperium. Its early symptoms (e.g. fever and upper respiratory manifestations), mimic those of a common cold, making early diagnosis di cult. Therefore, affected patients can worsen suddenly and develop systemic shock within days of onset of initial symptoms [3]. According to a Japanese study that analyzed all maternal deaths that occurred during pregnancy and within ≤ 42 days after childbirth, 13 (4.1%) of the 317 (4.3/100,000) cases of STSS recorded from 2010-2017 were caused by GAS. This causative organism is responsible for the highest maternal mortality rate due to severe infection, as reported by our earlier study of the backgrounds and characteristics of maternal deaths caused by GAS [3,4].
A comparative analysis of characteristics and course of treatments between patients who died and of those who survived after developing pregnancy-related STSS can reveal information that may be useful for both early detection and management of this condition. However, pregnancy-related STSS is a rare disease, making it di cult to nd enough cases to conduct an epidemiological analysis. Thus, nationwide case-control analyses of the condition are lacking, although a few case series and systematic reviews on the topic are available [5][6][7][8].
The purpose of this study was to conduct a nationwide analysis of all patients diagnosed with pregnancy-related STSS caused by GAS, irrespective of the eventual outcome. We aimed to conduct a comparative analysis of background factors and treatment course between patients who survived and those who did not in order to formulate an initial approach for this fulminant disease.
Materials And Methods

Study design and population
We conducted a retrospective observational study and sourced clinical data of pregnant women diagnosed with STSS from two national organizations.
a. Data of patients who died due to STSS The Japan Association of Obstetricians and Gynecologists (JAOG) is responsible for registering all maternal deaths and holds detailed reports on almost all maternal deaths that have occurred in Japan since 2010. In this study, we included data from all maternal deaths caused by STSS as determined by the Japan Maternal Death Exploratory Committee (JMDEC) (Chairman: Ikeda, T.) based on JAOG data from 2010-2017. JMDEC consists of doctors from several elds, including 23 obstetricians, 6 anesthesiologists, 3 pathologists, 2 forensic doctors, 2 psychiatrists, 1 emergency physician, 1 cardiologist, and 1 neurosurgeon. The JMDEC routinely analyzes causes of all maternal deaths registered with the JAOG and based on their ndings, provides recommendations for decreasing maternal mortality. In Japan, maternal death is de ned as that which occurs during pregnancy and within ≤ 42 days of childbirth due to any reason that is related to pregnancy or its management, regardless of the duration and the site of gestation [9].
b. Data of patients who survived STSS We collated clinical data of pregnant women diagnosed with STSS who were treated from 2015 to 2017 in institutions registered with the Perinatal Research Network in Japan (PRNJ) and survived until discharge. PRNJ is an organization that includes obstetricians working at both general and local perinatal medical centers and was established to improve the overall quality of perinatal medical care. As of April 2018, 192 institutions are registered with the PRNJ, which accounts for 99% (107/108) and 29% (85/298) of general and local perinatal medical centers in Japan, respectively. We rst conducted a primary survey of these institutions to identify eligible (survivors of STSS) patients in their records and requested their cooperation for collaborative research. If they consented to participate and shared that they had records of patients with relevant symptoms, we then sent a detailed questionnaire to collect all relevant clinical information.

Case de nitions of STSS
We de ned STSS cases, as speci ed by the Japanese Ministry of Health, Labour and Welfare (MHLW), as those that satis ed the following conditions simultaneously: 1) the patient had evidence of systemic shock with organ dysfunction (e.g., acute kidney failure, acute respiratory distress syndrome, necrotizing fasciitis); and 2) GAS was isolated and identi ed from otherwise sterile sites (e.g., blood, ascitic uid, and necrotizing soft tissues) [10]. These criteria are similar to those used for de ning STSS in the UK [11] and the US [12].

Statistical analysis
The Statistical Package for Social Science (SPSS) software program (Windows version 20.0J; Chicago, IL, USA) was used for data analysis.

Ethical considerations
This study was approved by the ethics committee of the Toho University Omori Medical Center before the commencement of the study (Approval No: M18055). The requirement for written informed consent was waived by the committee due to the retrospective nature of the study.

Results
One hundred and twenty-six institutions registered with the PRNJ consented to participate in this study and provided data of STSS survivors. Study patients were divided into two groups of STSS-related deaths and survivors for the comparative analysis, and the characteristics of the two study groups are presented in Table 1. We found no statistical differences between groups related to age, parity, and season, gestational age or patient location (home/hospital) at the time of symptom onset. In both groups, multiparas were more susceptible (deaths: 76.9%; survivors: 86.7%), and the disease occurred more frequently in winter and spring (deaths: 61.5%; survivors: 66.7%). The majority of patients experienced an onset of symptoms outside medical facilities (deaths: 69.2%; survivors: 66.7%).

Discussion
This study had three principal ndings. Firstly, no background factors (i.e. age, parity, season, gestational period, initial symptoms, and location of disease onset) were found to be different between STSS patients who survived and those who died. Secondly, intensive care tended to be initiated soon after the onset of fulminant disease (within 1 hour) in the group of survivors, since these patients were given systemic antibiotics at their rst visit to a clinic where they presented with initial symptoms. Further, fetal survival despite the onset of fulminant STSS tended to be higher among the group of maternal survivors, with an associated improved prognosis of the newborn following delivery. Thirdly, in patients experiencing non-speci c initial symptoms of GAS infection, such as upper abdominal pain and genital bleeding, early intervention was offered based on a positive family history of GAS infection or a positive rapid antigen test for GAS.
Background characteristics were not indicative of disease prognosis. Pregnancy-related STSS occurs more commonly in parous women and is more likely to occur in winter and spring [6]. GAS infection is seasonal and usually occurs in children, and can be transmitted either via droplet infection or contact between members of a family. Our ndings also re ected this tendency in both mortality and survival groups. Several studies have reported that the maternal mortality rate decreases if antibiotics are administered within 1 hour of the diagnosis of sepsis [13][14][15]. The Surviving Sepsis Campaign guidelines strongly recommend initiating treatment with systemic antibiotics within 1 hour of sepsis diagnosis, since every consecutive hour of delay in antibiotic administration increases mortality by 7.6% [16]. Our study ndings support this recommendation. Fulminant GAS infection is likely to progress to sepsis rapidly. If fulminant infection is suspected, the patient should be urgently transferred to a specialized center where treatment with systemic high-dose, broad-spectrum antibiotics can be initiated without awaiting con rmation of the diagnosis on culture tests [17]. This is important since the rate of fetal survival, and the subsequent neonatal outcome, was higher in the maternal survivor group, even with the onset of fulminant disease. We hypothesized that compared to the mortality group, patients in the survivor group were likely diagnosed at a stage where STSS was still reversible and responsive to treatment. STSS causes sepsis and MOF due to systemic GAS invasion that occurs after the initial colonization of the gravid uterus and subsequent induction of severe in ammation [18]. Consequently, much stronger uterine contractions are induced during delivery, releasing large quantities of germs and toxins into the maternal systemic circulation [19]. Intrauterine fetal death is assumed to occur during this process, due to spread of in ammation to the fetus through the uterine muscular wall, associated with concomitantly low umbilical blood ow as a result of reduced systemic perfusion in the mother due to sepsis. Therefore, if the fetus is still alive in women suspected of having STSS, therapeutic interventions that can save both mother and infant are needed, and therefore patients should be urgently transferred to a specialized medical institution that can provide intensive care for both.
Patients with non-speci c symptoms of STSS received early treatment only if they reported a positive family history of streptococcal infection or tested positive for GAS antigen. Since close contact with patients with streptococcal infection is associated with the transmission of STSS [17,20,21], it is important to ask pregnant women about their medical history and to consider GAS infection as part of the differential diagnosis of women presenting with symptoms of the common cold, e.g., sore throat. If GAS infection is suspected, the need for a rapid antigen test or the administration of antibiotics can be assessed using the Centor criteria [22]. To prevent unnecessary use of antibiotics, it is recommended that treatment should be delayed until results of the GAS rapid antigen test or throat swab culture test are available, even if a patient scores 3 or 4 points according to the Centor criteria [23]. However, since early treatment of patients with GAS infection can improve prognosis, we believe that the GAS rapid antigen test and/or throat swab culture test should be performed urgently in suspected patients. Further, empirical administration of antibiotics should be considered based on the clinical picture, even if the patient scores low as per the Centor criteria.
This study collated nationwide data for the systematic study of an uncommon but serious pregnancyrelated infection. However, it has certain limitations. Firstly, since this was a retrospective observational study designed to investigate a rare disease, evaluation of rates of incidence and prevention of STSS was impossible. In developed countries, national-level surveillance is often conducted for all patients with STSS since reporting of con rmed cases is often mandatory [7,24,25]. Some of these countries have instituted guidelines for the prevention of pregnancy-related GAS infections [11]. Second, it is still unclear why the results differ from previous reports with regard to the timing of STSS onset. Further prospective studies are warranted, including studies investigating the association between perinatal antibiotic use and severe infections. Based on the results of this study, the JMDEC is aiming to prospectively accumulate survival data of rare diseases, including STSS, in the future. Third, since this was a retrospective study performed using questionnaires, we cannot exclude the possibility of recall bias. It is possible that treatment effects were overrated based on doctors' experience of success with the surviving patients, while the same ndings could be conversely underrated in the mortality group.

Conclusions
No background factors were found to be predictive of mortality or survival in patients with STSS. We found that intensive care, including systemic administration of antibiotics, offered immediately (within 1 hour) after the onset of fulminant disease may contribute to better survival. In addition to evaluating clinical symptoms in affected patients, it is also important to elicit whether there is a family history of GAS infection and to promptly conduct a rapid antigen test to determine whether early intervention in affected patients is required. This study was approved by the ethics committee of the Toho University Omori Medical Center before the commencement of the study (Approval No: M18055). The requirement for written informed consent was waived by the committee due to the retrospective nature of the study.

Consent for publication
Not applicable.

Availability of data and materials
All data supporting the ndings in the study are contained within the manuscript.

Competing interests
The authors declare that they have no competing interests.