TYPE Original Research PUBLISHED 17 November 2022 DOI 10.3389/fpubh.2022.1017024 Glycemic control and the risk of OPEN ACCESS tuberculosis in patients with diabetes: A cohort study in a EDITED BY Marija Petek Šter, University of Ljubljana, Slovenia REVIEWED BY Enrico Girardi, Mediterranean city National Institute for Infectious Diseases Lazzaro Spallanzani (IRCCS), Italy Violeta Antonio-Arques1,2 , Joan A. Caylà3 , Jordi Real1,4 , Paola Del Carmen Guerra De Blas, Instituto Nacional de Ciencias Médicas Antonio Moreno-Martinez5,6 , Àngels Orcau6,7 , y Nutrición Salvador Zubirán (INCMNSZ), Mexico Didac Mauricio1,4,8,9 , Manel Mata-Cases1,4,10 , Josep Julve4,11 , Anabel Ordaz Vazquez, Elena Navas Mendez12 , Rai Puig Treserra1 , Joan Pau Millet6,7 , Instituto Nacional de Ciencias Médicas y Nutrición Salvador Zubirán Jose Luis Del Val García12,13 , Bogdan Vlacho1* and (INCMNSZ), Mexico Josep Franch-Nadal1,4,14 *CORRESPONDENCE 1 Bogdan Vlacho DAP-Cat Group, Unitat de Suport a la Recerca Barcelona, Fundació Institut Universitari per a la bogdan.vlacho@gmail.com Recerca a l’Atenció Primària de Salut Jordi Gol i Gurina (IDIAPJGol), Barcelona, Spain, 2 Primary Health Care Center La Ràpita - Alcanar, Gerència d’Atenció Primària Terres de l’Ebre, Institut Català SPECIALTY SECTION de la Salut, Tortosa, Spain, 3 Tuberculosis Research Unit Foundation of Barcelona, Barcelona, Spain, This article was submitted to 4 CIBER of Diabetes and Associated Metabolic Diseases (CIBERDEM), Instituto de Salud Carlos III, Family Medicine and Primary Care, Madrid, Spain, 5 Department of Infectious Diseases, Hospital Clínic de Barcelona, Barcelona, Spain, a section of the journal 6 CIBER of Epidemiology and Public Health (CIBERESP), Instituto de Salud Carlos III (ISCIII), Madrid, Frontiers in Public Health Spain, 7 Epidemiology Service, Agència de Salut Pública de Barcelona, Barcelona, Spain, 8 RECEIVED 11 August 2022 Department of Endocrinology and Nutrition, Hospital Universitari de la Santa Creu i Sant Pau, ACCEPTED 26 October 2022 Barcelona, Spain, 9 Department of Medicine, University of Vic—Central University of Catalonia, PUBLISHED 17 November 2022 Barcelona, Spain, 10 Primary Health Care Center La Mina, Gerència d’Atenció Primària Barcelona Ciutat, Institut Català de la Salut, Barcelona, Spain, 11 Department of Biochemistry, Institut de CITATION Recerca de l’Hospital de la Santa Creu i Sant Pau, Barcelona, Spain, 12 Fundació Institut Universitari Antonio-Arques V, Caylà JA, Real J, per a la Recerca a l’Atenció Primària de Salut Jordi Gol i Gurina (IDIAPJGol), Barcelona, Spain, Moreno-Martinez A, Orcau À, 13 Unitat d’Avaluació, Sistemes d’informació i Qualitat, Gerència d’Àmbit d’Atenció Primària Mauricio D, Mata-Cases M, Julve J, Barcelona Ciutat, Institut Català de la Salut, Barcelona, Spain, 14 Primary Health Care Center Raval Navas Mendez E, Puig Treserra R, Sud, Gerència d’Atenció Primària Barcelona Ciutat, Institut Català de la Salut, Barcelona, Spain Millet JP, Del Val García JL, Vlacho B and Franch-Nadal J (2022) Glycemic control and the risk of tuberculosis in patients with diabetes: A cohort study Background: Diabetes mellitus (DM) is one of the leading chronic diseases in a Mediterranean city. globally and one of the most common causes of death, morbidity, and poor Front. Public Health 10:1017024. doi: 10.3389/fpubh.2022.1017024 quality of life. According to the WHO, DM is also one of the main risk factors for developing active tuberculosis (TB). Subjects with DM are at a higher risk COPYRIGHT © 2022 Antonio-Arques, Caylà, Real, of infections, in addition to frequent micro and macrovascular complications, Moreno-Martinez, Orcau, Mauricio, and therefore sought to determine whether poor glycemic control is linked to Mata-Cases, Julve, Navas Mendez, Puig Treserra, Millet, Del Val García, a higher risk of developing TB. Vlacho and Franch-Nadal. This is an Methods: We used a retrospective cohort of diabetic subjects to predict the open-access article distributed under the terms of the Creative Commons incidence of TB. All DM patients were recruited from Ciutat Vella (the inner- Attribution License (CC BY). The use, city of Barcelona) from January 2007 until December 2016, with a follow-up distribution or reproduction in other forums is permitted, provided the period until December 2018 (≥2 years). Data were extracted from Barcelona’s original author(s) and the copyright Primary Care medical record database - SIDIAP, and linked to the Barcelona TB owner(s) are credited and that the Control Program. The incidence of TB and the impact of glycemic control were original publication in this journal is cited, in accordance with accepted estimated using time-to-event curves analyzed by Cox proportional hazard academic practice. No use, distribution regression. Hazard ratios (HRs) and 95% confidence intervals (CIs), unadjusted or reproduction is permitted which and adjusted by potential confounding variables, were also assessed, which does not comply with these terms. included age, sex, diabetes duration, macrovascular and microvascular signs, BMI, smoking habit, alcohol consumption and geographical origin. Frontiers in Public Health 01 frontiersin.org Antonio-Arques et al. 10.3389/fpubh.2022.1017024 Results: Of 8,004 DM patients considered for the study (equating to 68,605 person-years of follow-up), 84 developed TB [incidence rate = 70 (95% CI: 52–93) per 100,000 person-years]. DM subjects with TB were younger (mean: 52.2 vs. 57.7 years old), had higher values of glycosylated hemoglobin (HbA1c) (7.66 vs. 7.41%) and total triglycerides (122 vs. 105 mg/dl), and had twice the frequency of diabetic nephropathy (2.08 vs. 1.18%). The calculated incidence rate increased with increasing HbA1c: 120.5 (95% CI 77.2–179.3) for HbA1c ≥ 7.5%, 143 (95% CI 88.3–218.1) for HbA1c ≥ 8% and 183.8 (95% CI 105–298) for HbA1c ≥ 9%. An increase in the risk of TB was also observed according to a poorer optimization of glycemic control: adjusted HR 1.80 (95% CI 0.60–5.42), 2.06 (95% CI 0.67–6.32), and 2.82 (95% CI 0.88–9.06), respectively. Conclusion: Diabetic subjects with worse glycemic control show a trend toward a higher risk of developing TB. KEYWORDS comorbidities, diabetes mellitus, diabetes complications, glycemic control, Primary Health Care, social determinants of health, tuberculosis Introduction cavitary pulmonary TB and hospitalization (17) and more risk of treatment failure, death, relapse and multidrug resistant TB Diabetes mellitus (DM) is one of the leading chronic diseases (MDR-TB) (18). in the world (1) and one of the most common causes of death Both DM and TB can be considered syndemic conditions (2). Diabetes-associated mortality particularly impacts patients (19). Countries with an increasing prevalence of DM also have in low and middle-income countries due to poor control of the a higher incidence of TB, especially seen in South-East Asia (2). disease. Type 2 DM (T2DM) is the most frequent type of DM, Therefore, this clinical issue has become a priority for the World accounting for ∼90% of DM worldwide (3). Health Organization (WHO) in recent years (20). Focusing on In patients with DM, chronic hyperglycemia underlies Europe, although most countries have a low TB incidence, inner- the development of macrovascular and microvascular city districts of big cities, usually disadvantaged urban areas, complications, these complications are associated with the show a higher notification rate (21). greatest burden on patients, caregivers and the health system (4– In the current study, we used data from the diabetic 6). The aim of DM management is to prevent the development population of Ciutat Vella, an inner-city district of Barcelona of these complications by achieving and maintaining glycemic that has a three times higher incidence of TB than other city control. A legacy effect is already known, i.e., early intensive districts (22). optimization of glycemic control is frequently associated with a We aimed to estimate the incidence and characteristics of TB favorable reduction in the risk of myocardial infarction, death in patients with DM according to different HbA1c thresholds to from any cause and microvascular disease, with the benefits analyze if higher values of HbA1c are linked to a higher risk of of improved glycemic control in newly diagnosed patients developing TB. maintained over time (7). Besides vascular complications, DM has also been linked to an increased incidence of infections (8, 9), especially those related to lower respiratory, gastrointestinal, and urinary Materials and methods tract infections, and bacterial and mycotic skin and mucous membrane infections (8, 10). Current evidence suggests that DM Study design patients with a poorer glycemic control are at an increased risk of these infections and consequent hospitalization and mortality We performed a subanalysis from a retrospective (11, 12). In addition to the infections mentioned above, DM is observational cohort. The methodology related to the primary considered an important risk factor for developing tuberculosis data has been previously published (23). In the present study, (TB) (2, 13), and poor glycemic control has been associated with we selected a cohort of diabetes patients registered between a higher risk of developing this infection and worse outcomes January 1st, 2007 and December 31st, 2016, with a follow-up (14–16). Moreover, patients with DM have an increased risk of period until December 2018. The inclusion date was January Frontiers in Public Health 02 frontiersin.org Antonio-Arques et al. 10.3389/fpubh.2022.1017024 FIGURE 1 Flow chart of the DM cohort within the study population. 1st, 2007, but for those participants who became diabetic during characteristics and specific data on DM were extracted from the the study period the date of diagnosis was used. Primary Care medical record database from Barcelona. Specific data on TB were obtained from the Barcelona TB Prevention and Control Program. Both sources were linked through Study population and sample an anonymized unique identifier. Patients’ informed consent was unnecessary because the databases consisted of pseudo- The primary study was carried out in the district of Ciutat anonymized data. This study obtained an Ethics Committee Vella, the inner city of Barcelona, with a population of 108,000, approval in 2016 (code P16/023). that is characterized by a lower socioeconomic level and a higher percentage of immigrants (50.1%), compared to the rest of the city (24), and also has the highest incidence of infectious diseases (25). A cohort of 8,004 subjects with diabetes was analyzed in this Variables study (Figure 1). The primary outcome was defined as the occurrence and date of a TB diagnosis during the follow-up period. Inclusion and exclusion criteria We obtained the following information on the subjects at baseline: (1) demographic characteristics, smoking Patients over 18 years who fulfilled the criteria for a DM habits, and excessive alcohol consumption [defined by the diagnosis were included. People younger than 18 years or recorded International Classification of Disease 10 (ICE.10) without appointments registered in their Primary Health Care F10.1], (2) clinical variables, (3) laboratory data, (4) history Centre during the study period were excluded. of comorbidities. When analyzing the country of origin, those patients from South-East Asia (Pakistan, Bangladesh, and India) (24) were Data sources grouped in one category: Hindustani origin. We considered subjects to have DM if they had an ICD-10 All data were treated as confidential according to the ethical (International Classification of Diseases) diagnostic code for principles of the Helsinki Declaration of 1964, revised by the DM in their medical record (E10–E14) or if they were taking World Medical Organization in Fortaleza, Brazil, in 2013, and any class of antidiabetic drug other than metformin (sometimes the Spanish Organic Law 3/2018 of Data Protection. Clinical used to treat other conditions). Diabetes-related information Frontiers in Public Health 03 frontiersin.org Antonio-Arques et al. 10.3389/fpubh.2022.1017024 TABLE 1 Baseline characteristics of the study variables in diabetic patients without and with the occurrence of tuberculosis during the follow-up period. Variable Type 2 DM without TB Type 2 DM with TB p-values (n = 7,956) (n = 48) Age (years) mean (SD) 57.7 (14.2) 52.2 (13.4) 0.007a Gender: n (%) 0.353b Men 4,865 (61.1%) 33 (68.8%) Women 3,091 (38.9%) 15 (31.2%) Origin 0.008b Spain/high-income countries 3,172 (39.8%) 17 (35.4%) Hindustan 1,069 (13.4%) 14 (29.2%) Other 965 (12.1%) 7 (14.6%) Unknown 2,750 (34.6%) 10 (20.8%) Alcohol abuse 398 (5.0%) 3 (6.25%) 0.733b Smoking habit 587 (7.38%) 1 (2.08%) 0.260b Homeless 6 (5.94%) 3 (6.25) 0.837b Unemployment 9 (14.5%) 15 (32.6%) <0.001b High social risk (Medea > 2.3) 50 (42.0%) 12 (34.3%) 0.541b Duration of DM (years) mean (SD) 3.22 (5.77) 2.59 (4.04) 0.280a Total medical appointments 86.2 (80.7) 115 (84.1) 0.021a HbA1c (%) mean (SD) 7.41 (1.85) 7.66 (2.24) 0.613a HbA1c ≥ 7.5% 1,200 (36.3%) 11 (55%) 0.135b HbA1c ≥ 8% 906 (27.4%) 6 (30%) 0.996b HbA1c ≥ 9% 565 (17.1%) 5 (25%) 0.271b BMI (kg/m2 ) mean (SD) 30.1 (5.30) 29.8 (4.54) 0.798a BMI < 25 427 (15.9%) 0 0.077b BMI 25–30 1,010 (37.6%) 9 (64.3%) BMI > 30 1,249 (46.5%) 5 (35.7%) SBP (mmHg) mean (SD) 137 (17.8) 135 (19.9) 0.570a DBP (mmHg) mean (SD) 79.3 (11.2) 78.5 (12.5) 0.791a Total cholesterol (mg/dl) mean (SD) 207 (48.0) 205 (45.7) 0.850a HDL cholesterol (mg/dl) mean (SD) 48.1 (13.5) 45.5 (7.62) 0.165a LDL cholesterol (mg/dl) mean (SD) 124 (36.2) 131 (42.4) 0.504a Triglycerides (mg/dl) mean (SD) 105 (27.2) 122 (21.2) 0.025a Hemoglobin 13.5 (1.24) 12.7 (1.57) 0.101a Platelets 259 (71.4) 326 (80.4) 0.138a ESR 22.9 (19.1) 40.6 (43.1) 0.411a DM complications Macrovascular disease 640 (8.04%) 4 (8.33%) 0.793b Diabetic retinopathy 296 (3.72%) 2 (4.17%) 0.699b Diabetic nephropathy 94 (1.18%) 1 (2.08%) 0437b Diabetic neuropathy 95 (1.19%) 1 (2.08%) 0.115b Heart failure 221 (2.78%) 1 (2.08%) 1.000b DM treatment Non-pharmacological 5,197 (65.3) 31 (64.6) 0.909b Metformin 2,293 (28.8%) 15 (31.2%) 0,833b Secretagogues 826 (10.4%) 6 (12.5%) 0.632b DPP4-i 7 (0.1) 0 0.837b iSGLT2 1 (0) 0 0.938b Insulin 550 (6.91%) 3 (6.25%) 0.854b BMI, body mass index; DBP, diastolic blood pressure; DPP4-i, inhibitors of dipeptidyl peptidase 4; ESR, erythrocyte sedimentation rate; HDL, high-density lipoprotein; iSGLT2, Sodium-glucose cotransporter type 2 inhibitors; LDL, low-density lipoprotein; SBP, systolic blood pressure; SD, standard deviation. a t-test. b Fisher’s exact test. Frontiers in Public Health 04 frontiersin.org Antonio-Arques et al. 10.3389/fpubh.2022.1017024 TABLE 2 Characteristics of tuberculosis patients according to basal glycemic control. Variable HbA1c < 7.5% HbA1c ≥ 7.5% p-valuesa (n = 22) (n = 24) Localization 1.000 Pulmonary 16 (72.7%) 17 (70.8%) Extrapulmonary 6 (27.3%) 7 (29.2%) Radiography 0.575 Normal 6 (27.3%) 7 (29.2%) Cavitary 2 (9.09%) 5 (20.8%) Anormal non-cavitary 13 (59.1%) 12 (50%) TST 0.683 Positive 7 (31.8%) 9 (37.5%) Negative 2 (9.09%) 3 (12.5%) Unknown 13 (59.1%) 12 (50%) Bacteriology 0.683 Positive culture 11 (50%) 8 (33.3%) Negative culture 3 (13.6%) 7 (29.2%) ADA 1 (4.55%) 1 (4.17%) PCR (+) 0 (0%) 1 (4.17%) DM treatment 1.000 Non-pharmacological treatment 13 (59.1%) 14 (58.3%) NIAD 8 (36.4%) 9 (37.5%) Insulin 1 (4.55%) 1 (4.17%) aFisher’s exact test. TST, tuberculine skin test; NIAD, Non-insulin antidiabetic treatment; ADA, adenosine deaminase; PCR, protein chain reaction. was collected including glycated hemoglobin (HbA1c) during Statistical methods the follow-up, DM duration and treatment (i.e., lifestyle management, oral medication: metformin, secretagogues, Baseline comorbidities and demographic and clinical DPP4-I, ISGLT2, arGLP1, and/or insulin), microvascular characteristics were described by frequencies (n) and (neuropathy, nephropathy, retinopathy) and macrovascular percentages (%). Continuous variables were described using complications (stroke, peripheral arteriopathy, and ischemic mean and standard deviation (SD). Comparison tests were heart disease), and heart failure. As per the Barcelona TB computed as appropriate (t-test, and Fisher’s exact test) using Program, an active case of TB was considered when the study compareGroups R package (version 4.5.1) (28). In Kaplan-Meier subject had a maintained prescription of an anti-TB drug. analysis time-varying HbA1c was considered. The diagnosis of TB was based on the recommendations We conducted a time-to-event analysis to estimate the of the Spanish Consensus (26). The priority method was incidence of TB globally and by HbA1c group. Time-dependent the bacteriological study (baciloscopy and/or culture), but variables with Cox models were performed to examine the when this was not possible, PCR techniques, histology, association between glycemic control levels and the risk of biochemical methods (ADA) were used, or radiological and/or TB (unadjusted and adjusted for confounding variables, as clinical impression. appropriate). A different grouping strategy was assessed, and This program has an active epidemiological surveillance risk profiles were constructed based on HbA1c continuous system that makes under-detection very unlikely, with a good and risk cut-off points: ≥7.5%, ≥8%and ≥9%, respectively, follow-up of patients that includes the recording of TB cases and according to local guidelines (29). their characteristics, diagnostic procedure (smear observation, A database was constructed for the time-dependent models, culture, tuberculin skin test), and location (pulmonary or and HbA1c values were updated every time patients had a new extrapulmonary) and treatment. The socioeconomic deprivation value available. Thus, the last HbA1c was carried over to the next index (MEDEA) was collected in patients with TB. This index value (or change) or the end of the follow-up (TB event, death, classifies each study subject by living area, whereby values with or end of study). an average of 0 and a standard deviation of 1 and higher values Cox proportional risk regression models for clustered indicate a more unfavorable socioeconomic situation (27). data with constant and varying (HbA1c) variables over Frontiers in Public Health 05 frontiersin.org Antonio-Arques et al. 10.3389/fpubh.2022.1017024 TABLE 3 Incidence and risk of TB (unadjusted and adjusted hazard ratio) in DM patients depending on HbA1c level (%). Total person Number Incidence 95% CI HR unadjusted HR adjusted HR adjusted time (years) of TB rate per model 1 model 2 cases 100,000 p –y HR p-value HR p-value HR p-value (95% CI) (95% CI) (95% CI) All DM patients (n 68,605 48 69.97 51.6–92.8 – – – = 8,004) DM patients with 51,188 46 89.9 65.8–120.0 – – – HbA1c available (n = 6,733) HbA1c < 7,5% 31,276 22 70.3 44.1–106.5 Ref Ref Ref HbA1c ≥ 7,5% 19,912 24 120.5 77.2–179.3 2.22 0.016 1.87 0.064 1.89 0.293 (1.16–4.28) (0.98–3.64) (0.60–5.42) HbA1c < 8% 36,468 25 68.6 44.4–101.2 Ref Ref Ref HbA1c ≥ 8% 14,721 21 143.0 88.3–218.1 2.50 0.005 2.07 0.031 2.06 0.207 (1.31–4.78) (1.07–4.01) (0.67–6.32) HbA1c < 9% 42,481 30 70.6 47.6–100.8 Ref Ref Ref HbA1c ≥ 9% 8,707 16 183.8 105–298.4 4.36 <0.001 3.62 <0.001 2.82 0.082 (2.28–8.33) (1.86–7.04) (0.88–9.06) Model 1: adjusted by age and sex. Model 2: adjusted by age, sex, years of evolution of DM, microvascular and macrovascular complications, alcohol, smoking habit and geographical origin. The bold values are the hazard ratios for each HbA1c thresholds in each model. FIGURE 2 Cumulative incidence per 100,000 person-year of TB according to HbA1c level. Frontiers in Public Health 06 frontiersin.org Antonio-Arques et al. 10.3389/fpubh.2022.1017024 FIGURE 3 Forest plot of the factors associated with DM that can influence the appearance of TB. time were fitted using the survival R package (30). Hazard 57.7 years old), had higher levels of triglycerides (122 vs. 105 ratios (HR), unadjusted and adjusted, with their 95% mg/dl), had a higher prevalence of diabetic nephropathy and confidence intervals (CIs), were estimated. CIs and p-values neuropathy, and were more frequently from Hindustan (29.2 vs. were computed with robust standard errors to account for 13.4%) (Table 1). cluster (subjects). A sensitivity analysis of different fitted models, including different adjusted variables, was performed. A complete- Differences on TB characteristics cases analysis was performed with values available. Statistical significance was established as a p-value < 0.05. Data Overall, baseline HbA1c values were available for 46 DM management and analysis were performed with the R version subjects with TB. No statistically significant differences were 3.6.3 package (31). observed in subjects with an HbA1c ≥ 7.5% vs. those with HbA1c values < 7.5% for localization of TB disease, radiography Results and tuberculin skin test (Table 2). At the time of TB, 88.8% were symptomatic, and 44% General characteristics were smear-positive (21.3% were smear-negative, but the culture was positive). A total of 8,004 diabetes subjects were analyzed, equating to Regarding treatment, the most commonly used regimen 68,605 person-years of follow-up (PYFU), with 48 developing was the standard treatment with 4 drugs (47.2%), followed by TB during the follow-up period. Compared to DM subjects 6 months with 3 drugs (11.2%). In 13.5% of the cases had without TB, DM subjects with TB were younger (52.2 vs. resistant TB. Frontiers in Public Health 07 frontiersin.org Antonio-Arques et al. 10.3389/fpubh.2022.1017024 Risk of developing TB according to China, where patients with DM showed higher risks of active glycemic control (3.11), culture-confirmed (3.08), and pulmonary (3.63) but not extrapulmonary tuberculosis with baseline hemoglobin A1c > The TB incidence rate was 70 per 100,000 PYFU in the 7% (38). overall DM group, and was 90 per 100,000 in the DM group In addition, as well as influencing the onset of active with any HbA1c available. The incidence of TB increased with TB, a poor glycemic control may also increase the risk of increasing HbA1c cut-off values (i.e., highest in the group with latent TB infection. According to this previous study, glycemic HbA1c ≥9% followed by ≥8% and then ≥7.5%) (Table 3). The control assessed by glycosylated hemoglobin values influences incidence curves for TB over time according to the HbA1c group the capacity of the host to control the infection by altering also showed that the group with HbA1c ≥ 9% was more prone immune response, and had increased susceptibility to pathogens to TB than the other HbA1c groups (Figure 2). (39). Supporting this, in a recent meta-analysis, it was concluded The association between HbA1c and risk of TB according that poor glycemic control (defined by an HbA1c > 7%) would to the different HbA1c cut-off points (HbA1c ≥ 7.5, 8, and double the prevalence of TB (40). 9%) was further analyzed by estimating the unadjusted and Other studies were conducted without finding any relation adjusted HR: with higher levels of glycated hemoglobin, a higher between the risk of TB and the metabolic control. In a incidence of TB was observed, reaching the highest values in population-based study placed in Denmark, the risk of TB in those subjects with values over 9%. When we adjusted by age patients with DM was lower than previously expected (HR 1.18) and sex (model 1), we found statistically significant differences and no significant association with glycemic control was found. if HbA1c ≥ 8 and ≥9%, but not when adjusting by age, sex, Probably the main limitation of this work was the low number years of evolution of DM, macrovascular complications, alcohol of DM patients analyzed (7 subjects with HbA1c between 7 or smocking habit, BMI and geographical origin (model 2). and 7.9%, 16 for HbA1c ≥ 8% and 20 subjects with unknown Overall the risk increased with higher values of HbA1c (Table 3; values of HbA1c) and the low probability of TB in low-incidence Figure 3). countries (41). With similar conclusions, a study performed in UK showed that the increased risk of TB among DM patients was 1.3, it was considered moderate, and no evidence for a higher Discussion risk linked to a worse glycemic control was observed, but DM patients with the lowest and highest rates of chronic disease DM is one of the main risk factors for developing TB (2), management had a higher risk of TB (42). and both diseases frequently coexist, even in countries with The results of our work are in line with the publications a low incidence of TB (32). In our work, we observed an mentioned above: although the results were not statistically increased incidence of TB specially in patients with DM with significant, probably due to the small number of patients, the higher values of HbA1c; the increased HRs also suggested an study was done in a high incidence of TB zone, and a relation increased risk. This study took place in an inner city district of can be observed: with worse glycemic control, the risk of TB is Barcelona, where we found in previous work that DM patients greater. In the inner cities of developed countries, always with a had 1.90 (CI: 1.18-3.07) higher risk of developing TB (23). high incidence of TB, a good control of DM patients is necessary Without considering the metabolic control, a prospective cohort to avoid complications as TB. study showed that DM patients had more than 2-fold risk of When adjusting by age, sex, years of evolution of developing TB, and this risk increased with increasing number DM, microvascular and macrovascular complications, alcohol, of DM complications (33). smoking habit and geographical origin, the HR in the 9% cut- The higher risk of TB in patients with DM and poor off point changes from 3.62 to 2.82. This fact can be explained metabolic control has been studied before (34–36). In a cohort because the addition of the variable “geographical origin”. study conducted in Taiwan with 120,000 patients, with a People from Hindustan are the group with the highest incidence multivariate analysis, it was observed that diabetic patients with of TB in our city. They have much more latent TB [as they come poor glycemic control [described as a basal fasting plasma from high incidence countries (2)] and, therefore, a greater risk glucose (FPG) > 130 mg/dl] had twice the risk of developing of developing active TB. TB than non-diabetic patients. While there was no difference Disruptions of immunological mechanisms because of in the risk of TB between non-diabetic subjects and diabetic hyperglycemia may explain the increased TB risk. In this patients with good glycemic control, the risk of TB increased respect, excess glucose may affect cell activation, phagocytic with increasing FPG in diabetic patients (14). The results also capacity, microbicide mechanisms of alveolar macrophages align with another study conducted in China at the hospital and neutrophils, leukocyte transmigration, and chemotaxis, level, where patients with poor metabolic control (also defined as potentially delaying antigen presentation and, consequently, basal FPG > 130 mg/dl) were 2.66 times more likely to develop activation of the necessary immunological mechanisms to fight pulmonary TB (37), and with another work also performed in against Mycobacterium tuberculosis (43). Of note, DM patients Frontiers in Public Health 08 frontiersin.org Antonio-Arques et al. 10.3389/fpubh.2022.1017024 show the most severe clinical presentation of TB (17), and a Conclusion higher risk of mortality and multi-drug resistant TB (18, 44). Although we did not find significant differences between In summary, poor glycemic control is related to an increased groups when when comparing pulmonary or extrapumonary TB risk of TB development. Given the results of this research, and, on the other hand, the localization of the lesions detected improved glycemic control can be pursued in DM not only to via chest radiography, other studies have described differences reduce the risk of vascular events but also to decrease the risk in those with poorer glycemic control. In a study of more than of TB and its complications, especially in areas with a high 600 diabetic patients diagnosed with TB, patients with poor prevalence of TB. The increasing worldwide prevalence of DM metabolic control, defined by HbA1c > 7%, had more severe TB, and failures to eradicate TB, coupled with the synergy between more extensive lung disease, and more cavitations (15). Another DM and TB, make it necessary to consider managing these two work also described more involvement of lower lung fields and diseases together. increased likelihood of cavitation with poor glycemic control (45). In the work of Huang et al., diabetic patients with a Hb1Ac > 8% had more extensive lung lesions, atypical findings, more Data availability statement lymphadenopathy, more cavitation, and a greater likelihood of involvement of all lung lobes (16). Regarding symptoms, The datasets analyzed for this study can be found in one study concluded that diabetic patients, especially those the following link: https://github.com/USR-DAPCAT/TBC_ with HbA1c > 9%, had more clinical signs of TB: hemoptysis, Glicada. asthenia, and weight loss (46). In our work, although differences were insignificant, patients with poor glycemic control had more Ethics statement cavitary lesions on chest-x-ray. The main limitation of this work is the low number of The studies involving human participants were reviewed and TB cases found during the follow-up period but our inner- approved by Primary Health Care University Research Institute city has a high incidence of TB compared with the rest of the Jordi Gol, in 2016 (Code: P16/023). Written informed consent city (22) what facilitated this study. Fortunately, TB is not a for participation was not required for this study in accordance frequent disease nowadays in Western European countries but with the national legislation and the institutional requirements. its incidence in the inner cities of developed countries is still too high (47). Moreover, we limited the selection of subjects to those with both diseases: DM and TB. It is a very concrete Author contributions group of subjects, but, in our opinion, this analysis has a clinical and epidemiological value. According to the literature, other VA-A, JC, JF-N, JR, ÀO, and AM-M participated in the study studies which reported a relationship between glycemic control design. ÀO, JD, and JR worked on data collection. JR, EN, RP, and TB had a moderate number of subjects with both TB+DM and JF-N performed all statistical work. VA-A, JF-N, and JC were diseases: 63 and 214 in two studies in Taiwan (14, 16), and 105 major contributors in writing the manuscript. MM-C, DM, JJ, in a study in China (37). Within the limitations of working JM, BV, and AM-M reviewed and corrected the manuscript. BV with real world data, we could not have a systematic follow- contributed to prepare the manuscript according to the journal up of glycated hemoglobin as this was a retrospective study. policies. All authors contributed to the article and approved the The small number of cases with TB has made a clear statistical submitted version. significance difficult. Working in a real-world-data environment may have some inherent limitations, but it has the great strength Funding of reflecting the conditions of standard clinical practice in almost all patients. This work was supported by grants given by the Instituto Studies with large databases may involve different types of de Salud Carlos III PI16/01751 (Spanish Ministry of Economy), bias. Our study has tried to minimize the diagnostic errors of the Institut Universitari per a la Recerca a l’Atenció Primària DM and TB by basing their diagnosis both on the ICD-10 clinical de Salut Jordi Gol i Gurina (Catalan Health Institute) coding and on the incorporation of complementary variables PREDOC_ECO-19/2, and the Fundación redGDPS (Beca de such as the use of specific treatment for these pathologies. This apoyo José Luis Torres a la Investigación). method has been previously validated and published with our database (48), but we assume that, in some cases, it may be a limitation. On the other hand, the fact that some data on the Acknowledgments bacteriology of TB diagnosis is not available is a limitation, so the diagnosis has been complemented with other variables, as The authors thank Amanda Prowse for the grammatical detailed in the Section Methods. revision of the manuscript. Frontiers in Public Health 09 frontiersin.org Antonio-Arques et al. 10.3389/fpubh.2022.1017024 Conflict of interest organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or The authors declare that the research was conducted in the claim that may be made by its manufacturer, is not guaranteed absence of any commercial or financial relationships that could or endorsed by the publisher. be construed as a potential conflict of interest. Supplementary material Publisher’s note The Supplementary Material for this article can be found All claims expressed in this article are solely those of the online at: https://www.frontiersin.org/articles/10.3389/fpubh. authors and do not necessarily represent those of their affiliated 2022.1017024/full#supplementary-material References 1. WHO. Enfermedades No Transmisibles. Available online at: https://www.who. 16. Huang L-K, Wang H-H, Lai Y-C, Chang S-C. The impact of glycemic status int/es/news-room/fact-sheets/detail/noncommunicable-diseases (accessed May on radiological manifestations of pulmonary tuberculosis in diabetic patients. PLoS 15, 2021). ONE. (2017) 12:e0179750. doi: 10.1371/journal.pone.0179750 2. World Health Organization. Global Tuberculosis Report 2021. (2021). Available 17. Moreno-Martínez A, Casals M, Orcau, Gorrindo P, Masdeu E, Caylà online at: https://www.who.int/teams/global-tuberculosis-programme/tb-reports/ JA. Factors associated with diabetes mellitus among adults with tuberculosis global-tuberculosis-report-2021 (accessed April 28, 2022). in a large European city, 2000-2013. Int J Tuberc Lung Dis. (2015) 19:1507– 12. doi: 10.5588/ijtld.15.0102 3. International Diabetes Federation. IDF Diabetes Atlas Ninth. Dunia: IDF (2019). p. 168. 18. Huangfu P, Ugarte-Gil C, Golub J, Pearson F, Critchley J. The effects of diabetes on tuberculosis treatment outcomes: an updated systematic review and 4. Stratton IM, Adler AI, Neil HAW, Matthews DR, Manley SE, Cull CA, et al. meta-analysis. Int J Tuberc Lung Dis. (2019) 23:783–96. doi: 10.5588/ijtld.18. Association of glycaemia with macrovascular and microvascular complications of 0433 type 2 diabetes (UKPDS 35): prospective observational study. Br Med J. (2000) 321:405–12. doi: 10.1136/bmj.321.7258.405 19. Antonio-Arques V, Franch-Nadal J, Caylà JA. Diabetes and 5. Stettler C, Allemann S, Jüni P, Cull CA, Holman RR, Egger M, et al. Glycemic tuberculosis: a syndemic complicated by COVID-19. Med Clínica. (2021) control and macrovascular disease in types 1 and 2 diabetes mellitus: meta-analysis 157:288–93. doi: 10.1016/j.medcli.2021.04.004 of randomized trials. Am Heart J. (2006) 152:27–38. doi: 10.1016/j.ahj.2005.09.015 20. World Health Organization, International Union Against Tuberculosis and 6. Tracey ML, McHugh SM, Fitzgerald AP, Buckley CM, Canavan RJ, Kearney Lung Disease. Collaborative framework for care and control of Tuberculosis and PM. Risk factors for macro-and microvascular complications among older adults Diabetes, Geneva Vol. 314 (2011). p. 2. with diagnosed type 2 diabetes: findings from the Irish longitudinal study on 21. de Vries G, Aldridge RW, Caylã JA, Haas WH, Sandgren A, van ageing. J Diabetes Res. (2016) 2016:5975903. doi: 10.1155/2016/5975903 Hest NA, et al. Epidemiology of tuberculosis in big cities of the European 7. Holman RR, Paul SK, Bethel MA, Matthews DR, Neil HAW. 10-year follow-up union and European economic area countries. Eurosurveillance. (2014) 19:1– of intensive glucose control in type 2 diabetes. N Engl J Med. (2008) 359:1577– 8. doi: 10.2807/1560-7917.ES2014.19.9.20726 89. doi: 10.1056/NEJMoa0806470 22. Àngels Orcau i Palau, Carmen Gallego Cortés, Joan A. Caylà i Buqueras 8. Abu-Ashour W, Twells LK, Valcour JE, Gamble J-M. Diabetes and the CR i G, Recull. La tuberculosi a Barcelona. Informe. (2017). Available online occurrence of infection in primary care: a matched cohort study. BMC Infect Dis. at: https://www.aspb.cat/wp-content/uploads/2020/10/tuberculosi-barcelona- (2018) 18:67. doi: 10.1186/s12879-018-2975-2 2017.pdf (accessed April 28, 2022). 9. Baiju R, Janet E. Quantifying the risk of infectious diseases for people with 23. Antonio-Arques V, Franch-Nadal J, Moreno-Martinez A, Real J, Orcau diabetes. Diabetes Care. (2003) 26:510–3. doi: 10.2337/diacare.26.2.510 À, Mauricio D, et al. Subjects with diabetes mellitus are at increased risk for developing tuberculosis: a cohort study in an inner-city district of Barcelona 10. Muller LMAJ, Gorter KJ, Hak E, Goudzwaard WL, Schellevis FG, Hoepelman (Spain). Front public Heal. (2022) 10:789952. doi: 10.3389/fpubh.2022.789952 AIM, et al. Increased risk of common infections in patients with type 1 and type 2 diabetes mellitus. Clin Infect Dis. (2005) 41:281–8. doi: 10.1086/4 24. Perfil de la població estrangera a Ciutat Vella i els seus barris. Gener 31587 (2020). Available online at: https://ajuntament.barcelona.cat/estadistica/catala/ Estadistiques_per_temes/Poblacio_i_demografia/Documents_relacionats/pobest/ 11. Hine JL, de Lusignan S, Burleigh D, Pathirannehelage S, McGovern A, a2020/part1/nt27.htm (accessed September 15, 2021). Gatenby P, et al. Association between glycaemic control and common infections in people with Type 2 diabetes: a cohort study. Diabet Med. (2017) 34:551– 25. Agència de Salut Pública de Barcelona. Infobarris (2021). Available online 7. doi: 10.1111/dme.13205 at: https://www.aspb.cat/docs/infobarris/pagina-interna.html?codi=1#8 (accessed October 8, 2022). 12. Critchley JA, Carey IM, Harris T, DeWilde S, Hosking FJ, Cook DG. Glycemic control and risk of infections among people with type 1 or type 2 26. González-Martín J, García-García JM, Anibarro L, Vidal R, diabetes in a large primary care cohort study. Diabetes Care. (2018) 41:2127– Esteban J, Blanquer R, et al. Documento de consenso sobre diagnóstico, 35. doi: 10.2337/dc18-0287 tratamiento y prevención de la tuberculosis. Arch Bronconeumol. (2010) 46:255–74. doi: 10.1016/j.arbres.2010.02.010 13. Jeon CY, Murray MB. Diabetes mellitus increases the risk of active tuberculosis: a systematic review of 13 observational studies. PLoS Med. (2008) 27. Domínguez-Berjón MF, Borrell C, Cano-Serral G, Esnaola S, Nolasco A, 5:1091–101. doi: 10.1371/journal.pmed.0050181 Pasarín MI, et al. Construcción de un índice de privación a partir de datos censales en grandes ciudades españolas (Proyecto MEDEA). Gac Sanit. (2008) 14. Lee PH, Fu H, Lai TC, Chiang CY, Chan CC, Lin HH. Glycemic 22:179–87. doi: 10.1157/13123961 control and the risk of tuberculosis: a cohort study. PLoS Med. (2016) 13:1– 15. doi: 10.1371/journal.pmed.1002072 28. Subirana I, Sanz H, Vila J. Building bivariate tables: the comparegroups package for R. J Stat Softw. (2014) 57:1–16. doi: 10.18637/jss.v057.i12 15. Mahishale V, Avuthu S, Patil B, Lolly M, Eti A, Khan S. Effect of poor glycemic control in newly diagnosed patients with smear-positive pulmonary tuberculosis 29. Alemán JJ, Menéndez SA, Lachica LÁ, Puente JB de la. Guía de diabetes and type-2 diabetes mellitus. Iran J Med Sci. (2017) 42:144–51. tipo 2 para clínicos: Recomendaciones de la redGDPS. redGDPS (2018). Available Frontiers in Public Health 10 frontiersin.org Antonio-Arques et al. 10.3389/fpubh.2022.1017024 online at: https://www.redgdps.org/gestor/upload/colecciones/GuiaDM2_web.pdf based observational study. Clin Infect Dis. (2017) 65:2060–8. doi: 10.1093/cid/cix (accessed October 04, 2022). 632 30. Therneau TM, Grambsch PM. Modeling Survival Data: Extending the Cox 40. Chen Z, Liu Q, Song R, Zhang W, Wang T, Lian Z, et al. The association of Model. New York: Springer (2000). glycemic level and prevalence of tuberculosis: a meta-analysis. BMC Endocr Disord. (2021) 21:1–4. doi: 10.1186/s12902-021-00779-6 31. R Core Team1. The R Project for Statistical Computing (2020). Available online at: https://www.R-project.org/ 41. Leegaard A, Riis A, Kornum JB, Prahl JB, Thomsen VØ, Toft Sørensen H, et al. Diabetes, glycemic control, and risk of tuberculosis A population-based 32. New York City Health Department Annual Tuberculosis Summary, case-control study. Diabetes Care. (2011) 34:2530–5. doi: 10.2337/dc11-0902 2019 (2019). 42. Moore DAJ, Smeeth L, Prieto-Merino D, Wing K, Pealing L, Mathur R. 33. Baker MA, Lin H-H, Chang H-Y, Murray MB. The risk of tuberculosis disease Risk of tuberculosis in patients with diabetes: population based cohort study among persons with diabetes mellitus: a prospective cohort study. Clínical Infect using the UK Clinical Practice Research Datalink. BMC Med. (2015) 13:1– Dis. (2012) 54:818–25. doi: 10.1093/cid/cir939 16. doi: 10.1186/s12916-015-0381-9 34. Arliny Y, Yanifitri DB, Mursalin D. High glycosylated hemoglobin 43. Ayelign B, Negash M, Genetu M, Wondmagegn T, Shibabaw T. level as a risk factor of latent tuberculosis infection in patients Immunological impacts of diabetes on the susceptibility of mycobacterium with uncomplicated type 2 diabetes mellitus. Universa Med. (2022) tuberculosis. J Immunol Res. (2019) 2019:1–8. doi: 10.1155/2019/6196532 41:47–55. doi: 10.18051/UnivMed.2022.v41.47-55 44. Salindri AD, Kipiani M, Kempker RR, Gandhi NR, Darchia L, Tukvadze 35. Ghazanfarpour A, Preux P-M. Decrease of glycated hemoglobin N, et al. Diabetes reduces the rate of sputum culture conversion in patients with based on evaluation of diabetes intervention programs in low-and middle- newly diagnosed multidrug-resistant tuberculosis. Open Forum Infect Dis. (2016) income countries: a systematic review. Diabetes Epidemiol Manag. (2022) 3:1–10. doi: 10.1093/ofid/ofw126 6:100055. doi: 10.1016/j.deman.2022.100055 45. Chiang CY, Lee JJ, Chien ST, Enarson DA, Chang YC, Chen YT, et al. 36. Zhang YM, Wang F, Soolingen D van, Anthony RM. The impact Glycemic control and radiographic manifestations of tuberculosis in diabetic of diabetes mellitus on pulmonary TB. Public Heal Action. (2022) 12:64–7. patients. PLoS ONE. (2014) 9:e93397. doi: 10.1371/journal.pone.0093397 doi: 10.5588/pha.21.0070 46. Chiang CY, Bai KJ, Lin HH, Chien ST, Lee JJ, Enarson DA, et al. The influence 37. Wang YM, Dou M, Kou TM, Liu YM, Lv W, Han L, et al. of diabetes, glycemic control, and diabetes-related comorbidities on pulmonary Risk of having pulmonary tuberculosis in type 2 diabetes: a hospital- tuberculosis. PLoS ONE. (2015) 10:e0121698. doi: 10.1371/journal.pone.0121698 based matched case-control study. Asia Pac J Clin Nutr. (2021) 30:303– 10. doi: 10.6133/apjcn.202106_30(2).0015 47. European Centre for Disease Prevention and Control, WHO Regional Office for Europe. Tuberculosis Surveillance and Monitoring in Europe 2021 – 2019 Data. 38. Leung CC, Lam TH, Chan WM, Yew WW, Ho KS, Leung GM, et al. Copenhagen: WHO Regional Office for Europe (2021). Diabetic control and risk of tuberculosis: a cohort study. Am J Epidemiol. (2008) 48. Mata-Cases M, Mauricio D, Real J, Bolíbar B, Franch-Nadal J. Is 167:1486–94. doi: 10.1093/aje/kwn075 diabetes mellitus correctly registered and classified in primary care? A 39. Martinez L, Zhu L, Castellanos ME, Liu Q, Chen C, Hallowell BD, et al. population-based study in Catalonia, Spain. Endocrinol Nutr. (2016) 63:440– Glycemic control and the prevalence of tuberculosis infection: a population- 8. doi: 10.1016/j.endonu.2016.07.004 Frontiers in Public Health 11 frontiersin.org