Candida albicans ventriculoperitoneal shunt infection in an infant: A case report and review of the literature

Abstract

Ventriculoperitoneal shunt malfunction is commonly seen with infection. Fungal cases are rare. We report the case of an infant with a ventriculoperitoneal shunt infection who presented with accelerated head growth. Cerebrospinal fluid cultures demonstrated Candida albicans. She was successfully treated with device removal and 3 weeks of intravenous antifungal therapy. This case highlights challenges with the diagnosis and treatment of fungal shunt infections.

Case Presentation

A 6-month-old female infant with a history of obstructive hydrocephalus requiring a ventriculoperitoneal shunt (VPS) was admitted with accelerated head growth over the preceding 5 weeks. Her parents reported no fever, vomiting, or change in baseline behaviour. She had been born at 27 weeks gestation, and her course in the neonatal intensive care unit (NICU) was complicated by sepsis and presumed meningitis secondary to non-typeable Haemophilus influenzae at birth, which was treated with ampicillin. While on therapy, she developed Bacillus cereus sepsis that was treated with vancomycin. Shortly after completing her course of vancomycin, she developed culture-negative meningitis. Cerebrospinal fluid parameters showed pleocytosis, high protein, and low glucose, so she received 3 weeks of empiric cefotaxime and vancomycin and 2 weeks of rifampin. Following her second infection, she also developed grade IV intraventricular hemorrhage and obstructive hydrocephalus, necessitating a VPS around 2 months of age, close to the end of her third course of antibiotics. Both the proximal and distal shunt sites healed well post-operatively. She had no other significant medical issues, such as necrotizing enterocolitis or immunosuppression, during her admission. Following her shunt placement she had a stable course in the NICU and was discharged home at 3 months of age, after which she had been well with no clinical evidence of infection.

At 6 months of age, she was admitted with clinical and radiographic signs of hydrocephalus. Despite non-invasive adjustment to her Strata valve, her hydrocephalus did not improve, so a VPS revision performed. The intraoperative report described proteinaceous material around the shunt tip and valve reservoir (Figure 1). Intraoperative cerebrospinal spinal fluid (CSF) analysis showed a total nucleated cell count (TNCC) of 261 × 10⁶/L (0–7 × 10⁶/L), mainly neutrophils and lymphocytes, and an elevated CSF protein of 13.63 g/L (01.6–0.25 g/L). The CSF glucose was initially unavailable, but a subsequent level was 1.4 mmol/L (2.3–4.7 mmol/L). These unexpected findings prompted her team to initiate broad spectrum antibiotics for a bacterial VPS infection. Two days after her shunt replacement, microscopic examination of the distal shunt tip showed fungal hyphae (Figure 2 and Figure 3). Around the same time, the CSF fungal cultures grew Candida albicans. Her shunt was promptly removed and temporized with an external ventricular drain (EVD). Broad spectrum antibiotics were switched to amphotericin B. Cerebrospinal fluid sterility was documented 2 days after the removal of her second shunt. Subsequent serial CSF analysis from her EVD showed improvement in her CSF parameters. Her parameters on day 14 showed a TNCC of 7 × 10⁶/L, protein of 1.8 g/L, and glucose of 1.3 mmol/L. Her antifungal therapy was stepped down to intravenous fluconazole on day 12. A new shunt was placed on day 15 of antifungal therapy, and the patient completed 3 weeks of antifungal therapy. Her treatment was complicated by arterial hypertension that resolved after amphotericin B was discontinued. She has remained well after discharged with no evidence of accelerated head growth.

Figure 1:

Photograph showing distal catheter tip with slit valve occluded by reddish tissue

Figure 2:

Photomicrograph with Grocott–Gomori methenamine silver stain (GMS) showing branched, septated fungal hyphae in the tissue (original magnification x600)

Figure 3:

Photomicrograph of hematoxylin and eosin stain showing the dense collection of leukocytes in tissue removed from catheter (original magnification x100). Higher magnification (not shown) demonstrated a mix of neutrophils, lymphocytes, and fewer eosinophils.

Discussion

Ventricular shunt insertion is a neurosurgical intervention commonly performed on children with hydrocephalus. This procedure allows CSF to drain to other body cavities where it is then reabsorbed. While shunt placements are an effective intervention, they are frequently complicated by malfunction secondary to infection and obstruction.

The majority of shunt infections are caused by skin organisms such as coagulase-negative Staphylococcus and Staphylococcus aureus. Fungal shunt infections are rare (1). Most commonly, infections are introduced during a surgical procedure or manipulation of an external drain (1). Fungal central nervous system (CNS) infections can also be a result of hematogenous dissemination, especially in neonates with an underdeveloped blood-brain barrier that is easily breached (1). Candida is the most common fungus implicated in shunt infections (1). Most infections are caused by C. albicans; however, infections from C. parapsilosis, C. glabrata, and C. tropicalis have been reported as well (1,2). The majority of Candida shunt infections occur in the younger pediatric population. A review by Bhatti et al identified 55 cases of Candida ventriculoperitoneal shunt infections reported over the span of 60 years (3). Almost half of those cases were in children under the age of 2 years. Data from our centre showed 10 pediatric shunt infections from October 2012 to May 2018. Only two of those infections were proven fungal infections, both in infants and both secondary to C. albicans.

Although Candida is a common commensal skin organism, it can cause invasive disease, especially in those with risk factors such as altered immune defences, such as premature infants, patients with diabetes mellitus, and those with neutropenia (1). Other risk factors associated with Candidal shunt infections include the use of broad spectrum antibiotics, prior or concurrent bacterial meningitis, feeding by total parenteral nutrition, long term steroid use, and prior abdominal surgery (1–3). Our patient’s prolonged hospital stay and history of neurosurgical procedures following two episodes of bacterial meningitis were likely contributing factors to fungal colonization of her shunt leading to infection.

The diagnosis of VPS infections can be challenging in the pediatric population. Similar to infections of a bacterial etiology, fungal shunt infections can present with classic meningitic symptoms such as fever, vomiting, and altered level of consciousness. However, cases of patients with mild and non-specific symptoms have been described (1). Kumar et al even described an asymptomatic child with Aspergillus growth within the ventriculoperitoneal shunt tube (4). Investigations include imaging of the brain and analysis of cerebrospinal fluid. In fungal infections, the CSF analysis often shows low glucose, high protein, and pleocytosis, making it indistinguishable from a bacterial cause (1). If a device is removed, it should always be sent for microbiologic and histopathologic examinations to maximize the yield of diagnosis, as exemplified in our case. Neuroimaging is not diagnostic but may show findings that support an infection.

Treatment of an infected shunt with antifungals alone is challenging due to the development of microbial biofilm, which allows organisms to adhere to a surface and confer protection from antimicrobials (1). For this reason, it is crucial to remove the infected device as soon as possible. If the device is not removed, the infection can easily recur once medical therapy is discontinued. Limited evidence-based data are available for the optimal treatment of this rare infection. Both intravenous and intrathecal antifungal therapy have been described; however, there is a paucity of evidence to support the routine use of intrathecal/intraventricular antifungals (3,5). The Infectious Diseases Society of America (IDSA) recommends liposomal amphotericin B with or without oral flucytosine for the initial treatment of CNS candidiasis in patients outside the neonatal period (5). The empiric drug combination has demonstrated in vitro synergism along with excellent CNS penetration. For neonates, the benefit of adding flucytosine in this population is not established and should only be used as salvage therapy. Fluconazole is not recommended as empiric therapy for CNS infections because of its reduced activity against C. glabrata and C. krusei.

For other Candida species, fluconazole is recommended as step-down therapy, regardless of age (5). Currently, there are insufficient data to support the use of echinocandins to treat CNS infections. The optimal duration of treatment following device removal is not known. Case reports describe success with 3 to 6 weeks of antifungal treatment (1,3,6,7). Sterility of the CSF should be documented. Clinical and radiological abnormalities should be resolved before therapy is discontinued (5). There are no established recommendations for the appropriate time to reinsert a neurosurgical device. Our decision to reinsert the patient’s shunt was based on improving CSF parameters, persistent sterility, and stable clinical course.

Antifungal prophylaxis in premature neonates in the NICU varies between centres. At our centre, there is no standardized protocol and antifungal prophylaxis considered on a case by case basis in consultation with the Infectious Diseases service. The IDSA guidelines recommend that neonates with birth weights <1,000 g in nurseries with high rates (>10%) of invasive candidiasis receive fluconazole prophylaxis for 6 weeks (5). In our patient, repeated courses of antibiotics placed her at risk for fungal infections, but it is unclear if she would have benefitted from antifungal prophylaxis since her shunt infection presented 4 months post-operatively.

Conclusions

We describe an infant with a shunt infection secondary to C. albicans that clinically mimicked a shunt malfunction. Our case illustrates the benefit of routine pathological examination of a removed device and CSF analysis even if an infection is not initially suspected. Prompt removal of an infected neurosurgical device is crucial for cure of the infection. Although the optimal duration of treatment in the pediatric population is unknown, we report success with 3 weeks of intravenous antifungal therapy and device removal in our patient.

Competing Interests:

None declared.

Ethics Approval:

N/A

Informed Consent:

N/A

Registry and the Registration No. of the Study/Trial:

N/A

Funding:

No funding was received for this article.

Disclosures:

The authors have nothing to disclose.

Peer Review:

This article has been peer reviewed.

Animal Studies:

N/A