Cardiac Devices Emerge as the Strongest Signal in Metastatic Candidozyma auris Candidemia
Metastatic spread after Candidozyma auris candidemia was not rare in a large tertiary-hospital cohort from Spain. Among 108 adult patients treated between 2016 and 2024, 24 developed metastatic infection, a burden that was concentrated in patients with intracardiac hardware and prosthetic valves.
The pattern matters because C. auris is already a pathogen that strains infection-control systems and antimicrobial stewardship programs. Once it enters the bloodstream, the clinical question is no longer limited to clearing fungemia. The organism can seed endocardial surfaces, bone, the central nervous system, and hepatosplenic tissue, turning a bloodstream infection into a multi-organ problem with greater mortality risk.
Endocarditis Led the Spectrum of Dissemination
Endocarditis was the most frequent metastatic complication, appearing in 8 patients, or 7.4% of the full cohort. Osteoarticular infection, meningitis, and hepatosplenic infection each occurred in 4 patients, or 3.7%, while chorioretinitis appeared in 1 patient. Three additional patients developed later complications, including aneurysm sac infection and subcutaneous abscess.
These are not incidental complications. They reflect a pathogen capable of adhering to foreign material and persisting in protected anatomic niches, where bloodstream clearance does not guarantee sterilization of seeded tissue. In practice, that means the initial bloodstream isolate may only be the first visible sign of deeper disease.
Most patients, 84 of 108, had no evidence of dissemination beyond candidemia. Still, the 22.2% metastatic rate is high enough to shape bedside evaluation, particularly in patients whose anatomy or clinical course suggests a persistent intravascular reservoir for infection.
Intracardiac Hardware Carried the Clearest Risk Signal
A cardiac device or prosthetic valve was present far more often among patients with metastatic infection than among those without it: 54.2% versus 26.2%. In multivariable Fine–Gray analysis, that exposure was independently associated with metastatic infection, with a subdistribution hazard ratio of 2.89 and a 95% confidence interval from 1.21 to 6.91.
The association was less secure in a sensitivity analysis limited to patients who underwent echocardiography, which tempers the strength of inference without erasing the bedside implication. Even so, the direction of effect fits established clinical biology. Prosthetic surfaces and intracardiac devices create a substrate for fungal adherence, biofilm formation, and immune evasion, making metastatic infection more plausible when fungemia persists or fails to clear promptly.
A short paragraph from the clinical front line is unavoidable here: echocardiography belongs early in the workup when C. auris candidemia meets intracardiac hardware.
Persistent Fungemia Suggested Ongoing Seeding, Not a Definitive Threshold
Persistent candidemia showed a higher, though not statistically significant, risk for metastatic infection in the day-5 landmark analysis, with a subdistribution hazard ratio of 2.67 and a 95% confidence interval from 0.86 to 8.32. The wider interval leaves uncertainty, but the signal is biologically coherent. Ongoing recovery of the organism from blood after several days of therapy suggests failure of source control, inadequate antifungal activity, or a protected nidus such as a device or deep focus.
Median candidemia duration was longer in patients with metastatic infection than in those without it: 6 days versus 1 day. Older age also trended higher in the metastatic group, with medians of 68.9 and 59.7 years, respectively, although this comparison did not reach conventional statistical significance.
Early management did not clearly separate the groups. Central venous catheter removal within 48 hours, antifungal initiation within 48 hours, and resolution of the infectious focus within 48 hours were not significantly different between patients with and without metastatic infection. Initial treatment was most often echinocandin-based, and combination therapy was common in both groups. Those observations suggest that dissemination was not simply a marker of delayed recognition or obviously inferior initial management.
Mortality Rose When Metastatic Infection Was Present
Metastatic infection carried a measurable mortality penalty. In time-dependent analyses, metastatic disease was associated with increased deaths at 60 days, with an adjusted hazard ratio of 2.21 and a 95% confidence interval from 1.04 to 4.67, and at 90 days, with an adjusted hazard ratio of 2.26 and a 95% confidence interval from 1.09 to 4.67.
The time-dependent design is important because it treats metastatic infection as an exposure that can emerge after candidemia rather than assuming it exists from day 0. This approach better matches bedside reality, where the diagnosis of dissemination often follows initial bloodstream detection by days or weeks. It also reinforces the practical cost of missing seeded disease: mortality is not merely correlated with dissemination, it rises after dissemination is identified.
Forty-four patients died without metastatic complications, and 40 remained alive without documented dissemination. Death before metastatic diagnosis was handled as a competing event, which reduces the risk of inflating dissemination estimates in a population already at high risk of early death from severe illness.
Why Echocardiography Deserves a Lower Threshold
The study’s clearest clinical message is narrow but actionable: patients with C. auris candidemia and intracardiac devices need echocardiographic evaluation. The organism’s propensity for device-associated infection, together with the observed enrichment of prosthetic valves and other cardiac hardware in the metastatic group, makes structural cardiac assessment essential rather than optional.
This is especially relevant in hospitals where C. auris has become endemic and bloodstream infection can arise in both ICU and non-ICU patients. The yeast’s persistence in healthcare environments, combined with frequent colonization and multidrug resistance, means candidemia may represent only one phase of a longer infectious sequence.
Diagnostic confirmation of metastatic disease came from imaging in 18 patients, microbiology from the metastatic site in 19, and histopathology in 4, with overlap across modalities. Those findings suggest that dissemination often required more than one diagnostic lens to be recognized, which is exactly why a targeted search matters when fungemia does not resolve as expected or when prosthetic material is present.
The study does not define a universal screening protocol for every patient with C. auris candidemia. It does, however, identify a higher-risk phenotype: candidemia with cardiac devices or prosthetic valves, especially when fungemia persists. In that setting, the threshold for echocardiography should be low, and the threshold for looking beyond the bloodstream should be lower still.