Complex clinical management of group A Streptococcal pelvic inflammatory disease after bilateral tubal ligation in a small community hospital
- 1 College of Medicine, The University of Arizona College of Medicine Phoenix, Phoenix, Arizona, USA
- 2 Department of Family Medicine, Stanford Health Care, San Jose, California, USA
- Correspondence to Dr Emily Wolfenden; emily.wolfenden@hsc.utah.edu
Abstract
A 43-year-old woman with a history of bilateral tubal ligation and bilateral ovarian cysts presented to our hospital with progressively worsening right lower quadrant pain and abdominal distension. Her exam findings of vaginal discharge and cervical motion tenderness, in combination with her marked leucocytosis, were suggestive of pelvic inflammatory disease (PID). PCR for Chlamydia trachomatis and Neisseria gonorrhoeae was negative, however, our patient’s blood cultures grew group A Streptococcus. This exceptionally severe presentation of PID, in combination with uncommon laboratory findings, led to complex multidisciplinary clinical decision making guided by extensive literature review. Here, we present a rare case of group A Streptococcus PID after bilateral tubal ligation, and highlight the role of a family medicine primary team in the medical and surgical management of a complex case at a community hospital.
Background
Pelvic inflammatory disease (PID) is characterised by infection and inflammation of the female upper reproductive tract, including the endometrium, fallopian tubes, ovaries and pelvic peritoneum.1 Clinically, PID usually presents as lower reproductive tract inflammation and pelvic tenderness. Untreated PID can lead to scarring and fibrosis of the reproductive tract with subsequent infertility, ectopic pregnancy and pelvic pain.
PID is most often caused by Chlamydia trachomatis and Neisseria gonorrhoeae. However, an estimated 15% of cases are caused by respiratory or enteric pathogens, such as group A Streptococcus (GAS), that have colonised the vaginal canal.1
Pertinent to our case, while bilateral tubal ligation (BTL) was previously thought to be protective against the development of PID, our literature review revealed case reports detailing the diagnosis of PID after BTL.2
We present a case of GAS PID after a BTL, review the difficulties in diagnosis and treatment of rare presentations and discuss navigating multidisciplinary management in a small community hospital.
Case presentation
A 43-year-old gravida 4 para 4 with medical history of BTL and bilateral ovarian cysts presented as a transfer from a neighbouring community hospital with 8 days of dyspareunia, progressively worsening right lower quadrant abdominal pain and distension, and associated fever, nausea and vomiting.
On presentation, our patient was afebrile, tachycardic, normotensive and saturating 85% on room air. On physical examination, she had mildly decreased breath sounds at the lung bases bilaterally, a distended and diffusely tender abdomen, and white-watery discharge in the vaginal vault with cervical motion tenderness. She was dyspneic and unable to speak in complete sentences on interview.
Investigations
Laboratory findings demonstrated a leucocytosis to 36.17×109/L, haemoglobin of 125 g/L, platelets of 130× 109/L, creatinine of 2.10 mg/dL and total bilirubin of 5.0 mg/dL. Urinalysis showed >50 white blood cells/high power field, positive leucocyte esterase and positive nitrites. Pelvic ultrasound showed unremarkable ovaries with some free fluid in the pelvis but no tubo-ovarian abscess (TOA). A CT abdomen/pelvis showed bilateral trace pleural effusions with atelectasis, abdominal ascites, a 2.9 cm rim-enhancing fluid collection in the pelvis, soft tissue changes around the lower aspect of the uterus possibly secondary to PID, bladder wall thickening with air within the bladder and an appendix with borderline-normal size of 7 mm.
Differential diagnosis
On our patient’s initial presentation, our differential diagnosis was broad and included acute appendicitis, pyelonephritis, streptococcal toxic shock syndrome and PID.
She presented with the chief complaint of right lower quadrant abdominal pain, vomiting and fever. At admission, her abdomen was distended and tender on physical examination, her laboratory findings showed a leucocytosis and imaging revealed a non-specific rim-enhancing fluid collection. This prompted us to consider appendicitis in our differential diagnosis, and consult general surgery.3 However, there was no evidence of appendiceal dilation or wall thickening on CT, which allowed us to move this diagnosis further down our differential.4
We also considered pyelonephritis, given positive leucocyte esterase and nitrites on urinalysis, and bladder wall thickening with air within the bladder on imaging.5 However, no evidence of nephritis was seen on CT, which made this diagnosis less likely.6
We considered streptococcal toxic shock syndrome when laboratory findings showed an elevated creatinine, hyperbilirubinaemia and elevated liver enzymes with pleural effusions on imaging.7 However, the absence of hypotension or obvious soft tissue infection moved this diagnosis lower on our differential.
PID was high on our differential, due to the presence of white-watery discharge in the vaginal vault with cervical motion tenderness, free fluid in the pelvis, a rim-enhancing fluid collection in the pelvis and soft tissue changes around the lower aspect of the uterus.8
Treatment
General surgery recommended conservative management with antibiotics due to the absence of appendiceal pathology on CT. Obstetrics and gynaecology (OBGYN) recommended cefoxitin and doxycycline to cover suspected PID as per current Center for Disease Control and Prevention guidelines.8
On hospital day 2, our patient developed erythroderma of her nasal bridge and anterior lower extremities. Her leucocytosis improved to 23.00×109/L and her elevated creatinine normalised to 1.00 mg/dL, but she had a haemoglobin of 103 g/L, platelets of 101×109/L, persistently elevated total bilirubin to 2.9 mg/dL and progressive elevation of her alkaline phosphatase to 721 U/L. Her blood cultures were positive for GAS. Gonorrhoea, chlamydia and hepatitis testing was negative.
Although our patient was never hypotensive, in the setting of renal impairment, liver involvement, coagulopathy and erythroderma, the team had concern for streptococcal toxic shock syndrome.9 Infectious disease (ID) was consulted and had a wide differential including cystitis and PID as the source of entry for GAS. ID recommended adjusting antibiotics to ceftriaxone 2 g intravenous two times per day in addition to clindamycin to inhibit streptococcal toxin production. OBGYN again saw our patient, at which time they believed a gynaecologic source of infection was unlikely as there was no evidence of TOA on imaging and GAS is an unlikely source of PID. Pelvic drainage was not recommended. A transthoracic echocardiogram showed normal cardiac function and anatomy with no vegetations.
Over the course of hospital days 3–11, our patient’s leucocytosis gradually resolved; however, her abdominal pain and distention slowly worsened. Repeat CT scans of her abdomen and pelvis were performed on hospital days 3 and 6, which revealed progressively increasing areas of fluid collection in the abdomen and pelvis as well as a partially loculated pleural effusion. These fluid collections defied multiple efforts to address them non-surgically, including a paracentesis on day 3 and the placement of a CT-guided right abdominal drain and right chest tube on day 6. Our patient’s final CT chest/abdomen/pelvis on day 11 showed an enlarging right loculated pleural effusion and persistence of fluid components in the abdomen and pelvis concerning for abscesses despite drain placement and re-positioning. The family medicine team’s increasingly clarified belief that our patient had an unusual presentation of PID, coupled with her declining clinical status, ultimately led to a re-consultation of general surgery and OBGYN, a new consult with cardiothoracic surgery and the eventual decision to proceed to operative management.
On hospital day 17, general surgery performed a diagnostic laparoscopy with drainage of the abdominal and pelvic abscesses, and saw extensive abdominal adhesions and thick abscess material at the right pelvis and along the right colic gutter. OBGYN stated that the intraoperative appearance was consistent with fulminant PID. Cardiothoracic surgery then performed a bronchoscopy and thoracoscopy, which revealed an empyema of the right chest and laceration of the diaphragm with communication of pus from the abdomen. Empyema drainage, diaphragmatic repair and chest tube placement were performed. Due to findings during the surgery, doxycycline and ampicillin/sulbactam were initiated for management of PID, and our patient was transferred to the intensive care unit.
Outcome and follow-up
Our patient exhibited both clinical and symptomatic improvement postoperatively. Her right chest tubes were removed on postoperative day 4 with follow-up chest X-ray showing a stable loculated density in the right hemithorax. She was discharged from the hospital on hospital day 24, postoperative day 8 on a 14-day course of cephalexin and with instructions to follow-up with cardiothoracic surgery for monitoring of her right empyema. She decided to transfer her primary care to our outpatient family medicine resident training clinic.
Discussion
PID should be suspected in a sexually active woman presenting with lower abdominal pain and pelvic discomfort, both of which our patient exhibited on presentation. The typical bacterial aetiologies of PID include Neisseria gonorrhoea and C. trachomatis; however, up to 15% of cases of PID may be due to respiratory pathogens including GAS.1 Given our patient’s GAS positive blood cultures, we suspect this was the aetiology of this patient’s extensive disease course. GAS is known to colonise the vaginal canal, and previously published case reports detailing GAS peritonitis and GAS PID highlight that ascending vaginal flora are thought to contribute to the increased presentation of this disease among women.10–12 Alternate methods of GAS transmission that have been implicated in the literature include recent pelvic surgical instrumentation, childbirth, recent skin infection or oropharynx infection, or intrauterine device placement, none of which were present in our case.13 Our patient was ultimately diagnosed with PID by OBGYN via laparoscopic visualisation of the abdomen, which revealed pus in the peritoneal sac. This laparoscopic finding, in combination with our patient’s symptoms of possible endometritis including dyspareunia and cervical motion tenderness, has been found to have 81% sensitivity and 100% specificity for the diagnosis of PID.14
In addition to the unique bacterium suspected to cause PID in our report, our patient’s BTL 12 years prior to presentation also complicates our theory that GAS bacteria ascended into the abdominal cavity from the vaginal canal, as theoretically there would no longer be a communication between the two spaces. However, case reports do exist detailing instances of PID after tubal ligation.2 The pathophysiology of PID after BTL is thought to involve one of three processes: persistence of a passage between proximal and distal segments of the fallopian tube, iatrogenic infection caused by the introduction of bacteria during the surgical procedure and infection via haematogenous or lymphatic spread.15 The persistence of a passage between segments of the tube may be due to recanalisation, fistula formation or faulty ligation. As our patient’s BTL occurred 12 years prior to presentation and she did not exhibit an alternate source of infection, we believe her PID was most likely due to ascending infection via a recanalised fallopian tube or tubal fistula.
Our location within a small community hospital with many available specialists made this complicated case interesting for our primary team to navigate. Lack of an obvious nidus of infection in the abdomen or pelvis on presentation led to ambiguity regarding which surgical service would serve as the main consultants on this case, and resulted in our family medicine team becoming the primary drivers of our patient’s care. This lack of obvious infectious aetiology, in combination with the unique bacterium cultured in our patient’s blood, created clinical uncertainty of our patient’s suspected diagnosis and most appropriate treatment. It also led to conflicting recommendations of whether surgical management was indicated throughout our patient’s hospitalisation.
Guided by extensive literature review, our family medicine team was able to advocate for our patient’s care through the re-consultation of specialists, and ultimately played an extensive role in her medical and surgical management. After initial recommendations from consultants, it became our responsibility to determine when specialties needed to be re-consulted and when to consider repeat imaging. We used the persistence of clinical symptoms to guide much of our clinical management. The appearance of new respiratory symptoms late in our patient’s disease course prompted our consultation of cardiothoracic surgery, who ultimately coordinated with general surgery to complete surgical intervention together, with OBGYN on standby for the case. Interventional radiology also encouraged surgical teams to consider surgical intervention in the setting of sub-optimal drain output. Family medicine became the epicentre of this case, carefully and continuously evaluating the patient, driving decisions with each change and ultimately finding clinical resolve for our patient.
Here, we present a case of GAS PID after BTL. While cases of GAS PID7 and PID post BTL2 have been previously published in the literature, our case is unique in that our patient exhibited both of these rare disease presentations. This rare presentation at a small community hospital required family medicine to be at the forefront of clinical decision making, ultimately resulting in surgical intervention and clinical resolve for our patient with the help of a large multidisciplinary treatment team.
Learning points
-
In the absence of Neisseria or Chlamydia, group A Streptococcus (GAS) should be considered as the possible aetiology of suspected pelvic inflammatory disease (PID).
-
GAS is known to colonise the vaginal canal and can cause GAS peritonitis complicating a PID clinical course.
-
GAS PID may ascend into the abdomen despite bilateral tubal ligation.
-
At a small community hospital, family medicine can lead clinical decision making for a large, multidisciplinary team in a complex case, advocating for and achieving clinical resolve for a patient. Surgical subspecialist leadership is appropriate for surgical cases such as this when available.
Footnotes
-
Contributors EW helped in drafting and editing entire contents and obtaining patient consent. MM helped in drafting discussion and editing entire contents. RS helped in drafting learning points and editing entire contents.
-
Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.
-
Competing interests None declared.
-
Patient consent for publication Obtained.
-
Provenance and peer review Not commissioned; externally peer reviewed.
- © BMJ Publishing Group Limited 2020. No commercial re-use. See rights and permissions. Published by BMJ.
References
Use of this content is subject to our disclaimer