Ureaplasma urealyticum infection presenting as pyelonephritis and perinephric abscess in an immunocompromised patient

Abstract

We report a case of an immunosuppressed 67-year-old woman who presented with fever of unknown origin. Further investigation revealed multiple left renal and perinephric abscesses. These were managed with percutaneous drainage and broad-spectrum antibiotics; however, no clinical improvement resulted. No organism was identifiable on standard microscopy and culture of all drain, blood and urine samples taken. Left nephrectomy with right urinary diversion was performed for sepsis management and to protect the patient’s right kidney. Eventually, Ureaplasma species’ RNA was detected in the patient’s drain fluid using PCR and 16S ribosomal RNA gene sequencing. The patient was treated successfully with targeted antibiotic therapy and underwent extensive rehabilitation following this. Histopathology of the nephrectomy specimen suggested xanthogranulomatous pyelonephritis.

Background

Ureaplasma urealyticum is a commensal bacterium of the genitourinary tract and is usually associated with localised urethritis, vaginosis or perinatal infections.^{1 2} Renal or perinephric abscesses are a rare presentation of this pathogen. Common to all reported cases is the patient immunosuppression and a significant delay to diagnosis.³ The incidence of immunosuppressed patients is rising throughout the world; thus, our case report is relevant to all clinicians as it highlights: first, the importance of maintaining a broad pathogen differential in this vulnerable population and second, the invaluable role 16S ribosomal RNA (rRNA) PCR can play in clarifying diagnostic dilemmas by identifying atypical infective organisms.

Case presentation

A 67-year-old woman with indolent B-cell lymphoma was investigated over an 18-month period in an outpatient setting for dysuria, macroscopic haematuria and vaginal symptoms. CT intravenous pyelogram performed as part of the work up of symptoms found an inflamed thickened bladder but no other pathology. Cystoscopy identified only ‘light inflammation’ but no evidence of malignancy. The provisional diagnosis for her condition was paraneoplastic pemphigoid of the bladder. Paraneoplastic pemphigoid refers to an autoimmune-mediated blistering of mucous membranes in the context of malignancy. This diagnosis had not been confirmed histologically or immunologically. The patient had ceased her rituximab maintenance 4 months prior.

With the provisional diagnosis of paraneoplastic pemphigoid of the bladder, she was started on a course of dapsone and steroids. Her treatment, however, was complicated by lethargy, worsening urinary symptoms, fever and rigours. After 4 weeks of these symptoms, she was admitted acutely under a haematology team. During her admission, fever continued despite treatment with broad-spectrum piperacillin/tazobactam antibiotics.

Investigations

Inflammatory markers such as C-reactive protein remained elevated despite broad spectrum antibiotics (270 mg/L). Heinz bodies were present on blood film suggestive of ongoing haemolysis. Multiple blood cultures had been performed and aside from one contaminated sample, all cultures were negative. Urine cultures revealed significant leucocytosis (>500 /hpf) and haematuria (>500 /hpf) but likewise cultured no organism. Investigations for atypical infective organisms, including Q fever, HIV, cryptococcus and cytomegalovirus, were negative.

A positron emission tomography (PET) scan using fludeoxyglucose (FDG) tracer was performed to rule out an occult focus of infection and revealed multiple enhancing mass-like lesions within the left kidney (figure 1).

Figure 1.

Positron emission tomography (PET) CT using fludeoxyglucose (FDG) performed as part of patient investigative work up. (A) CT images reveal multiple low density nodular lesions within the left kidney. The largest is an exophytic mass arising from the lateral aspect of the lower pole of the left kidney measuring 33×23 mm. (B) PET scan reveals that these nodular lesions in the left kidney have increased FDG uptake. Abdominal and pelvic lymph nodes are non-avid.

Differential diagnosis

Two differential diagnoses were considered. The first was the patient having multiple perinephric abscesses. The other differential was a recurrence of the patient’s B-cell lymphoma. Nevertheless, given the findings of multiple mass-like lesions on PET CT scan and the patient’s history of urinary symptoms preceding the illness, renal abscesses were deemed the most likely diagnosis.

Treatment

The patient’s perinephric abscesses were percutaneously drained under CT guidance. Three drains were inserted (figure 2) and a large quantity of purulent fluid was drained. Bacterial, fungal and mycobacterial microscopy and culture of drain fluid did not identify any pathogen. The patient’s antibiotic regimen was changed to meropenem and vancomycin to provide broader antimicrobial cover and drain fluid samples were sent for bacterial 16S rRNA PCR.

Figure 2.

CT images demonstrating percutaneous drain placement for the patient’s perinephric abscesses.

Unfortunately, the patient’s febrile illness was refractory to treatment and all drains produced significant purulent discharge despite frequent flushing and the patient’s overall clinical status continued to decline secondary to sepsis. The patient’s case was discussed at a multi-disciplinary meeting and a consensus decision was made to perform a left radical nephrectomy, diversion of right ureter and urostomy procedure to provide sepsis control and protect the patient’s right kidney from communicating infection. The patient was afebrile for 36 hours postoperatively before the preoperative pattern of fever returned.

16S rRNA PCR of the patient’s fluid eventually returned Ureaplasma species. The nephrectomy specimen returned the same organism when cultured on a Ureaplasma-specific culture media. Furthermore, the histopathological examination of the total nephrectomy demonstrated xanthogranulomatous pyelonephritis indicative of recurrent urinary tract infection (figure 3). Put together, this confirmed that the patient’s febrile illness had been driven by recurrent pyelonephritis and abscess formation by U. urealyticum.

Figure 3.

Histopathology of the left nephrectomy specimen. (A) Intrarenal microabscess cavity (H&E, 40×). (B) Aggregates of foamy histiocytes to the left with neutrophils and karyorrhectic debris to the right. The cells represent acute on chronic inflammation that is characteristic of xanthogranulomatous pyelonephritis (H&E, 400×).

The patient was treated initially with a combination therapy of moxifloxacin and doxycycline. This was later changed to doxycycline monotherapy due to patient intolerance to moxifloxacin with intractable nausea and vomiting, and monotherapy was continued without further adverse effects.

Outcome and follow-up

The patient made complete recovery; however, she required significant inpatient rehabilitation. The patient is now 4 months since completion of treatment. The lower urinary tract and vaginal symptoms she had in the months prior to her admission have also resolved. This suggests that her original symptoms were likely secondary to U. urealyticum urethritis and vaginosis as opposed to paraneoplastic pemphigoid of the bladder. She continues to be followed up by her urology and haematology teams.

Discussion

Ureaplasma belong to the family of Mycoplasmataceae bacterium which is characterised by a lack of a peptidoglycan cell wall.² As demonstrated in our case, diagnosis of Ureaplasma is often challenging due to its inability to be visualised on Gram stain due to the absence of bacterial cell wall, along with the requirement of specialised culture media and resistance to most broad-spectrum antibiotics which act on bacterial cell walls.² Consequently, only antibiotics which have alternative mechanisms of action such as tetracyclines and quinolones are effective; in the literature, doxycycline, clindamycin and levofloxacin have been successfully used.³

Upper urinary tract infections secondary to U. urealyticum are rare. In the literature, there are only two cases of U. urealyticum presenting with renal or perinephric abscess.^{3 4} Eilers et al reported a case of a 19-year-old woman, a renal transplant patient, who developed multiple renal abscesses within her transplanted kidney.¹ This occurred on the background of post-transplant lymphoproliferative disease treated with sirolimus and rituximab. Following identification of the organism, the patient was treated with levofloxacin and rapidly recovered.¹ Yazdani and Thupil reported a case of Ureaplasma perinephric abscess in a patient with mantle cell lymphoma on rituximab. No mention was made of patient’s treatment outcomes.⁴ Both cases were like our own in that the patient had fever refractory to broad-spectrum antibiotics, significant delay to diagnosis occurred and all blood, urine and abscess samples were ‘sterile’ using conventional culture methods. Indeed, in all cases, the eventual diagnosis of Ureaplasma was made using molecular techniques rather than microscopy and culture. It is of interest that both cases of renal abscesses secondary to U. urealyticum occurred in the context of lymphoma. While cases are too few to confirm a pathophysiological mechanism, it has been noted by previous authors that there appears to be an association between systemic complications of Mycoplasmataceae bacteria and impaired humoral immunity.³

Two salient lessons arise from this case. First, our case emphasises the importance of vigilance for atypical organisms as a cause of pathology in immunocompromised patients. Our patient’s history and investigations were suggestive of a long-standing urethritis likely secondary to Ureaplasma, which unfortunately progressed to upper urinary tract infection and renal abscess. While identifying the causative pathogen may not have affected the eventual decision to proceed to nephrectomy in the context of the patient’s sepsis, early identification of this atypical organism may have prevented progression of her infection to renal abscesses in the first place. This could have been done via simply sending a urine specimen early in her clinical course for 16S PCR for atypical organisms, including U. urealyticum. Precedent exists in the literature; other centres have reported routinely investigating the urine samples of high risk immunosuppressed patients with molecular techniques such as 16S PCR.³

Second, our case is an important reminder of the limitations of conventional microscopy and culture techniques, and the importance of clinicians recognising the adjunct role molecular techniques, such as 16S PCR, can have in clarifying diagnosis where other ‘conventional’ diagnostic methods have only produced negative results.⁵ 16S PCR differs from traditional molecular techniques such as quantitative PCR (qPCR) in the breadth of bacterial specimens able to be identified. qPCR involves sequencing for a specific bacterial species. In contrast, 16S PCR allows for broad-range amplification and identification of any bacterial DNA within a microbiological sample. While this allows for identification of organisms not initially considered in the differential diagnosis such as in the case presented, the test is vulnerable to contamination and is of reduced utility in polymicrobial infection.⁵ Furthermore, the negative predictive value of 16S molecular testing is poor. However, as highlighted in our case, the appropriate use of this investigation played a critical role in establishing the aetiological pathogen and altered therapy to allow cure of the infection.

Learning points.

• Ureaplasma urealyticum is normally a commensal bacterium of the genitourinary tract; however, in immunosuppressed patients, it can produce significant pathology.

• Molecular techniques such as 16S ribosomal RNA PCR are an important adjunct in identifying atypical infections, where conventional microscopy and culture techniques produce negative results.

• In patients with genitourinary symptoms with no other identifiable pathology, infective causes such as infection with U. urealyticum should be considered.