Clinical patterns, species-specific correlations, and therapeutic outcomes of nocardiosis in immunocompetent individuals: a systematic analysis based on literature

Abstract

Background

Nocardiosis, traditionally an opportunistic infection, lacks comprehensive characterization in immunocompetent individuals. This study aimed to describe and analyze the demographics, clinical features, and outcomes of Nocardia infections in this population based on the literature.

Methods

A systematic analysis of 530 immunocompetent Nocardia cases (2020–2024) was conducted using reports from PubMed, CNKI, and MedNexus. Demographics, exposures, clinical characteristics, laboratory, treatment, and outcome were recorded and analyzed.

Results

Patients (median age: 59 years; 69.8% male) exhibited pulmonary (46.5%), cutaneous (20.0%), or cerebral (15.7%) involvements. Risk factors included soil/dust exposure (97/530, 18.3%), trauma (44/530, 8.3%), and comorbidities (255/530, 48.1%, e.g., diabetes, bronchiectasis). Cure/improvement (82.6%) group exhibited younger age (P = 0.01) and higher prevalence of local infections (P < 0.001), while more cases of disseminated infection (P < 0.001), N. farcinica (P = 0.01), and unclassified Nocardia spp. (P = 0.02) were in the deterioration/mortality group. Trimethoprim-sulfamethoxazole (71.9%) was the primary therapy. Species-specific patterns demonstrated: N. brasiliensis preferentially linked to cutaneous infections (P < 0.001), N. farcinica to cerebral disease (P < 0.001), and N. cyriacigeorgica to pulmonary involvement (P < 0.001).

Conclusion

Nocardiosis in immunocompetent hosts demonstrates distinct risk profiles, site-specific species associations, and prognostic factors. Early diagnosis, species identification, and antimicrobial therapy are critical for optimizing outcomes. This study underscores the need for heightened clinical suspicion and improved diagnostic protocols in non-immunocompromised populations.

Introduction

Nocardia spp., ubiquitous environmental actinomycetes of the order Corynebacteriales/ formerly Mycobacteriales, have long been recognized as opportunistic pathogens predominantly affecting immunocompromised populations, including transplant recipients, individuals with hematological malignancies, and those with HIV/AIDS or chronic corticosteroid use [1–3]. Studies reported up to 36% of cases occurring in patients without overt immunodeficiency [4].

Nocardiosis in immunocompetent populations is poorly characterized, presenting with clinical heterogeneity such as subacute pneumonia, cutaneous nodules resembling mycobacterial infections, or disseminated lesions involving muscles, bones, and the central nervous system [5]. This often leads to delayed diagnosis and suboptimal outcomes due to nonspecific symptoms and diagnostic challenges.

Despite advances in molecular diagnostics like metagenomic next-generation sequencing (mNGS), knowledge on demographic characteristics, species-specific tropisms, and prognostic factors in immunocompetent populations remains fragmented, relying mainly on case reports or small studies. This study systematically analyzes 530 immunocompetent Nocardia cases globally to (1) describe demographic and clinical characteristics, (2) identify species-specific infection patterns and prognostic correlates, and (3) evaluate diagnostic and therapeutic practices.

Method

Data collection

A systematic and comprehensive literature search was performed across three databases (PubMed, CNKI, and MedNexus) using the Medical Subject Headings term “Nocardia”, targeting publications from January 2020 to December 2024. To ensure the inclusion of higher-quality publications in our dataset, we included individual case reports and case series from the literature that contain data on the specific elements in our study. Moreover, our search strategy imposed no language restrictions in PubMed, while also incorporating CNKI and MedNexus. Although CNKI predominantly hosts Chinese-language publications, a portion provides English abstracts or full texts. Therefore, duplicate records exist across the CNKI, MedNexus, and PubMed databases. The retrieved records were manually screened, with priority given to records from PubMed in cases of duplicate entries. After exclusion criteria were applied, relevant articles aligned with the study objectives were identified. The study intended to collect information on Nocardia infections in immunocompetent patients, excluding cases reporting on immunodeficient patients and patients on immunosuppressive therapy, non-Nocardia infected cases, non-human studies, cases without details, and articles with unavailable full-text access. In our study, inclusion was limited to cases meeting clinical diagnostic criteria for infection—specifically, those exhibiting compatible symptoms supported by corroborating those exhibiting compatible symptoms supported by corroborating microbiological evidence, and cases of colonization were excluded.

Specifically, immunodeficiency disorders included human immunodeficiency virus, systemic lupus erythematosus, vasculitis, immune thrombocytopenic purpura, pulmonary alveolar proteinosis (specifically the idiopathic form characterized by anti-GM-CSF antibody positivity), and chronic granulomatous disease. The latter two conditions are considered immunocompromised states. Immunosuppressive therapy encompasses long-term exposure to corticosteroids, immunosuppressive agents used for organ transplantation or nephrotic syndrome, and chemotherapy. Cases without details referred to some single- or multi-center retrospective or prospective studies that lack detailed clinical information, making it impossible to extract data on immunocompetent patients from them. This systematic approach ensured the inclusion of high-quality, relevant studies for further evaluation.

Data categories

We systematically collected data across 16 distinct categories to ensure comprehensive analysis. These categories encompassed: (1) publication type, (2) year of publication, (3) gender, (4) age, and (5) author’s origin country. Additionally, we documented (6) exposure history (including soil/dust ingestion, trauma, invasive therapy, and contact lens wear), (7) underlying medical condition (e.g., diabetes, hypertension, dyslipidemia, bronchiectasis, chronic obstructive pulmonary disease (COPD), tuberculosis, tumor, hypothyroidism). The dataset further included (8) clinical manifestations, (9) infection sites, (10) diagnostic methods, (11) antimicrobial susceptibility testing (AST), (12) Nocardia spp., (13) co-infections (bacterial, viral, fungal, and mold), (14) therapeutic drugs, and (15) outcomes. This multidimensional data collection framework enabled robust characterization of Nocardia infections in an immunocompetent host.

Definitions

To ensure consistency in data extraction, we established standardized definitions for relevant parameters. Occupational exposures (workers, farmers, stonemasons, herdsmen, horticulturists, construction workers, and bricklayers), exposure to post-rainwater accumulation, and wilderness habitation were all defined as soil and dust ingestion. Trauma included injuries from wood splinters, thorn pricks, dog bites, and falls. Invasive therapy was defined to include lung biopsy, pacemaker implantation, pericardiocentesis, and acupoint injection. Clinical presentations were systematically classified: swelling in hands, arms, upper and lower extremities were all categorized as skin swelling, while lumbar and thoracic vertebrae were grouped as vertebrae. Disseminated infection was defined as either the involvement of more than two anatomic locations or central nervous system infections. Diagnostic methods were standardized as follows: culture and mNGS of secretions, sputum, lung tissues, pleural effusion, bronchoalveolar lavage fluid, cornea, pus, and brain tissues were all classified as tissue cultures and tissue mNGS. Suspected Nocardiosis was classified as unclassified Nocardia species. Outcomes were defined as follows: complete clinical and radiographic resolution, as well as reported as “recovered” or “complete recovery” were classified as cured. Cases described as “clinical status gradually improved” were grouped into the improved group. Patients who discontinued treatment or had unmentioned outcomes were categorized as lost to follow-up. These standardized definitions were implemented to facilitate statistical analysis of heterogeneous data sets.

Data encoding

For subsequent data processing using R software, the variables were coded as follows: Gender was encoded as 1 for male and 2 for female. In addition to being represented numerically, age was also dichotomized with 1 representing patients aged ≥ 65 years and 0 for those aged < 65 years. Medical history variables, including exposure history (soil and dust ingestion, trauma, invasive therapy, and contact lens wear) and underlying medical condition (diabetes, hypertension, dyslipidemia, bronchiectasis, COPD, tuberculosis, tumor, and hypothyroidism) were coded as 1 when present and 0 when absent or unspecified. Co-infections (bacterial, viral, and fungal) were designated 1 and 0 for absence. Disseminated infections are coded as 1, while non-disseminated infections are marked as 0. Laboratory parameters included binary coding for AST (1 = performed, 0 = not performed). Outcomes were categorized as 1 (cure/improvement) and 0 (deterioration/death), with NA assigned to lost follow-up cases. All unrecorded or missing data elements were systematically encoded as NA to facilitate consistent missing value handling in statistical analyses.

Data analysis

Statistical analysis and data visualization were performed using R software (version 4.2.3), which was chosen for its capabilities to perform not only statistical analyses but also visualization of geographic distributions and anatomical organs. Dichotomous variables were compared using chi-square tests, Fisher’s exact tests, or logistic regression analysis, while continuous variables were assessed using Mann-Whitney U tests, with statistical significance defined as P < 0.05. The “rworldmap” package was used to visualize the global distribution of Nocardia cases, supplemented by bar charts illustrating case counts and Nocardia species proportions in different countries. Nocardia species composition was represented through pie charts. Sites of Nocardia infection were mapped using the “gganatogram” package for the visualization of anatomical organs. Rank correlations between data categories were quantified using Spearman’s correlation and presented as heatmaps. The diagnostic method was visualized through Venn diagrams combined with donut charts, since multiple diagnostic modalities may be used simultaneously in a single case. Antimicrobial patterns were displayed via the Nightingale rose chart. All visualizations were implemented through ggplot2 extensions to maintain graphical consistency, with alpha shapes employed for proportional representation of overlapping categorical data. Missing values were coded as NA. For chi-square tests and Fisher’s exact tests, we employed Row-wise NA Removal to handle missing data. For correlation analyses and between-group comparisons, we applied pairwise deletion to address missing values.

Result

Screening results and paper types

Following systematic literature retrieval across PubMed, CNKI, and MedNexus, 1,524 articles were initially identified. After deduplication (n = 25), 1,499 articles encompassing 6254 case reports were retained, comprising 4768 cases from PubMed, 1,305 cases from CNKI, and 181 cases from MedNexus (Fig. 1). Exclusion criteria were rigorously applied: immunodeficiency (n = 655), immunosuppressant use (n = 376), insufficient clinical data (n = 3428), non-Nocardia infections (n = 805), and non-human studies (n = 460). Among the final 530 included cases, 204 were sourced from CNKI, 41 from MedNexus, and 285 from PubMed. Notably, to ensure data integrity, a dual-reviewer process was implemented with methodological rigor: one researcher conducted full-text screening and data extraction for all included articles, while a second investigator performed independent verification through random sampling of records (n = 50).

Fig. 1.

Flowchart of the selection process for Nocardia infection cases in immunocompetent patients

Demographic and clinical characteristics

The outcomes of the final obtained 530 cases included cure/improvement in 438 cases (82.6%), deterioration/death in 42 cases (7.9%), and 50 cases lost to follow-up (9.4%) (Table 1). Patients had a median age of 59 years (IQR 47–70 years; range 1–96 years), with a predominance of young and middle-aged adults (199/530, 37.5% ≥ 65 years vs. 329/530, 62.1% < 65 years). Male patients predominated (319/530, 60.2%) over females (206/530, 38.9%). Among 530 cases, Soil and dust ingestion constituted the primary risk factor (97/530, 18.3%), followed by invasive therapy (47/530, 8.9%) and trauma (44/530, 8.3%). The underlying medical condition affected 48.1% (255/530), predominantly diabetes (70/530, 13.2%), bronchiectasis (62/530, 11.7%), hypertension (60/530, 11.3%), and COPD (42/530, 7.9%). Immunocompetent patients with nocardiosis predominantly presented with non-disseminated infections occurred in 373 cases (70.40%). Concurrent infections affected 10.9% (58/530), primarily bacterial co-infections (34/530, 6.4%).

Table 1.

Comparison of demographic and clinical characteristics between the cure/improvement and deterioration/death groups

	Total (n = 530)		Cure/Improvement (n = 438/480)		Deterioration/Death (n = 42/480)		Univariate
	No.	%	No.	%	No.	%	P-value
Age
Range, years	1–96		1–90		16–91		0.000
< 65 years old	329	62.1	278	63.5	18	42.9
≥ 65 years old	199	37.5	159	36.3	23	54.8	0.013
Not mentioned	2	0.4	1	0.2	1	2.4
Gender
Male	319	60.2	257	58.7	26	61.9
Female	206	38.9	177	40.4	15	35.7	0.601
Not mentioned	5	0.9	4	0.9	1	2.4
Exposure
Have a history	181	34.2	157	35.8	10	23.8
Soil and dust ingestion	97	18.3	87	19.9	5	11.9	0.217
Trauma	44	8.3	37	8.4	1	2.4	0.195
Invasive therapy	47	8.9	39	8.9	4	9.5	0.893
Contact lens wear	3	0.6	3	0.7	0	0.0	0.987
Not mentioned	349	65.8	281	64.2	32	76.2
Underlying medical condition
Without an underlying condition	275	51.9	229	52.3	17	40.5
With underlying conditions	255	48.1	209	47.7	25	59.5	0.144
Diabetes	70	13.2	59	13.5	6	14.3	0.883
Hypertension	60	11.3	51	11.6	5	11.9	0.050
Dyslipidemia	12	2.3	9	2.1	1	2.4	0.888
Bronchiectasis	62	11.7	55	12.6	0	0.0	0.985
COPD	42	7.9	32	7.3	6	14.3	0.117
Tuberculosis	19	3.6	16	3.7	2	4.8	0.719
Tumor	23	4.3	19	4.3	2	4.8	0.898
Hypothyroidism	9	1.7	9	2.1	0	0.0	0.896
Others	70	13.2	57	13.0	8	19.0
Disseminated
Yes	157	29.6	121	27.6	23	54.8
No	373	70.4	317	72.4	19	45.2	0.000
Nocardia spp.
Unclassified	153	28.9	115	26.3	18	42.9	0.024
N. farcinica	88	16.6	67	15.3	13	31.0	0.011
N. cyriacigeorgica	61	11.5	59	13.5	1	2.4	0.070
N. brasiliensis	54	10.2	52	11.9	0	0.0	0.896
N. otitidiscaviarum	45	8.5	37	8.4	4	9.5	0.812
N. asteroides	23	4.3	20	4.6	0	0.0	0.896
N. beijingensis	12	2.3	11	2.5	0	0.0	0.984
N. abscessus	16	3.0	14	3.2	0	0.0	0.988
Others	84	15.8	69	15.8	6	14.3
Co-infection status
No co-infection	472	89.1	392	89.5	37	88.1
Co-infections	58	10.9	46	`10.5	5	11.9	0.778
Bacterial	34	6.4	28	6.4	3	7.1	0.850
Viral	8	1.5	6	1.4	2	4.8	0.124
Fungal	18	3.4	12	2.7	2	4.8	0.463
Others	3	0.6	3	0.7	0	0.0
AST
Yes	190	35.8	157	35.8	16	38.1
No	300	56.6	251	57.3	17	40.5	0.258
Not mentioned	40	7.5	30	6.8	9	21.4

COPD chronic obstructive pulmonary disease, AST antimicrobial susceptibility testing

Notable prognostic associations emerged in immunocompetent Nocardia patients. Patients in the cure/improvement group were younger than those in the deterioration/death group (278/530, 63.5% in the cure/improvement group < 65 years vs. 18/42, 42.9% < 65 years in the deterioration/death group, P = 0.01), while the proportion of disseminated infection was higher in the deterioration/death group (121/438, 27.6% in the cure/improvement group vs. 23/42, 54.8% in the deterioration/death group, P < 0.001). Additionally, the proportions of unclassified Nocardia spp. (115/438, 26.3% vs. 18/42, 42.9%, P = 0.02) and N. farcinica (67/438, 15.3% vs. 13/42, 31.0%, P = 0.01) were lower in the cure/improvement group.

Infection site and clinical symptoms

Nocardiosis reported in immunocompetent individuals exhibited multi-system involvement affecting the central nervous, circulatory, respiratory, endocrine, and musculoskeletal systems (Fig. 2). The most frequently involved organs were the lungs (46.5%), followed by the skin (20.0%) and brain (15.7%), consistent with Nocardia’s primary transmission via the respiratory route and skin-to-skin contact (This referred to the patients’ skin exposure to environmental sources -such as soil, dust, or puddles- rather than to direct contact with the skin of infected individuals). Pulmonary infections predominantly manifested as cough (22.9%), phlegm (17.8%), fever (15.8%), and breathlessness (5.0%). Cutaneous infections primarily presented with fever (9.5%) and localized lesions such as skin ulcers (12.5%), swelling (9.9%), and subcutaneous mass (11.0%). Cerebral infections were characterized by fever (16.1%), headache (14.5%), consciousness (9.3%), and fatigue (6.4%). Overall, Nocardia infections in immunocompetent hosts demonstrated localized symptomatology accompanied by systemic manifestations, including fever, fatigue, and weight loss.

Fig. 2.

Anatomical distribution mapping of Nocardia infections in immunocompetent patients

Diagnostic and therapeutic modalities

Microbiological culture, particularly tissue culture, remains the primary diagnostic modality for Nocardiosis (Fig. 3A). Clinically, multi-method approaches combining tissue culture with 16S ribosomal RNA (16S rRNA) gene sequencing, time of flight mass spectrometry (MALDI-TOF MS), mNGS, or polymerase chain reaction (PCR) are routinely employed to facilitate the identification of Nocardia species (Fig. 3B). As not all reports documented the year of disease onset, the time scale in Fig. 3B represented the year of publication. While the figure suggests an increasing trend in mNGS testing over time, this interpretation must be viewed with caution due to the potential bias introduced by publication delay. Blood-based diagnostics demonstrated limited utility. Among 157 disseminated Nocardia cases, only 21 (13.4%) were confirmed through blood cultures or mNGS. Trimethoprim-sulfamethoxazole (TMP-SMX) (381/530, 71.9%) persisted as the cornerstone therapeutic agent, supplemented by carbapenems (imipenem/cilastatin, meropenem, biapenem) (167/530, 31.5%), oxazolidinones (linezolid, tedizolid) (112/530, 21.1%), aminoglycosides (94/530, 17.7%), and cephalosporins (73/530, 13.8%) in descending order of prescription frequency (Fig. 3C). In addition, surgical interventions were employed in some cases (73/530, 13.8%). Obtaining infected tissue specimens for diagnostic confirmation, such as craniotomy for brain abscesses, and percutaneous or surgical drainage for subcutaneous abscesses. Meanwhile, surgically removing abscesses accelerated infection resolution.

Fig. 3.

The diagnosis and medication characteristics of immunocompetent Nocardia infections. A Diagnostic modalities visualization was achieved through complementary the donut chart and Venn diagram. The peripheral donut chart showed the quantitative distribution of diagnostic modalities employed in the case. The Venn diagram of the central demonstrated overlapping application of diagnostic modalities. B Proportions of different diagnoses from 2020 to 2024 depicted in a stacked bar chart. C The Nightingale Rose chart visualized pharmacotherapeutic utilization frequencies, with the chromatic sector corresponding to distinct pharmacological categories and each bar denoting one specific therapeutic agent

Correlation between categories

The correlations between gender, age, exposures, underlying medical conditions, co-infection, infection sites, Nocardia species, disseminated infections, and clinical outcomes were evaluated. Spearman’s correlation analysis revealed significant associations between outcomes and multiple variables, including age, infection sites, Nocardia species, comorbidities, and disseminated infections (Fig. 4A). Consistent with previous findings, age (P < 0.001), disseminated infections (P < 0.001), and unclassified Nocardia spp. (P < 0.05), and N. farcinica (P < 0.01) negatively correlated with the prognosis for cure/improvement, which showed a positive correlation with the underlying disease of bronchiectasis (P < 0.05). Moreover, pulmonary (P < 0.01), cerebral (P < 0.05), and hematogenous infections (P < 0.01) showed positive associations with deterioration/death outcomes. We also found that disseminated infection was positively correlated with infection site, including brain (P < 0.001), blood (P < 0.001), and N. farcinica (P < 0.001). Our analysis further revealed significant associations between Nocardia species and infection sites, exposures, and underlying conditions (Fig. 4A). More specifically, N. farcinica showed positive associations with cerebral infections (P < 0.001). N. brasiliensis demonstrated strong correlations with cutaneous (P < 0.001), while exhibiting inverse relationships with pulmonary (P < 0.001). N. cyriacigeorgica was preferentially linked to pulmonary infection (P < 0.001), yet inversely associated with cutaneous infection (P < 0.001).

Fig. 4.

Correlation analysis between categories. A The heatmap elucidated multidimensional correlations among gender, age, infection sites, Nocardia species, underlying diseases, co-infections, disseminated infections, and prognostic outcomes. B The heatmap visualized the association between therapeutic modalities, infection sites, Nocardia species, and outcomes. The correlations were measured using the Spearman rank correlation coefficient. Red and blue represented positive correlations and negative correlations, respectively. Color saturation denotes correlation strength. *P < 0.05; **P < 0.01; ***P < 0.001

Antimicrobial treatment regimens were found to be significantly related to Nocardia species and infection sites (Fig. 4B). Carbapenem antibiotics demonstrated positive correlations with central nervous system (P < 0.001) and pulmonary (P < 0.001) infections but showed negative associations with cutaneous (P < 0.001) infections. Oxazolidinones were positively associated with central nervous system infections (P < 0.001).

Distribution of Nocardia species and cases

Our statistical analysis revealed distinct geographical patterns in nocardiosis reporting, with China documenting the highest number of immunocompetent Nocardia cases globally, followed by the United States and India (Fig. 5A and B). This finding may be influenced by our inclusion of the CNKI database. While CNKI broadened the language coverage of the databases in our study, it also increased the clinical cases reported from China. Stacked column chart analysis demonstrated no significant country-specific distribution patterns for Nocardia species. Among 530 total cases, species-level identification was achieved in 377 cases (71.1%), with four predominant species collectively accounting for nearly half of all characterized cases: N. farcinica (88/530, 16.6%), N. cyriacigeorgica (61/530, 11.5%), N. brasiliensis (54/530, 10.3%), and N. otitidiscaviarum (45/530, 8.5%). Notably, 153 cases (28.9%) remained unclassified, underscoring persisting diagnostic challenges in microbial characterization.

Fig. 5.

Global distribution and species of Nocardia infections among immunocompetent hosts. A Global distribution of reported Nocardia infections among immunocompetent patients (2000–2024). The lighter and darker colors represented fewer and more cases on the map. B The bar chart showed the country-specific case distribution of Nocardia infections, with China representing the highest caseload, followed by the United States and India. C Stacked bar charts demonstrated the distribution of Nocardia species across nations, while their proportional representation is illustrated through segmented column charts. D The pie chart visualization depicted the proportion of Nocardia species subtypes within 530 cases. Each color represented a species of Nocardia

Discussion

This study provides a systematic analysis of Nocardia infections in immunocompetent individuals, highlighting key demographic, clinical, and outcome insights. Pulmonary (47.4%) and cutaneous (18.2%) infections predominated, reflecting primary transmission routes—respiratory inhalation and direct skin inoculation. Notably, cerebral infections (16.6%) underscore severe CNS involvement risks, even in immunocompetent patients, necessitating heightened clinical suspicion for neurological symptoms [5]. Ocular nocardiosis, though rare (primarily affecting the cornea), is consequently prone to underdiagnosis or delayed diagnosis [6]. It is linked to invasive procedures, trauma, and contact lens use, warranting vigilance in exposed individuals. Cardiac involvement, exceptionally rare, was associated with invasive surgeries in our study.

Soil/dust exposure (56.4%) and trauma (23.8%) emerged as major risk factors, reinforcing the environmental reservoir hypothesis for Nocardia. In our analysis, trauma was critically associated with cutaneous nocardiosis in immunocompetent patients. Beyond individuals engaged in agricultural work, urban populations can also be exposed to Nocardia pathogens via aerosols originating from urban soil, dust, and water bodies. Research on Nocardia pathogens in urban seepage systems indicates that urban residents, when exposed to damp and polluted environments, can acquire infection through either direct contact or inhalation of aerosolized Nocardia pathogens [7]. Furthermore, occupational exposure to industrial textile dust without proper protective measures represents another potential exposure source [5]. Diabetes and bronchiectasis were frequent comorbidities, suggesting that chronic underlying diseases may predispose immunocompetent individuals to infection through impaired mucosal barriers or immune dysregulation. Diabetes mellitus has been described as immunocompromised in some cases [8, 9]. Although diabetes mellitus is not classified as a primary immunodeficiency disorder, the history of the disease has also been suggested to be an important susceptibility factor for Nocardia infections. Bronchiectasis was a common CT finding in pulmonary nocardiosis, with a disproportionately higher prevalence observed among immunocompetent patients compared to immunocompromised individuals with pulmonary nocardiosis [10, 11]. The role of Nocardia species isolated from the respiratory tract of bronchiectasis patients – whether as a colonizing bacterium or a true pathogen – remained controversial. Evidence suggested that Nocardia can be cultured from the sputum of some cystic fibrosis patients in the absence of clinical infection or pulmonary tissue destruction, suggesting probable colonization [12]. Notably, all patients in the Nocardia-positive case reports and series analyzed in our study demonstrated unequivocal clinical manifestations of infection, corroborated by robust clinical evidence. Interestingly, bronchiectasis history was independently associated with cure/improvement, possibly due to early detection in this cohort or the presence of specific host-pathogen interactions. This finding is consistent with the results of a retrospective study, which reported a higher survival rate among immunocompetent patients with bronchiectasis (39 out of 40 cases) [13]. Our findings also revealed an inverse correlation between bronchiectasis and disseminated infection, whereas a positive correlation was established with pulmonary infection. These were consistent with the study’s findings, in which all included Nocardia patients with bronchiectasis had pulmonary infections, but only one case presented with disseminated disease [13].

The identification of species-specific infection patterns demonstrated that N. farcinica was strongly linked to cerebral infections and worse outcomes, aligning with its known virulence and ability to invade the bloodstream. Current evidence characterizes N. farcinica as a particularly virulent and most common Nocardia species, demonstrating not only heightened dissemination capacity and resistance to antibiotics but also carrying significant mortality risks [14, 15]. A study detected the presence of virulence genes relA, icl, and mbtH, along with resistance to 11 antibiotics, including rifampicin, polymyxin-B, and penicillin G, in a clinically isolated N. farcinica sample [16]. In another retrospective analysis of 374 patients with Nocardia infections, N. farcinica was among the two most commonly isolated species, and infected patients had significantly higher 1-year mortality [17]. This species exhibits a distinct predilection for central nervous system involvement, with cerebral abscesses complicating approximately one-third of infections [14, 18]. Characteristic neurotropic manifestations include solitary multiloculated abscesses, multiple cerebral abscesses, or abscess formation concomitant with meningitis [18]. Conversely, N. brasiliensis was associated with cutaneous/lymphatic infections, as well as traumatic exposure. Our findings provided robust confirmation of previous literature regarding N. brasiliensis infections, which are primarily acquired through traumatic percutaneous inoculation, typically via environmental exposures such as thorn injuries or wood splinter penetration. In immunocompetent individuals, these infections predominantly manifest as three distinct clinical presentations: actinomycetoma, superficial cutaneous infections, and lymphocutaneous syndrome [19]. The characteristic localized disease progression and indolent clinical course of N. brasiliensis infections [20], as reported in the literature, might provide a pathophysiological basis for the lower deterioration/ mortality rate observed in our study, reflecting its tropism for superficial tissues, which has implications for targeted antimicrobial therapy and prognostic stratification. In our study, N. cyriacigeorgica (12.2% in the cure/improvement group vs. 2.4% in the deterioration/death group) emerged as another clinically significant species demonstrating comparable prognostic profiles to N. brasiliensis (11.9% in the cure/improvement group vs. 0% in the deterioration/death group) in immunocompetent hosts. However, its clinico-epidemiological profile diverged markedly, exhibiting a predilection for pulmonary involvement rather than cutaneous manifestations. Our correlation analysis in immunocompetent individuals revealed that N. cyriacigeorgica demonstrates not only associations with pulmonary nocardiosis, but also significant relationships with chronic pulmonary comorbidities—particularly chronic obstructive pulmonary disease (COPD) and bronchiectasis. Actually, among 99 patients with pulmonary nocardiosis included in a retrospective study, N. cyriacigeorgica was also identified as the most prevalent Nocardia species [21].

While histopathological tissue cultures and blood cultures remain the gold-standard diagnostic modalities for Nocardia identification in our study, novel molecular diagnostic platforms—including PCR-restriction fragment length polymorphism analysis, 16S rRNA gene sequencing, matrix-assisted laser desorption ionization, MALDI-TOF MS, and mNGS are being increasingly integrated into clinical workflows [22–25]. These advanced methodologies demonstrate superior diagnostic efficiency in achieving rapid speciation and precise taxonomic classification of Nocardia isolates. Despite advances in molecular diagnostics, a significant proportion of cases (25.1%) remained unclassified, which was notably associated with suboptimal clinical outcomes, indicating the necessity for precise species-level characterization to optimize patient management strategies. Although 16S rRNA gene sequencing is the conventional method for Nocardia identification, it demonstrates limited discriminatory power for closely related species, such as the Nocardia nova complex [26]. Subsequent studies have revealed that target genes with higher sequence diversity - including hsp65, gyrB, secA1, and sodA - demonstrate superior interspecies discrimination and polymorphic resolution compared to 16 S rRNA [27–30]. Multilocus sequence analysis (MLSA) assigns species on the basis of a concatenated sequence consisting of housekeeping genes (e.g., gyrB, 16 S rRNA, secA1, hsp65, and rpoB), enabling simultaneous and rapid comparison of multiple strains. MLSA-based species classification shows 95% concordance with identification by microscopic, biochemical, and cellular fatty acid analyses, while offering higher discriminatory power [31]. Notably, China accounted for the highest volume of mNGS-related publications (70%), which may correlate with its broader adoption in clinical practice and the subsequent higher diagnostic yield observed in species identification. The limited utility of blood cultures (only 8.57% of disseminated cases confirmed) underscores the importance of tissue sampling for definitive diagnosis. In addition to these techniques, new molecular techniques increasingly have emerged as pivotal tools for delineation of Nocardia species, such as real-time PCR-based high-resolution melting [32] and TaqMan real-time PCR assay [33]. Clinician vigilance for nocardiosis diagnosis is also critical. Even in immunocompetent patients, physicians must consider Nocardia infection when specific risk factors coexist, particularly underlying comorbidities (e.g., bronchiectasis, diabetes, or prolonged use of corticosteroid inhalers), significant environmental exposures (gardening, soil/organic dust handling, or contact with decaying plant material), or failed response to standard antibiotic therapy. Only through such clinical suspicion can targeted diagnostic measures be initiated: optimal specimen collection from relevant sites, specific laboratory testing requests, and explicit instructions for microbiological screening for Nocardia.

TMP-SMX remained the first-line therapy (71.9%), consistent with current recommendations establishing its role as the empirically supported frontline regimen for Nocardia infections [34]. Combination therapy was predominantly reserved for disseminated or severe Nocardia infections, with empirically selected companion drugs including third-generation cephalosporins (e.g., ceftriaxone), amikacin, and carbapenems (e.g., imipenem), particularly when addressing suspected multidrug-resistant strains or optimizing therapeutic penetration across infected tissues [35]. In our study, carbapenems (33.2%) and oxazolidinones (22.4%) accounted for a substantial proportion of therapeutic interventions in clinical practice. This observed preference may reflect clinicians’ concerns about resistance or the severity of infection. The association of carbapenems with CNS infections supports their role in managing life-threatening complications. Surgical intervention (13.8%) was crucial for resolving abscesses, particularly in the brain and soft tissues, underscoring the multimodal management approach required.

This systematic analysis is subject to selection bias, as data were derived from published literature rather than prospectively collected cohorts. Despite concerted efforts to include additional studies, certain case reports and series with incomplete research elements remained ineligible for inclusion. This unavoidable exclusion may introduce selection bias into our analysis. Our study remains constrained by significant language barriers: CNKI’s inclusion of Chinese publications explains China’s case overrepresentation, yet key languages (e.g., French/Spanish) remain uncovered, narrowing geographical scope despite comprehensiveness efforts. Missing data (e.g., AST results, detailed treatment regimens) may have introduced analytical limitations. Additionally, the exclusion of immunocompromised patients necessitates cautious extrapolation to broader populations. Besides, prospective studies are needed to validate risk factors and species-tropism associations.

In summary, this study enhances the understanding of Nocardia infections in immunocompetent hosts, while emphasizing the clinical need for early microbiological diagnosis, species-specific therapeutic approaches, and multidisciplinary care coordination in clinical management, which may add clinical alert to Nocardia infections in immunocompetent individuals, ultimately reducing morbidity and mortality associated with this neglected pathogen.

Author contributions

L. L designed the study, collected data and wrote the draft of the manuscript. W. D wrote the draft of manuscript, collected data and did the data analysis. L. M did data collection and data analysis. Y. X did the literature review and data collection. Z. K, Z. Z and L. C did data collection. S. W supervised and designed the study. All authors had full access to the data in the study, did the data interpretation, contributed to writing and revision of the final version of the manuscript, and had final responsibility for the decision to submit for publication.

Funding

Wang D was funded by the Postdoctoral Foundation of Jinling Hospital (Grant No. 49623). Yang X and Lu L were funded by the Jiangsu Province Cadre Healthcare Research Program (Grant No. BJ24026).

Data availability

No datasets were generated or analysed during the current study.

Declarations

Ethics approval and consent to participate

Not applicable.

AI usage statement

During the preparation of this work the author(s) used Deepseek and kimi in order to help literature review, text polishing and grammar optimization. After using this tool/service, the author(s) reviewed and edited the content as needed and take(s) full responsibility for the content of the publication.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.