Postoperative dengue haemorrhagic shock and Trichosporon fungal sepsis: a multidisciplinary rescue

Abstract

A woman in her 30s had robotic pyeloplasty done for right ureteropelvic junction obstruction. Incidentally she developed dengue viral fever starting on postoperative day 1 itself, which progressed to dengue haemorrhagic shock by 1 week, complicating pyeloplasty due to pelvicalyceal haematoma. Dengue associated shock was superimposed with subsequent gram-negative bacterial sepsis, further complicated later with Trichosporon fungal sepsis. She was managed under multidisciplinary care, involving urology, infectious disease and ICU care. Her diagnostic and difficult management issues due to these rare sequential medical issues in an otherwise usually uncomplicated postsurgical phase are discussed along with short review of literature. This case highlights the importance of early diagnosis, timely supportive care and appropriate management in such tropical infections with significant associated mortality.

Background

Dengue in postoperative period may be fatal for patients with a significant mortality rate due to progression to haemorrhagic fever and shock. It is therefore imperative to have a high degree of suspicion for diagnosis, especially in endemic areas. Similarly, Trichosporon fungal infection is also rare but potentially fatal infection.¹ We present a rare and unique case of dengue haemorrhagic shock, bacterial sepsis and Trichosporon fungal infection all occurring in the immediate postoperative period following pyeloplasty. To our knowledge, this is first such case report in surgical practice. This case also reflects the importance of collaboration from various specialties in successful management of such a critical patient.

Case presentation

A woman in her 30s presented to us with complaint of left flank pain from past 10 months. She had a history of hypothyroidism which was controlled with medications.

She was evaluated with ultrasound, CT scan, renal dynamic scan and diagnosed to have right ureteropelvic junction obstruction. She underwent right robotic Anderson-Hynes pyeloplasty. There was no intraoperative complication with blood loss less than 50 mL. She had high grade (>103°F) intermittent fever spikes on postoperative day 1 with severe upper abdominal discomfort and retching. TLC was normal initially followed by leucopenia (2.96×10³/mm³), thrombocytopenia (40×10³/mm³) and increased liver enzymes. Dengue virus NonStructural Protein 1 (NS1) antigen test on postoperative day (POD) 3 confirmed dengue fever. Adequate hydration and supportive management were provided in consultation with infectious medicine department. On POD 6, she developed haematuria, flank pain and hypotension, necessitating multiple platelet transfusions. Ultrasound abdomen showed right renal intrapelvic haematoma with no other intra-abdominal collection (figure 1). Antibiotics were upgraded for broad spectrum coverage. She improved briefly with supportive care, but fever persisted and she required admission to intensive care unit for new onset respiratory distress, lactic acidosis and hypotension on POD 9.

Investigations

Point-of-care ultrasound on POD 10 confirmed bilateral pleural effusion, normal cardiac function on echocardiography and metabolic acidosis on arterial blood gas analysis. Single time therapeutic pleural aspiration was done revealing transudative collection. Contrast enhanced CT scan (POD 14) showed right moderate hydronephrosis filled with clots and non-excreting kidney (figure 2).

Differential diagnosis

In view of raised serum procalcitonin (11.7 ng/mL) and continuing high spikes of fever (which tends to improves in a typical dengue fever), a differential diagnosis of bacterial urosepsis or hospital acquired pneumonia were considered over a background of dengue haemorrhagic shock. Given the presence of clots in pelvicalyceal system (figures 1 and 2) along with JJ stent and urethral catheter in situ, a life-saving nephrectomy was contemplated. But considering her general condition, only a percutaneous nephrostomy (PCN) was placed. Urine culture confirmed Escherichia coli and blood culture revealed Acinetobacter pittii. Antibiotics were changed from meropenem and teicoplanin to colistin, fosfomycin and nitrofurantoin based on sensitivity reports. She improved clinically requiring no circulatory or respiratory support, though her PCN output remained nil and fever spikes continued. Later on POD 25, culture from PCN tube was positive for Trichosporon asahii.

Figure 1.

Ultrasonography of abdomen showing clots and JJ stent in renal pelvis.

Figure 2.

Contrast enhanced CT scan showing clots in pelvis with poor nephrographic enhancement.

Treatment

She was started on injection voriconazole 300 mg two times per day for one day followed by 200 mg two times per day. She became afebrile after 2 days of antifungal treatment. Gradually PCN output increased to 1.5 L/day and she was finally discharged on POD 37 on oral voriconazole 200 mg two times per day.

Outcome and follow-up

Patient was afebrile and voiding well with clear urine output from right PCN tube. Voriconazole was stopped after 3 weeks and follow-up contrast enhanced CT scan (figure 3) showed clearance of clots, no hydronephrosis, mild perinephric stranding and normally excreting right kidney. JJ stent was removed after 13 weeks and PCN was clamped. It was removed at 15 weeks. The patient is doing well presently after 3 months of follow-up post PCN removal.

Figure 3.

Follow-up contrast enhanced CT scan showing PCN in situ, mild perinephric stranding, normally functioning right kidney with contrast excretion in delayed phase.

Discussion

There are reported cases pertaining to dengue infection in postoperative period, particularly postrenal transplant cases² but no case reported postpyeloplasty. Most available studies report dengue haemorrhagic shock with its antecedent morbidity, also raising the concern that dengue in a postoperative setting may present more often in its severe form due to associated postsurgical stress. Our patient suffered from viral (dengue), followed by bacterial (E. coli and Acinetobacter), followed by fungal (T. asahii) infection. Testing for COVID-19 and malarial antigen was also done given the ongoing pandemic and malarial endemicity in our part of the world.

Dengue viral fever should be suspected in all, including postoperative cases, with typical bone-break symptoms and fever pattern,³ particularly in endemic zones. A mortality of 8.9% has been reported in patients with dengue fever in renal transplant patients.² There may be a heightened need for careful observation and investigations in postoperative patients who develop dengue, particularly in early convalescent phase of day 4–7.⁴ At this time, the patients tend to improve symptomatically, fever subsides and viraemia settles, though fall in platelets occurs and coincides with setting in of antibody response and haemorrhagic shock. Persistence of fever during this phase pointed towards a superadded bacterial sepsis in our case, which was confirmed on serum procalcitonin and blood culture. Additional antibiotics were necessitated in our case, but in a background of dengue viral fever, indwelling urethral catheter and PCN they aided in further superadded fungal infection. This should always be considered as a differential in patients not responding to more than a week of antibiotic treatment.

T. asahii is an emerging superantigen and resistant to multiple available antifungal agents. In a previous case series of six cases,⁵ all patients were women with catheter induced infection and sensitive to voriconazole. This fungus in our case could be a contaminant but continuous fever which was not responding to multiple antibiotics and showed a dramatic response to voriconazole makes it less likely to be a contaminant.

In our case, aggressive management of dengue with early platelet transfusion; bacterial sepsis with inotropes, antibiotics and supportive management and; fungal infection with voriconazole saved the patient while preserving the kidney as well. This case also highlights that all efforts should be made for mosquito-control in the postoperative care wards to prevent such vector borne preventable diseases.

Learning points.

• Postoperative patients may develop dengue viral infection with heightened risk of haemorrhagic fever.

• Early detection is necessary to guide fluid therapy, platelet transfusion and organ support in such patients.

• Days 4–7 from onset of symptoms is the critical phase after dengue infection and should be carefully looked after.

• Fungal sepsis should be identified and managed aggressively.

• Non-excretion of contrast in acute infection or bleeding is not representative of true function of the kidney.

• Multi-departmental intensive collaboration is the key to success.

Ethics statements

Patient consent for publication

Consent obtained directly from patient(s).