Malassezia restricta as an unexpected cause of infectious osteomyelitis diagnosed by metagenomic sequencing: a case report and literature review

Background Malassezia restricta, a lipophilic and lipodependent yeast belonging to the basidiomycetes group, is an opportunistic fungal pathogen associated with various skin diseases, including seborrheic dermatitis and dandruff. Typically, Malassezia infection in neonates manifests as fungemia or hematogenous dissemination to the bone or lungs. However, vertebral osteomyelitis caused by these fungi is rarely reported owing to non-specific clinical presentations and laboratory/imaging findings. The Pathogen Metagenomics Sequencing (PMseq) technique enables direct high-throughput sequencing of infected specimens, facilitating the rapid and accurate detection of all microorganisms in clinical samples through comprehensive reports. Case presentation A 52-year-old male was admitted to our hospital on July 20, 2022 with a 3-month history of ambulatory difficulties and localized low back pain. Magnetic Resonance Imaging (MRI) examination of the spinal column revealed irregular bone destruction affecting the L2, L3, and L5 vertebral bodies. Additionally, low T1 and high T2 intensity lesions were observed at the intervertebral discs between L3 and L5. The presumptive diagnosis of tuberculous spondylitis was made based on the imaging findings, despite negative results in all mycobacterium tests. However, the patient exhibited no improvement after receiving regular anti-tuberculosis treatment for 3 months. Subsequent MRI revealed an expansive abnormal signal within the vertebral body, leading to progressive bone destruction. The absence of spinal tuberculosis or other infective microorganisms was confirmed through culture from blood and pathological tissue from the L4 vertebral body. Subsequently, PMseq was performed on the specimens, revealing M. restricta as the predominant pathogen with the highest relative abundance value. The pathological examination revealed the presence of fungal mycelium in the L4 vertebral body, with positive findings on periodic Schiff-methenamine and periodic acid-Schiff staining. The anti-tuberculosis treatment was discontinued, and an antifungal combination of fluconazole and voriconazole was administered. All symptoms were resolved after 7 consecutive months of treatment, and the patient was able to ambulate autonomously. Vertebral lesions were reduced on MRI during the 13-month follow-up. Conclusions M. restricta is not a commonly recognized pathogen associated with infectious vertebral osteomyelitis. However, PMseq can aid in diagnosis, timely treatment, and decision making for some non-specific infectious diseases.

Case presentation A 52-year-old male was admitted to our hospital on July 20, 2022 with a 3-month history of ambulatory difficulties and localized low back pain.Magnetic Resonance Imaging (MRI) examination of the spinal column revealed irregular bone destruction affecting the L2, L3, and L5 vertebral bodies.Additionally, low T1 and high T2 intensity lesions were observed at the intervertebral discs between L3 and L5.The presumptive diagnosis of tuberculous spondylitis was made based on the imaging findings, despite negative results in all mycobacterium tests.However, the patient exhibited no improvement after receiving regular anti-tuberculosis treatment for 3 months.Subsequent MRI revealed an expansive abnormal signal within the vertebral body, leading to progressive bone destruction.The absence of spinal tuberculosis or other infective microorganisms was confirmed through culture from blood and pathological tissue from the L4 vertebral body.Subsequently, PMseq was performed on the specimens, revealing M. restricta as the predominant pathogen with the highest relative abundance value.The pathological examination revealed the presence of fungal mycelium in the L4 vertebral body, with positive findings on periodic Schiff-methenamine and periodic acid-Schiff staining.The anti-tuberculosis treatment was discontinued, and an antifungal combination of fluconazole and voriconazole was administered.All symptoms were resolved after Malassezia restricta as an unexpected cause of infectious osteomyelitis diagnosed by metagenomic sequencing: a case report and literature review Guoxuan Peng 1 † , Yuan Lin 1,2 † , Qiang Zou 1 , Hongcheng Peng 1,2 , Anyi Lei 1,2 , Xu Zou 1,2 , Zhe Xu 2,3 , Hong Sun 1 , Xu Ning 1  and Mingzhi Huang 1*

Background
The etiology of spinal osteomyelitis encompasses a wide range of pathogens, including bacteria, viruses, fungi, Mycobacterium tuberculosis, and parasites [1].Although tuberculous osteomyelitis is the most prevalent spinal infection, fungal infections of the spine, particularly in immunocompromised or critically ill patients, can also be caused by Candida, Cryptococcus, Aspergillus species, and other molds [2][3][4].In the case of the other fungi, spinal involvement typically occurs owing to hematogenous or direct spread of organisms from an initial pulmonary source of infection.Moreover, involvement of the vertebral bodies can result in severe lumbar pain, impaired ambulation, vertebral compression fractures, and pronounced spinal deformity [5].Malassezia restricta is an opportunistic fungal pathogen associated with a range of dermatological conditions, including seborrheic dermatitis and dandruff [6,7].To date, occurrences of vertebral infection caused by M. restricta are exceedingly uncommon.Recently, Limon et al. [8] demonstrated that M. restricta elicits innate inflammatory responses largely linked to the presence of an inflammatory bowel diseaseassociated polymorphism through the signaling adaptor gene of CARD9 in Crohn's disease.Owing to the nonspecific clinical manifestations and imaging characteristics of fungal vertebral osteomyelitis, accurate diagnosis is typically challenging [9].Additionally, a minimum of 3 days of culture at temperatures ranging from 23 to 28 °C is typically required for Malassezia growth.It should be noted that false-negative results may occur in the culture, serum 1-3-β-D-glucan test, and serum galactomannan (GM) test [10].
Pathogen Metagenomics Sequencing (PMseq) involves direct high-throughput sequencing of infected specimens, followed by comparison with microbial-specific databases and intelligent algorithm analysis.This approach enables the identification of species information for suspected pathogenic microorganisms, facilitation of rapid and accurate detection of challenging and critical infections, and comprehensive and in-depth genomics analysis of total DNA [11,12].Moreover, PMseq demonstrates a significantly reduced turnaround time of approximately 24 h encompassing sampling, nucleic acid extraction, library sequencing, data processing, and reporting in comparison to conventional methods.Consequently, it enables prompt pathogen identification in instances of inexplicable illnesses [13].We employed the PMseq method to successfully identify M. restricta as the probable causative agent of vertebral osteomyelitis in a patient who was initially diagnosed with tuberculosis (TB) but did not respond to regular anti-TB agents after 3 months of treatment.Furthermore, the implementation of antifungal therapy has demonstrated significant improvements in patient outcomes, encompassing alleviation of low back pain, restoration of ambulatory capacity, and mitigation of vertebral lesions.Herein, we present a case of fungal vertebral osteomyelitis potentially caused by M. restricta in an immunocompetent patient.

Case presentation
A 52-year-old male presented with a 3-month history of ambulatory difficulties and lumbar pain.Magnetic Resonance Imaging (MRI) examination of the spine revealed irregular bone destruction affecting the L2, L3, and L5 vertebral bodies.Additionally, low T1 and high T2 intensity lesions were observed at the intervertebral discs between L3 and L5 3 months earlier (Fig. 1A).The primary diagnosis was considered to be spinal TB; however, the T-SPOT.TB test and acid-fast staining of blood yielded negative results, whereas computed tomography (CT) did not reveal any signs of pulmonary tuberculosis.Therefore, the patient underwent a CT-guided percutaneous biopsy of spinal lesions owing to an unidentified diagnosis based on clinical manifestations and laboratory/imaging findings.Furthermore, both culture and pathological examination of vertebral specimens failed to find any evidence of pathogen infection.Despite the absence of conclusive evidence supporting Mycobacterium.tuberculosis infection, the potential for spinal TB remained substantial based on its imaging characteristics.Therefore, with the consent of family members, the patient underwent anti-TB treatment at a local hospital.However, despite receiving oral anti-TB medications for a duration of 3 months, there was no improvement in his symptoms; on the contrary, his back pain had notably intensified over the past month.On July 20, 2022, the patient was admitted to our hospital with recurrent severe low back pain (visual analog scale (VAS) score of 9 out of 10 points) and difficulty standing for 1 week.Laboratory investigations yielded the following results: serum C-reactive protein (CRP), 8.41 mg/L (normal value: < 5 mg/L); urine kappa-light chain (K-LC), 13.5 mg/L (normal value: 0.00 − 7.13 mg/L); erythrocyte sedimentation rate (ESR), 6.0 mm/h (normal value: 0-21 mm/h); Interleukin-6 (IL-6), 6.61 pg/ml (normal value: 0 − 7 pg/ml); procalcitonin, 0.03 ng/ml (normal value:<0.15ng/l); serum 1-3-β-D-glucan,<37.5 pg/ml (normal value: 0 − 70 pg/ml); serum GM<0.008 (normal value:<0.5S/CO); serum LAMBDA-light chain (λ-LC)<3.94mg/L (normal value: 0.00 − 3.94 mg/L); HIV (negative).The first-third doses of COVID-19 vaccine were administered in August, September, and October 2021, respectively.Furthermore, the patient exhibited a state of overall good health with no known allergies, other immune deficiencies, cardiovascular disorders, or diabetes mellitus.Additionally, the patient had no skin abnormality including atopic dermatitis, psoriasis, dermatophytosis, and tinea cruris.However, the patient has not received any acupuncture or moxibustion treatment before the onset of the disease.Positron emission tomography-CT scans showed metabolically active glands and inflammatory uptake in the destructive vertebral bodies of L2, L3, and L5.Repeat MRI showed an expansive presence of abnormal signals in the vertebral body, resulting in progressive bone destruction compared to previous MRI findings (Fig. 1B).After obtaining informed consent from family members, a CT-guided percutaneous biopsy procedure was conducted to obtain specimens of the spinal lesions located at the L4 vertebra.Acid-fast staining and TB DNA and mycobacterium DNA tests yielded negative results in the aforementioned specimens.Bacterial, mycobacterial, and fungal cultures of the aforementioned specimens yielded negative results as well.
To detect all potential pathogens and confirm a diagnosis, concurrent PMSeq of the L4 lesion specimens was performed in the Wuhan BGI clinical laboratory.Specifically, the vertebral lesion sample, ranging from 0.5 to 3 mL, was collected from the patient following standardized procedures.Briefly, DNA extraction and construction were performed following the manufacturer's protocol, which involved DNA fragmentation, end-repair, adapter-ligation, and polymer chain reaction (PCR) amplification.The generation of high-quality sequencing data involved the removal of low-quality reads, followed by computational subtraction of human host sequences that were mapped to the human reference genome (hg19) using Burrows-Wheeler Alignment.After eliminating low complexity reads, the remaining data were classified by simultaneously aligning them to four Microbial Genome Databases, consisting of bacteria, fungi, viruses, and parasites.The classification reference databases were downloaded from the National Center for Biotechnology Information (NCBI) (ftp://ftp.ncbi.nlm.nih.gov/genomes/).Ultimately, the detection of pathogens was reported in accordance with the described criteria [14].On July 25, 2022, PMseq identified M. restricta as the predominant pathogen with the highest relative abundance value (91.08%); the other pathogens identified were Cutibacterium acnes (relative abundance value: 21.22%), Kocuria palustris (relative abundance value: 5.12%), Acinetobacter junii (relative abundance value: 3.55%) and Pseudomonas monteilii (relative abundance value: 2.49%).However, M. tuberculosis complex was not identified (Table 1).Additionally, we conducted a pathological examination that revealed the presence of fungal mycelium in consecutive sections of the vertebral lesion tissue through periodic Schiff-methenamine (PASM) and periodic acid-Schiff (PAS) staining (Fig. 2).These results suggested that the fungal infection of M. restricta was implicated in vertebral body bone destruction, rather than this being a case of tuberculous spondylitis.Therefore, the anti-TB treatment was discontinued, and a combination of fluconazole and voriconazole was administered for antifungal therapy.After a 4-month course of antifungal therapy (4 weeks of oral fluconazole 400 mg/ day, then 3 months of oral voriconazole, 300 mg/day), all his symptoms resolved.At the 13-month follow-up, no significant signs of low back pain were evident (VAS score of 1 out of 10), independent walking was achieved, and the range of vertebral lesions was reduced as evidenced by MRI (Fig. 3).

Discussion and conclusions
Fungi have been widely acknowledged as an opportunistic causative agent of vertebral osteomyelitis, accounting for 0.5 − 1.6% of spinal infections.The most common etiological agents include Aspergillus, Candida, and Mucor [15,16].Patients with fungal spondylitis often present with a high-risk history of fungal infections or comorbidities such as diabetes, leukemia, tumors, and immunocompromised-related diseases [17].The predominant clinical manifestations of fungal spinal osteomyelitis include local pain, pyrexia, muscular debility, and additional  [56][57][58].However, Malassezia has rarely been identified as a pathogen involved in spinal infections, with its main association being chronic infections such as Crohn's disease, cytokine release syndrome, and respiratory diseases except for skin disorders [7].Upon retrospective analysis of all previous cases since 2010, we have identified only eight documented instances of Malassezia infection, including four patients with endocardial infection [59,60], three patients with pulmonary  infection [61,62], one patient with liver infection [63], and one patient with synovial infection [64] (Table 3).However, no reports of spinal infections caused by Malassezia were found.
In the aforementioned literature review, fungal spondylitis is typically diagnosed through positive cultures and histologic examination of infectious lesion specimens in conjunction with clinical symptoms, laboratory findings, and radiological features.However, our patient was a middle-aged man without any underlying medical conditions or history of abnormal immune function and belonging to an unsusceptible population.The diagnosis of spinal TB is primarily based on the involvement of multiple vertebral bodies, resulting in destruction of the spinal structure observed on CT imaging; however, in this case, evidence from skin, blood tests, or tissue cultures was insufficient to definitively support the presence of infection.The initial pathological examination, including hematoxylin & eosin staining and tissue culture, did not detect any pathogens.Furthermore, the subsequent treatment of anti-TB medication was deemed ineffective.However, it is worth noting that the puncture biopsy under the guidance of CT is recognized as a critical method of diagnosis in cases of unknown etiology of vertebral body destruction [65].Therefore, PMseq of specimens of the L4 vertebral lesion was performed concurrently to detect all possible pathogens.The results indicated that a fungal infection caused by M. restricta may be responsible for the patient's vertebral body destruction, rather than being a case of tuberculous spondylitis.Metagenomic sequencing, as an innovative culture-independent approach, possesses the capacity for rapid, sensitive, and accurate identification of pathogens.Numerous recent studies have demonstrated that metagenomic sequencing plays a pivotal role in successful pathogen recognition for infections characterized by atypical clinical symptoms that are unidentifiable through traditional etiological detection methods or caused by unknown pathogens [66,67].These findings underscore the pivotal role of metagenomic sequencing in guiding drug administration and enhancing clinical outcomes for patients with spinal infection, particularly when conventional methods fail to identify the pathogen [68].In this study, the pathological examination revealed the presence of fungal mycelium in serial sections of the vertebral lesion tissue, as indicated by positive results of PASM and PAS staining.Consequently, the anti-TB treatment was replaced with a combination of ketoconazole and itraconazole as antifungal therapy, which is widely acknowledged as essential and efficacious for managing fungal osteomyelitis [69].In our review, a favorable outcome was observed in 9 out of 40 cases (22.5%) who received prolonged treatment with antifungal agents.Notably, early identification and sustained administration of appropriate antifungal drugs such as fluconazole, voriconazole, etc., significantly enhanced the prognosis for patients with M. infection.Fortunately, the patient's symptoms of impaired ambulation and lumbago were successfully alleviated based on the VAS score, whereas the extent of vertebral lesions demonstrated a reduction on MRI.Although no precise pathogen evidence was detected through blood tests and tissue culture, the therapy was proved effective for the patient following guidance from PMseq of infected tissue.Furthermore, confirmation of Malassezia infection may be achieved through a combination of pathology, blood cultures, and other methods involving selective pathological examination and PCR testing [70].However, this study is limited by the absence of definitive evidence to ascertain the causative pathogen responsible for the infection in the tissue culture.Additionally, we did not rule out whether the injection of COVID-19 vaccine could be a risk factor for this infection because the time of onset of illness was within 6 months after the injection of the COVID-19 vaccine.Nevertheless, the PMseq results combined with suspected fungal presence were further supported by fungal mycelium observation from PASM and PAS staining, CT imaging, and the efficacy of antifungal treatment.These findings provide additional confirmation to the speculation that M. restricta is recognized as the most predominant pathogen .In summary, we present the first reported case of spondylitis caused by Malassezia and provide a comprehensive review of 40 published cases of fungal spondylomyelitis as well as eight cases of systemic organ infection caused by Malassezia.Given the atypical clinical symptoms and imaging findings observed in fungal spondylitis, along with the high rates of negative diagnosis for Malassezia, a single diagnostic method may not be sufficient to accurately diagnose Malassezia spondylitis.Therefore, employing PMseq from affected tissue could serve as a rapid and precise approach to guide drug management and clinical treatment for certain non-specific infectious diseases.

Fig. 1A
Fig. 1A CT showed an irregular bone destruction at the L2 vertebral body (a).MRI showed irregular bone destruction at the L2 vertebral body endplate (white arrow) with a T2-weighted image (c) and a lesion with low T1 (b) and high T2 intensity (c) at the L2 vertebral body and L2-L3 intervertebral disc (red arrows) on March, 2022.Fig. 1B CT showed a significant irregular bone destruction at the L2 and L3 vertebral body (a).MRI showed irregular bone destruction at the L2 and L3 vertebral body endplate (white arrow) with a T2-weighted image (c) and a lesion with low T1 (b) and high T2 intensity (c) at the L2 and L3 vertebral body and L2-L3 intervertebral disc (red arrows) on July, 2022

Fig. 3 Fig. 2
Fig.3Imaging manifestations after treated with anti-fungal therapy.CT showed an significant reduction of bone destruction in vertebral body (a).MRI showed reduced bone destruction at the L2 and L3 vertebral body (white arrow) compared with the imaging performance on July, 2022 (b&c)

Table 2
Main characteristics of spine fungal osteomyelitis

Table 3
Main characteristics of spine fugal osteomyelitis