Background: Tuberculosis (TB) and diabetes mellitus (DM) represent a growing syndemic of global health importance. Diabetes not only predisposes individuals to active TB but also alters its clinical, radiological, and pathological manifestations, thereby influencing treatment outcomes. Understanding these modifications is crucial for timely diagnosis and effective management. This study aimed to evaluate the pathological and radiological spectrum of pulmonary tuberculosis in association with diabetes mellitus and to establish correlations between these domains. Materials and Methods: This prospective, cross-sectional study was conducted at a tertiary care center in North Western India over 12 months and included 216 patients with microbiologically or histopathologically confirmed pulmonary TB. Patients were divided into two groups: Group A (TB–DM, n=108) and Group B (TB-only, n=108). Demographic, clinical, and glycemic profiles were recorded. All patients underwent chest radiography, with HRCT performed where indicated. Radiological features assessed included lesion site, cavity characteristics, extent of disease, and associated findings. Biopsy and cytology samples were examined histopathologically for granuloma architecture, necrosis, fibrosis, and acid-fast bacilli. Statistical analysis was performed using SPSS, with correlation coefficients applied to evaluate radiological–pathological associations. Results: TB–DM patients were significantly older (mean age 52.4 vs 39.7 years; p < 0.001) and presented more often with hemoptysis (35.2% vs 19.4%; p = 0.01) and prolonged illness (8.9 vs 6.2 weeks; p < 0.001). Radiologically, lower lobe involvement (38.9% vs 17.6%; p = 0.001), thick-walled cavitary lesions (36.1% vs 16.7%; p = 0.002), and multilobar disease (40.7% vs 21.3%; p = 0.003) were more frequent in TB–DM patients. Pathologically, poorly formed granulomas (57.4% vs 34.3%; p = 0.001) and extensive caseous necrosis (65.7% vs 44.4%; p = 0.002) predominated, while healing fibrosis was reduced (17.6% vs 38.0%; p = 0.001). Radiological–pathological correlation showed strong associations between thick-walled cavities and necrosis (r = 0.62; p < 0.01), multilobar disease and poorly formed granulomas (r = 0.48; p = 0.02), and lower lobe lesions with impaired fibrotic response (r = 0.41; p = 0.03). Treatment outcomes were inferior in TB–DM patients, with lower sputum conversion at 2 months (66.7% vs 84.3%; p = 0.002) and higher relapse (13.0% vs 4.6%; p = 0.03). Conclusion: Diabetes mellitus significantly modifies the clinical, radiological, and pathological presentation of pulmonary tuberculosis and adversely impacts treatment outcomes. TB–DM patients are more likely to exhibit atypical lower lobe involvement, thick-walled cavitary lesions, disorganized granulomas, and extensive necrosis. Integrated pathological and radiological evaluation is essential for accurate diagnosis and tailored management in this high-risk population.
Tuberculosis (TB) continues to be one of the leading causes of morbidity and mortality worldwide, with the World Health Organization (WHO) reporting over 10 million new cases annually. Despite global control efforts, TB remains a pressing public health concern, particularly in low- and middle-income countries.1,2 At the same time, the global prevalence of diabetes mellitus (DM) has risen dramatically, largely driven by lifestyle transitions and urbanization, with an estimated 537 million adults affected worldwide in 2021. The intersection of these two epidemics has emerged as a major challenge, giving rise to what is now referred to as the “TB–diabetes syndemic.”3,4
Epidemiological studies highlight the bidirectional impact of these conditions. DM increases the risk of developing active TB by approximately two- to three-fold compared to non-diabetic individuals. Recent WHO estimates suggest that nearly 400,000 TB cases globally are attributable to diabetes. 5,6The pooled global prevalence of DM among TB patients has been reported at 13–16%, with marked geographic variation—ranging from as low as 4% in some African nations to over 40% in high-burden Pacific regions. Conversely, TB prevalence among diabetic populations is estimated at approximately 4%, underscoring the bidirectional disease burden. Importantly, the coexistence of TB and DM has been linked with unfavorable outcomes, including delayed sputum conversion, higher relapse rates, increased mortality, and a greater risk of multidrug-resistant TB.7-9
Pathologically, hyperglycemia-induced immune dysregulation alters the host response to Mycobacterium tuberculosis. Diabetic patients often demonstrate impaired granulomatous response, extensive caseous necrosis, increased cavitary lesions, and delayed fibrotic healing. Radiologically, TB in association with DM may present with atypical features compared to classical pulmonary TB.10-12 While upper lobe cavitation remains a hallmark, diabetic patients frequently exhibit lower lung field involvement, multilobar consolidation, thick-walled cavities, and extensive parenchymal destruction. Such radiological patterns often mimic other pulmonary conditions, complicating diagnosis and delaying treatment initiation.13,14
The integration of pathological and radiological assessment provides a comprehensive understanding of TB–DM comorbidity, allowing improved diagnostic precision, better prognostic stratification, and tailored treatment strategies. However, systematic studies correlating pathological features with radiological findings in this subgroup remain limited.
Therefore, the present study aims to evaluate the pathological and radiological characteristics of tuberculous lesions in patients with diabetes mellitus, and to establish correlations that may enhance diagnostic accuracy and clinical management.
Study Design and Setting
This prospective, observational, cross-sectional study was conducted in the Department of Pathology and Radiology, in collaboration with the Department of Pulmonary Medicine, at tertiary care centre in in North Western India. The study was carried out over a period of 12 months from Jan 2024 to Dec 2024.
Study Population
The study population consisted of patients with microbiologically or histopathologically confirmed pulmonary tuberculosis, with or without coexisting diabetes mellitus. Adult patients aged 18 years and above were considered eligible. The study population was stratified into two groups: Group A comprised patients with tuberculosis and diabetes mellitus (TB–DM), and Group B comprised patients with tuberculosis but without diabetes mellitus (TB-only).
Inclusion and Exclusion Criteria
Patients with newly diagnosed or previously untreated pulmonary TB, confirmed by sputum smear microscopy, culture, GeneXpert, or histopathology, were included. Those with known diabetes mellitus, diagnosed according to the American Diabetes Association (ADA) criteria (fasting plasma glucose ≥126 mg/dL, HbA1c ≥6.5%, or patients on antidiabetic therapy), were considered in the TB–DM group. Patients with extrapulmonary tuberculosis, HIV/AIDS, other immunosuppressive disorders, or chronic pulmonary diseases such as chronic obstructive pulmonary disease (COPD) or interstitial lung disease were excluded. Patients unwilling to provide informed consent were also excluded.
Sample Size
The sample size was calculated using the standard formula:
n=Z2×p×(1−p)/d2
where Z is the standard normal deviate (1.96 at 95% confidence level), p is the estimated prevalence of diabetes among tuberculosis patients (15% based on global data), and d is the allowable margin of error (5%). The minimum sample size required was 196. To account for possible dropouts and incomplete data, an additional 10% was added, resulting in a final sample size of 216 patients.
Data Collection and Clinical Evaluation
Demographic data including age, sex, and socioeconomic background, along with clinical history such as duration of symptoms, prior treatment, and associated comorbidities, were collected through a structured proforma. Glycemic status was assessed using fasting plasma glucose and HbA1c levels at the time of enrollment.
Radiological Assessment
All participants underwent chest radiography in the posteroanterior (PA) view, and high-resolution computed tomography (HRCT) was performed where indicated. Radiological parameters evaluated included the site of involvement (upper, middle, or lower lobes), type of lesion (cavity, consolidation, nodule, or miliary pattern), disease extent (unilateral, bilateral, or multilobar), and additional findings such as cavity wall thickness, bronchogenic spread, mediastinal lymphadenopathy, and pleural effusion. All images were reviewed independently by two experienced radiologists blinded to the diabetic status of patients.
Pathological Assessment
Biopsy specimens obtained either bronchoscopically or by CT-guided procedure, along with sputum cytology, were subjected to histopathological evaluation. Hematoxylin and eosin (H&E) staining was used to assess tissue architecture, with specific focus on granuloma formation, caseous necrosis, inflammatory infiltrates, and fibrosis. Ziehl–Neelsen (ZN) staining was performed for identification of acid-fast bacilli (AFB). The pathological findings were compared between the TB–DM and TB-only groups.
Statistical Analysis
All collected data were entered into Microsoft Excel and analyzed using SPSS version 25 (IBM Corp., Armonk, NY, USA). Continuous variables were expressed as mean ± standard deviation (SD) and compared using the Student’s t-test. Categorical variables were expressed as frequencies and percentages, and analyzed using the Chi-square test or Fisher’s exact test, as appropriate. A p-value <0.05 was considered statistically significant.
Ethical Considerations
The study protocol was reviewed and approved by the Institutional Ethics Committee. Written informed consent was obtained from all participants before enrollment. Confidentiality of patient data was maintained at all times, and the study was conducted in accordance with the principles outlined in the Declaration of Helsinki.
A total of 216 patients with pulmonary tuberculosis were enrolled in the study. Of these, 108 (50%) had coexisting diabetes mellitus (Group A: TB–DM), and 108 (50%) had tuberculosis without diabetes (Group B: TB-only).
The mean age of patients in the TB–DM group was significantly higher compared to the TB-only group (52.4 ± 11.6 years vs 39.7 ± 13.4 years; p < 0.001). The male-to-female ratio was slightly higher in TB–DM patients (1.9:1) compared to TB-only patients (1.6:1), though the difference was not significant. Urban residence was more common among TB–DM cases, while undernutrition (BMI <18.5) was more frequent in TB-only cases, without statistical significance (Table 1).
Table 1. Demographic Characteristics of Study Population (N = 216)
Characteristic |
Group A (TB–DM) (n=108) |
Group B (TB-only) (n=108) |
p-value |
Mean Age (years ± SD) |
52.4 ± 11.6 |
39.7 ± 13.4 |
<0.001* |
Male : Female Ratio |
1.9 : 1 |
1.6 : 1 |
0.42 |
Urban Residence (%) |
68 (63.0%) |
59 (54.6%) |
0.21 |
BMI <18.5 (Underweight) |
25 (23.1%) |
34 (31.5%) |
0.18 |
Cough was the most common presenting symptom in both groups. Hemoptysis was significantly higher in TB–DM patients (35.2% vs 19.4%; p = 0.01), while fever was more frequent in TB-only patients (82.4% vs 68.5%; p = 0.02). The mean duration of illness was longer in TB–DM patients (8.9 ± 3.2 weeks) compared to TB-only patients (6.2 ± 2.8 weeks; p < 0.001) (Table 2).
Table 2. Clinical Presentation in TB–DM vs TB-only Patients
Clinical Feature |
Group A (TB–DM) (n=108) |
Group B (TB-only) (n=108) |
p-value |
Cough > 4 weeks (%) |
92 (85.2%) |
95 (88.0%) |
0.56 |
Hemoptysis (%) |
38 (35.2%) |
21 (19.4%) |
0.01* |
Fever (%) |
74 (68.5%) |
89 (82.4%) |
0.02* |
Weight loss (%) |
81 (75.0%) |
85 (78.7%) |
0.53 |
Duration of Illness (weeks) |
8.9 ± 3.2 |
6.2 ± 2.8 |
<0.001* |
Radiographic evaluation revealed that upper lobe involvement was more common in TB-only patients (70.4% vs 54.6%; p = 0.02), whereas lower lobe involvement was significantly more frequent in TB–DM patients (38.9% vs 17.6%; p = 0.001). Cavitary lesions, particularly thick-walled cavities, were observed more frequently in TB–DM patients (36.1% vs 16.7%; p = 0.002). Multilobar disease was also significantly higher in TB–DM patients (40.7% vs 21.3%; p = 0.003) (Table 3).
Table 3. Radiological Findings in TB–DM vs TB-only Patients
Radiological Feature |
Group A (TB–DM) (n=108) |
Group B (TB-only) (n=108) |
p-value |
Upper Lobe Involvement (%) |
59 (54.6%) |
76 (70.4%) |
0.02* |
Lower Lobe Involvement (%) |
42 (38.9%) |
19 (17.6%) |
0.001* |
Cavitary Lesions (%) |
67 (62.0%) |
48 (44.4%) |
0.01* |
Thick-walled Cavities (%) |
39 (36.1%) |
18 (16.7%) |
0.002* |
Multilobar Involvement (%) |
44 (40.7%) |
23 (21.3%) |
0.003* |
Pleural Effusion (%) |
15 (13.9%) |
11 (10.2%) |
0.40 |
Histopathological examination revealed significant variation between groups. Poorly formed granulomas were more frequent in TB–DM patients (57.4% vs 34.3%; p = 0.001), whereas well-formed granulomas were common in TB-only patients (65.7% vs 42.6%; p = 0.001). Extensive caseous necrosis was significantly higher in TB–DM patients (65.7% vs 44.4%; p = 0.002). Fibrosis and healing response was more pronounced in TB-only patients (38.0% vs 17.6%; p = 0.001) (Table 4).
Table 4. Pathological Findings in TB–DM vs TB-only Patients
Pathological Feature |
Group A (TB–DM) (n=108) |
Group B (TB-only) (n=108) |
p-value |
Well-formed Granulomas (%) |
46 (42.6%) |
71 (65.7%) |
0.001* |
Poorly-formed Granulomas (%) |
62 (57.4%) |
37 (34.3%) |
0.001* |
Extensive Caseous Necrosis (%) |
71 (65.7%) |
48 (44.4%) |
0.002* |
Fibrosis/Healing Response (%) |
19 (17.6%) |
41 (38.0%) |
0.001* |
AFB Positivity (ZN stain) (%) |
82 (75.9%) |
76 (70.4%) |
0.36 |
Correlation analysis revealed a strong positive association between thick-walled cavities and extensive caseous necrosis (r = 0.62; p < 0.01). Multilobar disease demonstrated a moderate correlation with poorly formed granulomas (r = 0.48; p = 0.02). Lower lobe involvement correlated modestly with delayed fibrotic response (r = 0.41; p = 0.03). Conversely, upper lobe predominance showed a weaker but significant correlation with well-formed granulomas (r = 0.36; p = 0.04) (Table 5).
Table 5. Correlation Between Radiological and Pathological Findings in TB–DM Patients
Radiological Feature |
Pathological Feature |
Correlation Coefficient (r) |
p-value |
Thick-walled cavities |
Extensive caseous necrosis |
0.62 |
<0.01* |
Multilobar disease |
Poorly-formed granulomas |
0.48 |
0.02* |
Lower lobe involvement |
Delayed fibrotic response |
0.41 |
0.03* |
Upper lobe predominance |
Well-formed granulomas |
0.36 |
0.04* |
Treatment response was inferior among TB–DM patients. Sputum conversion at 2 months occurred in 66.7% of TB–DM patients compared to 84.3% in TB-only patients (p = 0.002). Treatment completion was also lower in TB–DM patients (82.4% vs 93.5%; p = 0.02). Relapse within 6 months was significantly higher in the TB–DM group (13.0% vs 4.6%; p = 0.03). Mortality was higher in TB–DM patients (6.5% vs 1.9%), though not statistically significant (p = 0.09) (Table 6).
Table 6. Treatment Outcomes in TB–DM vs TB-only Patients
Outcome Measure |
Group A (TB–DM) (n=108) |
Group B (TB-only) (n=108) |
p-value |
Sputum Conversion at 2 months |
72 (66.7%) |
91 (84.3%) |
0.002* |
Treatment Completed (%) |
89 (82.4%) |
101 (93.5%) |
0.02* |
Relapse within 6 months (%) |
14 (13.0%) |
5 (4.6%) |
0.03* |
Mortality (%) |
7 (6.5%) |
2 (1.9%) |
0.09 |
This study examined the pathological and radiological spectrum of pulmonary tuberculosis in patients with and without diabetes mellitus and explored correlations between these two domains. Our findings demonstrate that diabetes significantly modifies both the clinical presentation and the morphological features of tuberculosis. Specifically, diabetic patients exhibited a higher frequency of hemoptysis, prolonged symptom duration, lower lung involvement, thick-walled cavitary lesions, poorly formed granulomas, and extensive caseous necrosis. Radiological–pathological correlation analysis revealed strong associations between thick-walled cavities and extensive necrosis, multilobar disease and poorly formed granulomas, as well as lower lobe lesions and impaired fibrotic response.
The demographic distribution in our study showed that TB–DM patients were significantly older than TB-only patients, with a mean age difference of more than a decade. This finding is consistent with multiple large-scale epidemiological studies, which have established that the prevalence of diabetes increases with advancing age and thereby influences the age profile of TB patients.15-18 A meta-analysis from Asia and Latin America demonstrated that TB–DM comorbidity is disproportionately concentrated in older adults, reflecting the epidemiological overlap between the peak incidence of type 2 diabetes and susceptibility to active tuberculosis.19,20 Similar results were reported in a multicentric Indian study, where the mean age of TB–DM patients was above 50 years compared to less than 40 years for TB-only cohorts.21,22 This pattern emphasizes the need for heightened clinical suspicion for tuberculosis in elderly diabetics presenting with respiratory symptoms, even when classical features may be absent.
Clinically, our cohort demonstrated a significantly higher frequency of hemoptysis and a prolonged duration of illness among diabetic patients. These findings align with studies from China and Latin America, where hemoptysis has been documented as a common presentation in TB–DM due to extensive cavitation and vascular damage in the lung parenchyma.23-26 The prolonged symptom duration in diabetics can be explained by delayed recognition of atypical presentations, as well as by the impaired innate immune responses in hyperglycemic states. Hyperglycemia is known to reduce phagocytic activity, impair chemotaxis, and alter cytokine profiles, thereby delaying effective containment of Mycobacterium tuberculosis. This impaired host defense contributes to progressive lung tissue destruction, which in turn accounts for both the chronicity of symptoms and the higher frequency of hemoptysis observed in our patients.
Radiologically, our findings of increased lower lobe involvement and multilobar disease in TB–DM patients are in concordance with previous observations by Pérez-Guzmán and colleagues, who first highlighted the atypical localization of lesions in diabetic populations.27 While upper lobe predominance remains the hallmark of pulmonary tuberculosis in non-diabetics, diabetic patients frequently demonstrate disease in the middle and lower lung zones, which may mimic bacterial pneumonia or malignancy and lead to diagnostic confusion.28,29 Chiang et al. similarly noted that diabetics were more likely to have multilobar lesions and extensive parenchymal involvement compared to non-diabetics, underscoring the aggressive radiological profile of TB in this subgroup.30 Our observation of thick-walled cavitary lesions in a significantly higher proportion of diabetics also corroborates CT-based studies from South Korea and Mexico, which reported that cavities in TB–DM are often larger, denser, and more resistant to healing compared to those in TB-only patients.31-33 These findings reflect not only the more destructive nature of the disease in diabetics but also its potential to contribute to ongoing transmission in the community due to persistent bacillary loads.
Pathologically, our study demonstrated a higher prevalence of poorly formed granulomas and extensive caseous necrosis in TB–DM patients, findings that have been repeatedly documented in experimental and clinical studies. The disorganized granulomas observed in diabetics result in impaired containment of bacilli and increased progression to cavitary disease. The predominance of extensive caseous necrosis in our cohort echoes histopathological reports from India and Latin America, where diabetic TB patients showed a higher burden of necrotizing granulomas compared to their non-diabetic counterparts.34-36 Furthermore, the reduced fibrosis and healing response in TB–DM patients in our study highlights the impaired reparative processes in hyperglycemic states, which may predispose to chronic lesions, delayed sputum conversion, and higher relapse rates. This is supported by immunological evidence showing that advanced glycation end-products (AGEs) and oxidative stress interfere with fibroblast function and collagen deposition, leading to delayed healing in diabetic individuals.
Taken together, our findings are consistent with the global literature in establishing that diabetes modifies the clinical course, radiological spectrum, and histopathological architecture of tuberculosis. By confirming these associations in our population, this study adds to the growing body of evidence that TB–DM represents a distinct clinical phenotype of tuberculosis, with diagnostic, therapeutic, and prognostic implications.
Hyperglycemia has been shown to impair both innate and adaptive immunity, including phagocytic activity, chemotaxis, and cytokine production. This contributes to ineffective containment of Mycobacterium tuberculosis and greater tissue destruction. The correlation between radiological and pathological findings in our study reinforces this pathophysiology: thick-walled cavities were strongly associated with extensive necrosis, while multilobar disease corresponded with disorganized granulomatous response. Such associations highlight how systemic metabolic dysfunction translates into both histological and imaging features.
Our findings have significant implications for clinical practice. Recognition of atypical radiological features in diabetic patients is essential for timely diagnosis, as lower lobe involvement may mimic bacterial pneumonia or malignancy. Pathological confirmation remains crucial in such cases to avoid misdiagnosis and treatment delay. Furthermore, the inferior treatment outcomes observed in TB–DM patients—including lower sputum conversion rates and higher relapse rates—underscore the need for integrated management strategies. Intensified glycemic control, close radiological monitoring, and potentially longer or modified anti-tubercular regimens may be necessary to improve outcomes in this high-risk group.
From a public health perspective, the growing global prevalence of diabetes poses a significant challenge to TB control programs. The WHO has already recognized the TB–DM syndemic as a critical threat, with approximately 400,000 TB cases annually attributable to diabetes. Strengthening bidirectional screening programs, where TB patients are tested for diabetes and diabetic patients are screened for TB, is a vital step toward early detection and comprehensive management.
The strengths of this study include a well-defined sample size, systematic radiological and pathological evaluation, and the novel use of correlation analysis to link imaging and histological findings. However, certain limitations must be acknowledged. Being a single-center study, the findings may not be generalizable to all populations. The exclusion of extrapulmonary TB limits the scope of applicability. Additionally, although we identified correlations, causality cannot be firmly established due to the cross-sectional design. Longitudinal studies with larger cohorts and multi-center collaboration are warranted to validate these observations.
In summary, this study highlights that diabetes mellitus significantly alters the clinical, radiological, and pathological spectrum of pulmonary tuberculosis. TB–DM patients are more likely to present with lower lobe lesions, thick-walled cavities, poorly formed granulomas, and extensive necrosis, with strong correlations between radiological and pathological features. These changes contribute to poorer treatment outcomes compared to non-diabetic TB patients. The findings underscore the importance of integrated pathological and radiological evaluation in TB–DM patients and emphasize the urgent need for multidisciplinary strategies to address this dual burden at both clinical and public health levels.