Expanding Epidemiology: A Case of Pulmonary Disease Caused by Acrophialophora angustiphialis

Abstract

Invasive fungal infections have emerged as a major cause of concern, especially in immunocompromised patients. The number of opportunistic filamentous fungi documented as etiologic agents of invasive diseases continues to escalate, with rare molds being increasingly recognized as occasional pathogens [1].

The genus Acrophialophora comprises thermotolerant soil fungi that grow well at temperatures of 45°C or higher [2]. Twenty-eight species of this genus have been identified so far (https://www.indexfungorum.org/names/Names.asp), of which Acrophialophora fusispora and Acrophialophora levis are the 2 well-known opportunistic human pathogens, responsible for cases of keratitis, pulmonary colonization and infection, and brain abscess. The rarity of the infection limits our understanding of its pathogenesis, clinical manifestations, and treatment [1]. With the increasing utilization of molecular identification techniques, newer pathogenic species are coming into the limelight. Here we discuss a case of pulmonary disease caused by Acrophialophora angustiphialis, a species that was earlier isolated only from soil. To the best of our knowledge, this is the first reported human infection caused by A angustiphialis.

CASE REPORT

A 61-year-old woman, with known hypertension for the past 10 years, presented to the emergency department in January 2023 with complaints of progressive shortness of breath and cough with expectoration for 2 months, which had worsened over the past 10 days. The patient recalled 1 episode of hemoptysis 1 month prior and an on-and-off fever. Additionally, the patient also complained of progressive jaundice for 2 weeks. On admission, altered parameters included an oxygen saturation of 80% on room air, total leukocyte count of 11 730 cells/μL (reference range, 4000–11 000 cells/μL), and elevated bilirubin and liver enzymes (total bilirubin: 9.5 mg/dL [reference range, 0.1–1.2 mg/dL]; serum glutamic-oxaloacetic transaminase: 447 IU/L [reference range, 8–45 IU/L]; serum glutamic-pyruvic transaminase: 557 IU/L [reference range, 7–56 IU/L]; and alkaline phosphatase: 684 IU/L [reference range, 44–147 IU/L]). Contrast-enhanced computed tomography of the chest revealed bilateral cystic bronchiectasis with multiple cavitary nodules, suggesting an infective etiology. Ultrasonography of the abdomen revealed extrahepatic biliary obstruction, likely carcinomatous in etiology. Initial sputum cultures failed to identify the causative agent. The patient was started on injection piperacillin-tazobactam empirically and percutaneous transhepatic biliary drainage was done to relieve the biliary obstruction. By day 6, given worsening respiratory function, the patient had to be intubated and put on invasive ventilation. Ten percent potassium hydroxide–calcofluor white staining (KOH-CFW) of the mini-bronchoalveolar lavage (mini-BAL) sample failed to show any mycelia. Serum and mini-BAL galactomannan (measured by Platelia Aspergillus Ag assay, Bio-Rad Laboratories, Hercules, California) indices were 0.473 and 5.11, respectively. Owing to this, the patient was started on isavuconazole; voriconazole was not preferred due to elevated liver enzymes. By day 9, the patient went into septic shock with procalcitonin level of 4.6 μg/L and β-D-glucan (BDG) level (measured by FungiXpert Fungus (1,3)-β-D-glucan Detection Kit, Genobio Pharmaceutical, Tianjin, China) of 0.0589 ng/mL. The anti-infectives were upgraded to meropenem, levofloxacin, and caspofungin. A mini-BAL sample collected on day 12 again failed to show any mycelia on direct microscopy, but it grew fungal elements subsequently on culture as discussed below. By day 15, the patient developed refractory hypokalemia and died from the illness.

Postmortem lung biopsy samples showed septate hyphae on KOH-CFW. Two days later, the mini-BAL sample collected on day 12 showed growth of a filamentous mold on Sabouraud dextrose agar (SDA) that was flat, velvety in texture, and initially white in color but soon became brownish-gray with a brownish-black reverse. The postmortem lung biopsy sample showed a similar growth. Lactophenol cotton blue mount (LPCB) of the growth demonstrated pigmented septate hyphae but failed to yield conclusive speciation. Microculture from the growth was performed and subsequent LPCB showed abundant production of flask-shaped phialides, swollen at the base and tapering abruptly to a narrow neck, arising terminally and laterally on vegetative hyphae. Conidia were subhyaline to brown, ellipsoid to fusiform, smooth-walled, and arranged in chains (Figure 1).

Figure 1.

A, Colony morphology of Acrophialophora angustiphialis cultured from the lung biopsy sample of a 61-year-old patient in India, on Sabouraud dextrose agar after 9 (left) and 12 (right) days of incubation at 25°C, showing filamentous, brownish-gray mold with a velvety surface. B–D, Microscopic morphology of A angustiphialis on lactophenol cotton blue mount, showing pigmented, septate hyphae (blue arrow) with abundant production of flask-shaped phialides, swollen at the base and tapering abruptly to a narrow neck, arising terminally (black arrow) and laterally (red arrow) on vegetative hyphae. Conidia are subhyaline to brown, ellipsoid to fusiform, smooth-walled, and arranged in chains. Scale bar = 10 µm.

Open in new tabDownload slide

For further identification, growth on the SDA tube was homogenized using liquid nitrogen and DNA was extracted using QIAmp DNA Mini Kit (Qiagen, Hilden, Germany) following manufacturer recommendations. Conventional polymerase chain reaction targeting the internal transcribed spacer (ITS) region of the 18S ribosomal DNA was run on the extracted nucleic acid. Primers used were ITS1 (5′ TCCGTAGGTGAACCTGCGG 3′) and ITS4 (5′ TCCTCCGCTTATTGATATGC 3′). The amplicons were purified and subjected to Sanger sequencing. Basic Local Alignment Search Tool (BLAST) search of the sequenced region on GenBank database (https://www.ncbi.nlm.nih.gov/) gave 100% identity with A angustiphialis (GenBank accession number for our isolate is PQ203343). Furthermore, the phylogenetic relatedness of our isolate with selected global isolates of Acrophialophora spp was derived using MEGA software version 11.0.13 (https://www.megasoftware.net/) (Figure 2).

Figure 2.

Phylogenetic relatedness of Acrophialophora angustiphialis cultured from the lung biopsy sample of a 61-year-old patient in India (bold) with selected global isolates of Acrophialophora spp. The tree was derived by using the maximum-likelihood method with the Jukes-Cantor model and 1000 bootstrap replicates implemented in MEGA 11.0.13 (https://www.megasoftware.net).

Open in new tabDownload slide

Susceptibility to antifungal drugs was tested by broth microdilution method following Clinical and Laboratory Standards Institute (CLSI) standard M38-A2. The drug concentration ranges over which the testing was performed were as follows: amphotericin B, 0.12–8 µg/mL; voriconazole, 0.008–8 µg/mL; posaconazole, 0.008–8 µg/mL; itraconazole, 0.015–16 µg/mL; isavuconazole, 0.015–16 µg/mL; caspofungin, 0.008–8 µg/mL; micafungin, 0.008–8 µg/mL; terbinafine, 0.015–16 µg/mL. Among the tested antifungals, voriconazole (minimum inhibitory concentration [MIC] of 2 μg/mL) had the best in vitro activity (Table 1).

DISCUSSION

Acrophialophora infections are rare; there are only 24 documented infections to date (Table 2), with the first reported in 1981 [3–15]. Infections have been reported in India, the United States, Spain, Saudi Arabia, Portugal, Japan, Taiwan, and China, with India accounting for 13 of the 24 documented cases. Both males and females of various age groups have been affected, ranging from an 11-year-old to an 81-year-old. Transplant recipients, particularly those who have undergone lung or kidney transplants, are at a higher risk of infection, with several cases reported in this subgroup. Additional risk factors for invasive disease include human immunodeficiency virus/AIDS, malignancy, neutropenia, and the use of immunosuppressive drugs. Ocular trauma and retained contact lenses are specific risk factors for Acrophialophora keratitis. The lungs, brain, and eyes are the most commonly affected organs. In our case, the malignancy causing extrahepatic biliary obstruction likely contributed as a potential risk factor.

Of the documented infections so far, 18 were caused by A fusispora, 4 by A levis, and 2 by unspecified Acrophialophora spp. It must be noted that only 18 of the aforementioned 24 cases were confirmed as Acrophialophora by sequencing; others were only identified morphologically. Conversely, challenges with accurate mold identification likely have contributed to limited reports of Acrophialophora as an agent of clinically relevant disease. In the past, Acrophialophora has been misidentified as Scopulariopsis chartarum [1] and Lomentospora prolificans [6], both possibly due to the similar appearance of the colonies. Additionally, the confusion with L prolificans is because both form flask-shaped conidiogenous cells. A careful examination of the conidia might resolve this confusion, as chains of conidia are not seen with L prolificans. Additionally, conidiogenous cells of L prolificans are in a brush-like arrangement, not single on hyphae as in Acrophialophora. Scopulariopsis chartarum can be differentiated by its cylindrical annellides and thick-walled, rough and spiny conidia. Microscopically, Acrophialophora may be confused with its closely related fungus, Paecilomyces spp, due to their similar arrangement of conidia; however, Paecilomyces never produces black colonies [1].

In our case, BAL galactomannan level was elevated, whereas serum BDG level was normal. While earlier reported cases of Acrophialophora have not mentioned galactomannan or BDG levels, dematiaceous fungi have been known to cause elevated galactomannan index [16]. Further studies are required to ascertain if this is due to cross-reactivity of the antigen produced by Acrophialophora or occult coinfection with Aspergillus spp. Likewise, it could not be ascertained if the low BDG levels were a limitation of the assay or a unique characteristic of A angustiphialis. Cases reported henceforth might provide insight into this.

There is limited knowledge of human infections caused by A angustiphialis. This is in part because the phenotypic speciation of Acrophialophora poses a challenge for medical mycology laboratories. All 3 pathogenic species have similar colony morphologies. However, microscopically, there are unique identifiers that can help differentiate the various species (Table 1 and Supplementary Figure 1) [2]. DNA sequencing targeting the ITS1 and ITS2 regions is the reference method for accurate identification, and all suspected clinical isolates should undergo molecular confirmation to avoid misdiagnosis [1].

A comparison of in vitro antifungal susceptibility testing among the 3 species reveals limited activity of echinocandins and amphotericin B. Azoles generally demonstrate good in vitro activity against the other 2 species, with our isolate showing the lowest MIC for voriconazole (Table 1). However, the MIC for voriconazole was still high (2 µg/mL). There are no established clinical breakpoints for dematiaceous fungi due to the unclear clinical relevance of testing this group of pathogens. Based on the working breakpoints outlined in CLSI M38-A2, voriconazole can be interpreted as intermediate, while the other drugs can be considered resistant. It is uncertain whether the patient would have responded to voriconazole, as there are no clinical data to support these breakpoints.

At present, there are no specific treatment guidelines for Acrophialophora infections, largely due to the limited number of documented cases. Clinical outcomes have varied with different antifungal treatments. Some patients have shown positive responses to voriconazole [4, 9, 13], while others have not [7, 8]. The most prudent approach may be to test the cultured strain and treat it with the antifungal drug exhibiting the strongest in vitro activity. Combination therapy, often involving amphotericin B and an azole, has been successful in certain cases [5, 8, 12]. For example, a patient with acute lymphoblastic leukemia complicated by brain and pulmonary abscesses from Acrophialophora did not show improvement after 26 days of amphotericin B treatment until itraconazole was added [5]. Treating Acrophialophora infections remains challenging, with several reported fatalities despite appropriate azole therapy [4, 7, 8, 14]. Effective management requires aggressive, long-term use of antifungal agents to address the risk of morbidity and mortality [1]. More research is needed to identify the optimal treatment strategies.

Our case highlights the importance of considering a fungal diagnosis early in the disease presentation, especially when bacterial cultures fail to yield significant growth. Nevertheless, a highly resistant isolate in our case possibly worsened the outcome, thus shedding light on the need for rapid methods of resistance determination in pathogenic fungi.

CONCLUSIONS

Acrophialophora spp are emerging fungal pathogens and, though rarely encountered in clinical specimens, they can cause invasive infections. The isolation of A angustiphialis from human samples sheds light on the expanding epidemiology of the genus and the need for molecular confirmation to avoid misdiagnosis. Further studies in this field are needed to better understand the prevalence, pathogenesis, and optimum management of Acrophialophora infections.

Supplementary Data

Supplementary materials are available at Open Forum Infectious Diseases online. Consisting of data provided by the authors to benefit the reader, the posted materials are not copyedited and are the sole responsibility of the authors, so questions or comments should be addressed to the corresponding author.

Notes

Author contributions. K. P. A.: Data curation, investigation, writing–original draft, writing–review and editing. S. Go.: Conceptualization, data curation, investigation, writing–review and editing. S. M. and S. A.: Data curation, investigation. I. X.: Conceptualization, investigation, validation, supervision. M. S. and N. K. V.: Validation, supervision. H. M. and M. P.: Investigation, software. S. Gu. and B. R.: Investigation, resources. G. S.: Conceptualization, investigation, supervision, writing–review and editing.

Patient consent. Written informed consent was obtained from the relatives of the patient for publication of this case report.

References

Author notes