Unique pathological features and drug resistance patterns in cutaneous tuberculosis

Cutaneous tuberculosis (CTB), a rare manifestation of extrapulmonary tuberculosis, often presents diagnostic challenges in clinical settings due to its atypical presentation. The definitive diagnosis relies heavily on pathological evaluation, underscoring the importance of understanding the distinct pathological characteristics and drug resistance patterns of CTB, which have not been extensively explored previously. This study conducted a comparative analysis of 59 CTB samples and 59 pulmonary tuberculosis samples, focusing on their clinicopathological features. Findings reveal that CTB can be characterized by subcutaneous irregular hypoechoic regions on ultrasound, localized soft tissue swelling, and flaky low-density shadows on CT scans, with MRI effectively determining the extent of bone and soft tissue involvement.

The study found no statistical difference in the positivity rate for acid-fast staining and molecular detection between CTB and pulmonary tuberculosis groups. Notably, the incidence of granulomatous lesions was higher in CTB compared to pulmonary tuberculosis, potentially due to a higher number of macrophages in the skin. However, other parameters such as caseous necrosis, coagulative necrosis, inflammatory necrosis, acute inflammation, hemorrhage, fibroplasia, and exudation showed no significant differences between the groups. Intriguingly, many significant differences in drug resistance patterns were observed between the CTB and control groups. When comparing the secondary CTB group to the control group, the only significant difference identified was in resistance to RFP + INH + STR. Overall, this study highlights unique pathological features and drug resistance profiles in CTB, providing valuable insights for more accurate clinical diagnosis and tailored therapeutic strategies.

Tuberculosis (TB), primarily caused by Mycobacterium tuberculosis, predominantly affects the lungs. However, the bacterium can invade other organs, leading to extrapulmonary tuberculosis (EPTB). Cutaneous tuberculosis (CTB) is notably rare, comprising only 1–1.5% of EPTB cases. CTB is typically acquired through hematogenous or lymphatic spread or direct contact. Its clinical presentation is influenced by factors such as the infecting strain’s load and pathogenicity, the route of infection, and the host’s immune status. Initial CTB lesions are often non-specific, manifesting as subcutaneous nodules, inflammation, ulcers, bleeding, and exudation, posing a diagnostic challenge.

This non-specificity makes CTB difficult to differentiate from bacterial infections, fungal infections, sarcoidosis, or tumors. The rarity of CTB further exacerbates the likelihood of misdiagnosis or delayed diagnosis, potentially leading to severe patient outcomes. Therefore, enhancing the accuracy of CTB diagnosis is of paramount clinical importance. Conventional diagnostic approaches include patient history, clinical symptom assessment, laboratory testing, and imaging studies. In suspected CTB cases, tests like tuberculin skin tests (TST) are employed, but TST results can be influenced by various factors and cannot distinguish between tuberculosis infection and immune response after BCG vaccination.

Pathological examination, including hematoxylin and eosin (H&E) staining, acid-fast staining, and molecular testing, remains the definitive diagnostic method. The hallmark pathological feature of TB is chronic granulomatous inflammation with caseous necrosis. In CTB, ulcers tend to persist longer than in pulmonary tuberculosis (PTB) and are frequently complicated by co-infections, adding to the complexity of the pathological profile. Prior to biopsy, patients often receive extensive local or systemic treatment, which can introduce additional non-specific pathological changes. Moreover, the skin’s unique immune microenvironment, rich in Langerhans cells and lymphocytes, contributes to further pathological distinctions between cutaneous and pulmonary TB.

The management of CTB necessitates a comprehensive and consistent approach to anti-tuberculosis medication, adhering strictly to the principles of early, appropriate, regular, and complete treatment. For patients with mild skin involvement, anti-TB pharmacotherapy alone is often sufficient. In cases where skin lesions significantly affect appearance or function of critical areas like the face or hands, surgical intervention may be considered for lesion removal and reparative procedures. However, surgery serves merely as an adjunct to medical therapy; the tuberculosis bacillus cannot be entirely eradicated through surgical means alone, necessitating the use of anti-TB drugs both pre- and post-operatively.

First-line anti-TB medications commonly employed include rifampicin (RFP), isoniazid (INH), pyrazinamide (PZA), streptomycin (STR), and ethambutol (EMB). The recently released endTB study results showed the non-inferiority of three fully oral, short-course regimens in the treatment of rifampin-resistant tuberculosis, all of which include PZA. While the overall incidence of tuberculosis has been declining annually, there has been a notable increase in the prevalence of drug-resistant and multidrug-resistant TB strains. This trend underscores the need for a deep understanding of drug resistance patterns in TB to tailor personalized treatment plans effectively and improve patient outcomes.

Skin samples, being more accessible and less invasive than samples from lungs or other vital organs, offer a unique opportunity for study. However, the literature lacks comprehensive reports on whether the drug resistance characteristics of CTB align with those of pulmonary tuberculosis and if these findings can guide TB treatment more broadly. This study investigated the clinical, pathological, and drug resistance features of CTB, analyzing the distinctions between CTB and pulmonary tuberculosis. These insights provide a crucial foundation for the clinical diagnosis and individualized treatment of this complex disease.

This study included patients admitted to the 8th Medical Center of Chinese PLA General Hospital from January 2012 to June 2023. Inclusion criteria were age over 18 years and clinical and pathological diagnosis of M. tuberculosis infection. Diseases like non-tuberculous mycobacteria (NTM), leprosy, and sarcoidosis can mimic TB characteristics. Cases with positive molecular detection of M. tuberculosis by PCR were included. Patients with malignant tumors, HIV or other immunodeficiency diseases, autoimmune diseases, severe fungal and bacterial infections, or hematologic malignancies were excluded, as these conditions can significantly affect pathological characteristics and immune status.

A total of 59 patients with CTB met these criteria. Among them, 33 patients had a history of TB at other sites, with secondary to pulmonary tuberculosis (16 cases) and pleural tuberculosis (7 cases) being the most prevalent. The remaining 26 cases were primarily cutaneous, with no TB at other sites. For the control group, 59 patients hospitalized during the same period with a pathological diagnosis of pulmonary tuberculosis (PTB) were selected. The methodology involved detailed histopathological staining and molecular testing procedures for sample analysis.

A 3 μm thick tissue section was cut from the sample for Hematoxylin and Eosin (H&E) staining. The section underwent baking, xylene treatment, and dehydration in a graded ethanol series. After washing, the section was stained with hematoxylin, differentiated with hydrochloric acid in ethanol, and washed again. It was then blued with ammonia water, washed, and stained with eosin. Following another wash, the section underwent further dehydration in graded ethanol, cleared in xylene, and mounted with neutral resin. This process allowed for detailed microscopic examination of tissue morphology.

For Ziehl-Neelsen staining, three 3 μm thick tissue sections were prepared. After initial preparation, these sections were processed similarly to the H&E staining method up to the washing post-ethanol dehydration. Subsequently, carbolic acid red dyeing solution was applied at room temperature for 2–3 hours. Decolorization was achieved using 1% hydrochloric acid in alcohol until a light pink color was attained, followed by a 30-second counterstain with hematoxylin. After further washing and differentiation steps, the sections were dehydrated through a graded ethanol series, cleared in xylene, and sealed with neutral resin to preserve the stained structures.

For strain identification, 8–10 tissue slices of 5–10 μm thickness were prepared from each sample and placed into 1.5 mL centrifuge tubes. The procedure involved dewaxing, cracking, digestion, and DNA extraction. In the PCR process, the Mycobacterium species identification gene test kit was used. Four microliters of sample DNA was added to the PCR tube. Amplification followed a specific protocol: 2 minutes at 50°C, 10 minutes at 95°C, then 30 cycles of 45 seconds at 95°C and 60 seconds at 68°C, and a final extension of 10 minutes at 68°C. Hybridization involved placing the membrane strip and amplified product in a test tube with solution A, boiling for 10 minutes, and then incubating at 59°C for 1.5 hours.

After washing with Solution B and incubating with a POD enzyme solution, the strips were developed in a color solution, shielded from light for 10 minutes, and terminated with purified or deionized water. The appearance of blue spots indicated detection, with both positive and negative controls included in each experiment. Figure 1 illustrates the sequence of the detection site on the membrane strip. This molecular method provided a sensitive and specific means of identifying mycobacterial species.

Drug resistance testing utilized the M. tuberculosis drug resistance mutation gene detection kit. Here, 4 μL of DNA was added to each PCR tube. The amplification process was followed by a similar hybridization and washing protocol as described for strain identification. Blue spots on the strips indicated the detection of drug resistance, and each test included positive and negative controls, as detailed by Li et al. (2022). This approach allowed for the rapid identification of genetic mutations associated with resistance to key anti-tuberculosis drugs.

Monoresistance refers to resistance to only one antituberculosis drug. Polyresistance refers to resistance to more than one antituberculosis drug but does not include resistance to RFP and INH. Multidrug resistance refers to resistance to at least RFP and INH simultaneously. Drug resistance to any drug refers to resistance to any one or more antituberculosis drugs, as depicted in Figure 2. No expression of drug resistance refers to the absence of detectable drug resistance genes, possibly due to low bacterial load. These classifications are crucial for understanding treatment challenges.

Data analysis was conducted using GraphPad Prism 5. Categorical data were presented as frequency (number of cases or strains) and proportion (percentage). The comparison among different groups was performed using the chi-square test or Fisher’s exact test, as appropriate. A p-value of less than 0.05 was considered indicative of statistical significance, ensuring robust conclusions from the collected data.

The control group comprised 59 pulmonary tuberculosis (PTB) patients, ranging in age from 19 to 78 years, with an average age of 51.78 years. This group included 34 males and 25 females. The CTB group also had 59 patients, aged between 19 and 88 years, with an average age of 45.71 years, including 37 males and 22 females. For the control group, samples were obtained via surgical excision or lung tissue puncture. CT scans revealed lobular and nodular high-density shadows, with cavity shadows observed in some lesions (Figure 3).

In the CTB group, the majority (52.54%) of samples were obtained through surgical excision, with 18 cases showing ulcer formation (Figures 4A,B). The sites of infection were predominantly on the trunk (64.41%), limbs (27.12%), and face and neck (8.47%). Ultrasound in CTB cases typically showed irregular hypoechoic areas under the skin, while CT scans indicated local soft tissue swelling and flaky low-density shadows (Figures 4C–E). In cases where M. tuberculosis spread to adjacent bone tissue or throughout the body, MRI revealed multiple changes in bone and soft tissue, including decreased T1WI signal and increased T2WI signal (Figures 4F,G).

In the control group, all 59 cases (100%) were positive for molecular detection, and 37 cases (62.71%) were positive for acid-fast staining. In the CTB group, all 59 cases (100%) were positive for molecular detection, with a higher positivity rate of 41 cases (69.49%) for acid-fast staining. There was no statistical difference in the positivity rate for acid-fast staining and molecular detection between the two groups, indicating the efficacy of these diagnostic methods across both forms of tuberculosis.

Pathologically, chronic granulomatous inflammation was most common in the control group, observed in 37 cases (62.71%), followed by caseous necrosis in 21 cases (35.59%), coagulative necrosis in seven cases (11.86%), inflammatory necrosis in 21 cases (35.59%), acute inflammation in 19 cases (32.20%), hemorrhage in three cases (5.08%), fibroplasia in seven cases (11.86%), and exudation in five cases (8.47%). In the CTB group, chronic granulomatous inflammation was again most common, present in 50 cases (84.75%), with caseous necrosis in 15 cases (25.42%), coagulative necrosis in nine cases (15.25%), inflammatory necrosis in 17 cases (28.81%), acute inflammation in 12 cases (20.34%), hemorrhage in one case (1.69%), fibroplasia in seven cases (11.86%), and exudation in two cases (3.39%).

Statistical analysis indicated a higher proportion of chronic granulomatous inflammation in the CTB group, but no significant differences in the proportions of other pathological features compared to the control group. There were no statistical differences in all pathological characteristics between secondary CTB and primary CTB. The results of immunohistochemistry indicated that the number of macrophages in the CTB group was over two times that of the control group, suggesting that macrophages play a significant role in the characteristic pathological changes associated with CTB. We also compared the number of macrophages and lymphocytes in normal tissues around the lesion and found that the number of macrophages and lymphocytes in the skin was significantly higher than that in the lung tissue (Figures 5, 6 and Table 1).

Clinically, skin biopsy is easier and less risky than lung puncture. Therefore, we compared the drug resistance characteristics to determine whether the drug resistance of secondary CTB can guide the treatment of PTB. In the control group, seven cases (11.86%) were sensitive to all four drugs tested, while 12 cases (20.34%) showed no expression of drug resistance. A total of 40 cases (67.80%) exhibited drug resistance to any drug, with monoresistance found in 13 cases (22.03%), predominantly against rifampicin (RFP) in nine cases. For multidrug resistance, 12 cases (20.34%) were resistant to both RFP and isoniazid (INH), and another 12 cases showed resistance to RFP, INH, and streptomycin (STR). Additionally, two cases (3.39%) displayed polyresistance, specifically to RFP and STR.

In the CTB group, only one case (1.69%) was sensitive to all four drugs, and a higher proportion of 36 cases (61.02%) showed no expression of drug resistance. Drug resistance to any drug was observed in 22 cases (37.29%), with monoresistance noted in eight cases (13.56%), primarily against INH. Among the multidrug resistance types, resistance to RFP and INH was the most common, observed in 11 cases. Only one case was found to be resistant to INH and STR, with no other types of polyresistance detected.

Statistical analysis revealed significant differences in drug resistance patterns between the CTB group and the control group. Compared to the control group, the CTB group had a higher proportion of cases with no expression of drug resistance and relatively lower proportions of drug resistance to any drug, RFP + INH + STR resistance, and RFP resistance. Additionally, when comparing the secondary CTB group to the control group, the only significant difference found was in resistance to RFP + INH + STR (Tables 2, 3). We further compared the differences between secondary CTB and primary CTB, and found that the incidence of INH resistance was higher in the secondary CTB group.

In the control group, mutations in the rifampicin (RFP) resistance gene were observed in 35 cases (59.32%). The D516V mutation was present in seven cases (11.86%), and single codon mutations including D516V, D516G, H526Y, H526D, and S531L were found in another seven cases (11.86%). Isoniazid (INH) resistance, detected in 28 cases (47.46%), frequently involved the -15M mutation. Streptomycin (STR) resistance was noted in 15 cases (25.42%), predominantly due to the 88M gene mutation. Ethambutol (EMB) resistance was characterized by the 306M2 mutation.

In the CTB group, there were 15 cases (25.42%) with RFP resistance gene mutations, primarily the D516V mutation. The distribution of other RFP resistance gene mutations was less pronounced. INH resistance was found in 20 cases (33.90%), all with the -15M mutation. STR resistance gene mutations were present in three cases (5.08%), with one case (1.69%) exhibiting the 88M gene mutation.

Among the 17 rpoB gene mutation classifications, the PTB control group had a significantly higher proportion of gene mutations compared to the CTB group. Notably, the D516V + D516G + H526Y + H526D + S531L mutation was present in 11.86% of the PTB control group but absent in the CTB group. The prevalence of STR resistance mutations and the 88M gene mutation was also higher in the PTB group compared to the CTB group. Further comparison between the secondary cutaneous TB group and the PTB control group revealed differences in RFP resistance gene mutations, D516V + D516G + H526Y + H526D + S531L site mutations, and STR resistance gene mutations (Tables 4, 5).

CTB, as a rare form of extrapulmonary tuberculosis, has seen a steady increase in its infection rate worldwide, garnering increasing attention from clinicians. Clinically, CTB often presents as ulceration and subcutaneous nodules, which can be challenging to distinguish from other skin diseases. This difficulty in differential diagnosis often leads to delayed identification, resulting in significant patient distress and potentially severe consequences. Pathology is an important diagnostic criterion for CTB, mainly characterized by granulomatous inflammation and caseous necrosis. The positive rate of acid-fast staining is higher in the CTB group, possibly due to the larger volume of biopsy specimens.

The advancement of molecular biology techniques has significantly bolstered the role of molecular detection in diagnosing and differentiating tuberculosis. This study indicates a high positive rate of molecular detection in both pulmonary and cutaneous tuberculosis cases. Additionally, molecular testing is crucial for differentiating CTB from infections caused by non-tuberculous mycobacteria. Consequently, for cases involving ulcers that are non-responsive to long-term treatment or have difficulty healing, timely skin biopsies and pathological examinations are recommended to establish a definitive diagnosis.

Numerous studies have underscored granulomatosis as a pivotal aspect of the immune pathogenesis in M. tuberculosis infection. The formation of granulomas, primarily composed of macrophages, is believed to be a response to the activation of the immune system upon infection. These macrophage-rich nodules act as a barrier, restricting the movement and proliferation of tuberculosis bacteria. This study corroborates this understanding, showing that chronic granulomatous inflammation is a predominant pathological feature in both CTB and PTB. Upon comparing the pathological characteristics of CTB and PTB, a relatively higher prevalence of granulomatous inflammation was observed in CTB.

However, other features such as caseous necrosis, coagulative necrosis, inflammatory necrosis, acute inflammation, hemorrhage, fibroplasia, and exudation did not exhibit significant differences between the two groups. This finding suggests that the variance in macrophage quantity might be a key factor influencing the distinct pathological features of CTB and PTB. Single-cell sequencing has revealed that macrophages, also known as Langhans giant cells, are the most abundant immune cells in the skin, playing a crucial role in maintaining the stability of the body’s immune barrier. The high concentration of macrophages in the skin could facilitate the effective containment and destruction of M. tuberculosis upon its invasion, potentially explaining the significantly lower incidence of CTB compared to PTB.

This observation highlights the importance of understanding the immune microenvironment in the skin and its role in the pathogenesis and clinical presentation of CTB. The lower incidence of CTB compared to other extrapulmonary tuberculosis forms may be partly attributed to the robust immune response in the skin, particularly the high concentration of macrophages. These immune cells become more active during infection, leading to a higher proportion of granulomatous inflammation. This response is especially pronounced when the immune system is compromised or when the bacterial load is excessive. Many cytokines, including TNF-α and IFN-γ, play crucial roles in granuloma formation, and some immunosuppressive molecules may also be important mechanisms.

Tuberculosis therapy faces a significant challenge with the rising incidence of drug-resistant tuberculosis. Many patients continue to experience inadequate responses or clinical deterioration, such as new lesions, persistent fever, weight loss, or relapse, even after prolonged standard anti-TB therapy. This study found a notable difference in the classification of “no expression of drug resistance” between the CTB group and the PTB control group. This category, characterized by no results in drug resistance determination areas or incomplete color development, may be due to a relatively low bacterial load in CTB samples, reducing the sensitivity and specificity of detection methods.

In comparing drug resistance in secondary CTB with the PTB control group, a significant difference was observed only in resistance to RFP + INH + STR. This finding suggests that secondary CTB foci may reflect the drug resistance profile of primary foci to some extent. D516V is located in the rifampicin resistance determinant region of the rpoB gene, which encodes the active center of the β subunit of RNA polymerase. RFP is the core drug in the treatment regimen for CTB, and its resistance significantly increases the difficulty of treatment and the risk of failure. Given the convenience and lower invasiveness of skin biopsies compared to lung punctures, skin biopsy in cases of concurrent pulmonary and cutaneous tuberculosis can provide valuable guidance for clinical medication.

In summary, CTB is characterized by a higher proportion of granulomatous lesions, and the high number of macrophages in the skin may be an important reason. The similarity in drug resistance profiles between secondary CTB and PTB offers a potential avenue for guiding personalized anti-tuberculosis treatment. Skin biopsy in cases of concurrent pulmonary and cutaneous tuberculosis can provide valuable guidance for clinical medication. This study thus contributes significant data for clinical diagnosis and the development of tailored treatment strategies.