Background

Pulmonary cryptococcosis (PC) is a lung infection caused by inhaling Cryptococcus neoformans or Cryptococcus gattii in the environment, which are two major cryptococcal pathogens in humans and animals; these two pathogens belong to the Cryptococcus neoformans species complex. When Cryptococcus spores are inhaled, they frequently enter the human lower respiratory tract. It is most common in immunocompromised patients. However, PC becomes one of the emerging diseases in immunocompetent patients [1], and epidemiological surveys in recent years have shown that the proportion of asymptomatic infections in immunocompetent PC patients is higher than that in immunocompromised patients (40.8% and 30.2%, respectively) [10]. In addition, the long-term empirical treatment with broad-spectrum antibiotics may increase drug resistance in some pathogens. Conventional tests for infectious pathogens, such as microbial culture, microscopy methods, antigen detection techniques, and serologic tests, have deficiencies like long average detection turn-around time (TAT), low sensitivity and low positive rate. These shortcomings increase the risk that a mixed pulmonary infection is missed or misdiagnosed, hindering patients from receiving effective care in the timeliest manner possible.

Metagenomic next-generation sequencing (mNGS), an emerging pathogen detection method in recent years, can directly detect and identify pathogens by sequencing without culturing or classifying. Here, we retrospectively analyzed a female diagnosed with bacterial pneumonia complicated by PC and showed how mNGS contributes to the identification of infectious pathogens during diagnostic procedures.

Case presentation

A 43-year-old female was admitted to the Department of Respiratory and Critical Care Medicine, the First People’s Hospital of Qinzhou, because she had been coughing for over a month. One day before hospital admission, her chest CT at another local hospital showed space-occupying lesions in the right lower lobe, along with cavities and multiple nodules. Her preliminary diagnosis was a pulmonary infection. She was admitted to the hospital for further diagnosis and treatment. About 10 months ago, she underwent right plantar cyst resection at the Department of Joint and Sports Medicine of the hospital; the surgery was successful and she made a smooth recovery. She denied a history of cigarette-smoking, tuberculosis, intrafamilial infectious diseases, and a family history of genetic diseases. Physical examinations at the time of admission were normal, with the following parameters: body temperature 36.2 °C, breathing rate 20 times/min, heart rate 112 times/min, and blood pressure 116/83 mmHg. Her right lower lung breath sounds were slightly coarse, while her right upper lung and left lung breath sounds were clear, and there were no obvious rales or pleural friction rub in both lungs. Auxiliary examinations were as followed: WBC 12.32 × 109/L, absolute neutrophil count = 8.48 × 109/L, CRP = 16.04 mg/mL, ESR = 72.00 mm/h; serum CrAg-LFA test was positive; the M. tuberculosis antibody TB-Dot test was weakly positive; no abnormalities were found in G and GM test, liver and kidney function, blood gas, tumor markers test, and squamous cell carcinoma antigen examinations. Her contrast-enhanced chest CT image revealed multiple nodules and plaques in her right lung, as well as the formation of cavities, which is considered the characteristic of tuberculosis, but the fungal infection could not be ruled out (Fig. 1b). She was diagnosed with pulmonary infection due to the nature of the space-occupying lesions and cavities in the right lower lobe, and moxifloxacin (0.4 g, qd, by intravenous drip) was used for empirical anti-infective therapy.

Fig. 1
figure 1

Comparisons of chest CT without infection one year before admission (a), contrast-enhanced chest CT on the second day of admission (b) and chest CT on the 18th day of admission (c). a The CT imaging showed no no abnormality. b The CT imaging showed plaque shadowing and obvious cavity formation in the right lung, as well as multiple nodules. c The lesions had obvious absorption

Full size image

Chronologically, on the3rd day of admission, her tuberculin skin test (PPD) was negative. On the 7th day, her body temperature was 36.5 °C. Her cough did not improve significantly, and she experienced occasional chest pain. No abnormality was found in bronchoscopy, and BALF samples were collected for microbial culture, G and GM test, M. tuberculosis DNA test, and fungal and Cryptococcus smear, all of which proved negative. In the meantime, a BALF specimen was directly sent to Clinical Genome Center, Guangxi Kingmed Diagnostics, for mNGS in accordance with the previous references [11, 15], the average TAT of conventional tests is over 3 days, while the TAT of mNGS only takes 2 days. In our case, the mNGS result was report within 40 h. Long TAT of traditional detection methods and a complex diagnosis and treatment route, cause the diagnosis of mixed infections and uncommon pathogens to take a long time, which is one of the reasons for poor prognosis [16]. What is more, long-term empirical treatment with broad-spectrum antibiotics has led to the emergence of drug resistance in some pathogens, and the drug resistance itself is increasing. Also, the early use of broad-spectrum antibiotics is one of the reasons for the low positive rate of conventional detection methods [17]. mNGS is unbiased and has broad pathogens coverage, which greatly offsets the shortcomings of conventional tests for the detection of mixed infection. It can quickly identify multiple pathogens of mixed pulmonary infection without bias as well as assumption and minimize missed diagnosis, lessening the abuse of antibiotics as a consequence of inexperience and the long TAT of conventional methods. Jiahui et al. [17] analyzed 55 cases of mixed pulmonary infection and found that the sensitivity of mNGS in diagnosing mixed pulmonary infection was about 7 times higher than that of conventional tests (97.2–13.9%, respectively). Besides, the results of mNGS can assist clinicians to adjust anti-infective therapy. Tianjun et al. [18] discovered that that 60.6% of the patients showed apparent improvement after following mNGS results, wheras only 37.9% of the patients diagnosed by conventional test improved. In our case, the mNGS result was a mixed infection of bacteria and Cryptococcus neoformans, while the bacterial culture and smear of BALF samples were both negative. Clinicians modified the anti-infectious treatment plan in time, which diminished the time of using broad-spectrum antibiotics.

It is noteworthy that mNGS can quickly and unbiasedly detect and identify pathogens in mixed-infected patients without culture, and it has more outstanding sensitivity than traditional tests [19], particularly in detecting fungi, viruses, and anaerobic bacteria. The high sensitivity of mNGS makes it easier for positive pathogens to be detected in samples even if these pathogens are with a low load. Because of the advantages of high sensitivity, broad coverage of pathogens, and short TAT [7], mNGS can help clinicians to quickly and accurately identify unknown etiologies [20]. The high pathogen coverage and unbiasedness of mNGS play a great role in the differential diagnosis of pathogens causing lung infections, including mixed pulmonary infections, rare pathogen infections, and the differential diagnosis of some bacteria and fungi. There have been numerous reports of misdiagnosed and missed-diagnosed patients whose diagnoses were modified after mNGS tests [21]. However, it is undeniable that mNGS still has certain limitations. In our case, one of the reasons for the low proportion of pathogen reads detected is that most of the detected microorganisms were reagent engineering microorganisms and colonization microorganisms (Malassezia, Sphingomonas, and Acinetobacter in our case). This drawback can be overcome by optimizing the bioinformatics analysis process and using negative control. In addition, other issues involve the interference of human source sequences with pathogen detection and low detection efficiency for pathogens with cell wall protection. The optimization can include removing human source sequences, and optimizing protocols to enable rupturing cell walls more easily for releasing pathogens’ nucleic acids.

In conclusion, we present a case where the diagnosis was modified from an unexplained pulmonary infection to PC and bacterial pneumonia after mNGS test, demonstrating the potential role of mNGS in distinguishing PC from pulmonary tuberculosis when conventional test results are inconsistent and precisely detecting pathogens in mixed pulmonary infection. Compared with some traditional methods, mNGS can detect pathogens more swiftly and accurately with higher sensitivity, which can assist clinicians to identify pathogens with similar imaging manifestations, and greatly reduce the time of diagnosis. Additionally, the patient in our case recovered after anti-infection therapy was adjusted to a suitable one, which also suggests that mNGS results have the potential ability to adjust treatment regimens. It is very meaningful because it may reduce the abuse of antibiotics, and providing clear ideas for precise anti-infectious treatment plans. Although there are still some limitations of mNGS at this stage, they are expected to be gradually improved with the constant update and optimization of the experimental protocols and bioinformatics analyses, and mNGS is suitable to be used for assisting clinical diagnosis of mixed pulmonary infection.