This case demonstrates the need for prompt, decisive care in a situation where a previously well, young, new mother becomes acutely and life threateningly unwell. The presence and involvement of senior clinicians at all stages allowed for confidence in adopting a conservative approach and avoiding a laparotomy, which may have added to risk. A prompt examination under anaesthesia with the assumption of, the very much more common, retained products as an aetiology of postpartum haemorrhage is a safe approach given the lack of specificity of ultrasound in excluding this in the immediate puerperium, and even had sepsis been suspected at that stage, it would have been important to exclude these were underlying. The postpartum haemorrhage in this situation was due to sepsis, but I would highlight the addition of the antifibrinolytic tranexamic intravenously to uterotonics in the management of any significant postpartum haemorrhage.

Good interdisciplinary communication aiding decision making and ease of access to intensive care support potentially all contributed to the prevention of the tragedy of a maternal mortality.
– Dr Rosemary Reid

A 28-year-old woman, gravida five para four, presented to the emergency department (ED) 24 hours after a normal vaginal delivery with heavy vaginal blood loss and increasing abdominal pain. She reported to have soaked five maternity pads. The estimated blood loss at time of delivery was 450 ml and a second-degree tear was repaired at the time. She attended ED at 11:00 via ambulance and was assessed in ED by gynaecology: heart rate 135, blood pressure 140/88, respiratory rate 20, saturations 99 per cent, apyrexial. Examination revealed heavy vaginal bleeding and a high tender uterine fundus well above the umbilicus. She required high doses of intravenous fentanyl for her pain. A Foley catheter was inserted into the bladder. She was given intramuscular syntometrine, intravenous tranexamic acid, cefuroxime and metronidazole, along with misoprostol rectally and an oxytocin infusion. With a high fundus ongoing vaginal loss, tachycardia and high analgesia requirements, the working diagnosis was of postpartum haemorrhage secondary to retained tissue and she was consented for an examination under anaesthetic (EUA). No imaging was performed prior to transfer.

Bloods: Hb 103, WCC 6.1, CRP 77.

She was transferred to the operating theatre and was given a general anaesthetic. The EUA was performed by a registrar with senior medical officer supervision (SMO); the perineal sutures were intact, the uterus was distended but only a small amount of clot found in the uterus. With fundal massage, the uterus contracted and was below the umbilicus at the end of the procedure.

In recovery, she had ongoing abdominal pain with increasing severity and ongoing tachycardia of 140. Intravenous fentanyl, morphine and ketamine was given. On palpation of the abdomen her fundus was at the umbilicus and she was peritonitic. The question was raised of an alternative diagnosis and a CT scan was requested, which showed mild to moderate intraperitoneal free fluid and diffuse periportal oedema, an enlarged uterus with a trace of endometrial fluid and gas which correlated with recent postpartum status and EUA, but no definite cause for her symptoms found. A transabdominal ultrasound in recovery also did not add to the diagnosis.

At 17:20 she was reviewed by an O&G SMO, the cause for her symptoms was still unclear and a surgical review was requested with no additions to the management plan. In the evening, she developed increasing oxygen requirements, a chest x-ray showed bilateral lower zone atelectasis.

Point of care haemoglobin at this time was 93.

She had ongoing diffuse abdominal pain with signs of peritonitis. A further surgical review was requested, it was felt appropriate investigations had been performed and no surgical cause had been identified.

At this point, the blood cultures taken in ED came back positive for Group A Streptococcus (GAS) sensitive to penicillin, clindamycin and vancomycin.

She was transferred to intensive care after 12 hours in recovery due to ongoing oxygen requirement, tachycardia and high pain requirements. Concern arose over the possibility of a pulmonary embolus due to the oxygen requirement and postpartum status, but a computed tomography pulmonary angiogram did not confirm this. It revealed increasing ascites and bilateral pleural effusions. She continued to describe severe diffuse abdominal pain.

The working diagnosis then became GAS spontaneous bacterial peritonitis.

A differential of uterine perforation and myometritis was raised, however, was not supported by a further ultrasound with no suspicious features and there had been no concerns over perforation at EUA.

On day three, another CT abdomen and pelvis was performed to investigate for any other surgical cause for ongoing pain and peritonitis. With the consideration of surgical washout due to ongoing pain.

Her white cell count and neutrophils remained normal the entire stay, however, her CRP was significantly raised at 324.

She received intravenous clindamycin and cefuroxime during her eight-day stay and was discharged on oral co-amoxiclav.

Discussion

Puerperal sepsis is a major cause of morbidity and mortality for women, accounting for 11 per cent of maternal deaths in Australia between 2008–2012.¹ GAS is a life-threatening cause of puerperal sepsis, accounting for 50 per cent of deaths from sepsis in New Zealand between 2006–2013.¹ New Zealand has higher rates compared to other countries and the rates have been increasing since 2002.² The onset of GAS can be insidious and progress rapidly as demonstrated in this case; therefore, early recognition and appropriate management is essential.

There are no reported cases in the literature of postpartum GAS spontaneous bacterial peritonitis. It is a rare and life-threatening infection, one that can be difficult to diagnose. It is different to other forms of primary peritonitis in that it affects mainly young healthy people; therefore, it is assumed secondary peritonitis and a surgical cause looked for.³

A previous PubMed literature search found 26 publications of case reports with 35 cases of spontaneous bacterial peritonitis, none of which were postpartum: 29 patients had CT scans showing free fluid and oedema but no surgical cause; 34 cases proceeded to laparotomy despite a negative CT scan.³

The question is raised as to whether a surgical procedure in this circumstance would help or hinder. Due to the rare occurrence of GAS peritonitis it is not known whether a procedure and washout is of benefit to the patient or not. In a previous review of three cases, one patient died and two patients remained in hospital for nearly 60 days with multiple laparotomies. Could this case report be an example of avoidance of surgery being beneficial to the patient?³

There is ongoing research into the development of a vaccine for GAS, which would hope to reduce one of the leading causes of maternal death having a significant impact on women’s healthcare.⁴

Research is currently in progress looking into rapid antigen testing to aid in the early diagnosis of invasive GAS, which is widely used in primary care for pharyngitis diagnosis.⁵ In this case, a positive rapid antigen test giving a much quicker result than blood cultures may have streamlined her management, reducing postpartum radiation exposure with multiple CT scans.

Conclusion

Sepsis was not initially suspected in this case as she was not pyrexial and had a normal white cell count; however, antibiotics were given rapidly in the ED as it was thought likely she had retained tissue. It is well documented that women with GAS sepsis can present with acute onset severe chest, abdominal or even limb pain.³ There has not been a reported case of spontaneous bacterial peritonitis in the postpartum period and this case demonstrates another atypical presentation of invasive GAS disease

1. Bowyer L, Robinson HL, Barrett H, et al. SOMANZ guidelines for the investigation and management sepsis in pregnancy. ANZJOG. 2017;57:540-51.
2. Palaniappan N, Menezes M, Willson P. Group A streptococcal puerperal sepsis: management and prevention. The Obstetrician & Gynaecologist. 2012;14:9-16
3. Malota M, Felbinger TW, Ruppert R, Nüssler NC. Group A Streptococci: A rare and often misdiagnosed cause of spontaneous bacterial peritonitis in adults. Int J Surg Case Rep. 2014;6C:251-5.
4. Invasive Group A Streptococcal disease in New Zealand, 2016. Available from: https://surv.esr.cri.nz/PDF_surveillance/InvasiveGAS/InvGAS2016infographic.pdf.
5. Gazzano V, Berger A, Benito Y, et al. Reassessment of the Role of Rapid Antigen Detection Tests in Diagnosis of Invasive Group A Streptococcal Infections. Journal of Clinical Microbiology. 2016;54(4):994-9.