Table 1 Characteristics of included studies in the systematic review

Magee et al. 2018 [8] (U.S.A.)

Cross-sectional study

Patients with diabetes identified by self-report and glycated hemoglobin (n = 4958)

Metformin medication identified by self-report

A diagnosis of LTBI infection identified by QuantiFERON-TB Gold In-tube

Patients without metformin medication

575

Latent TB infection prevalence was non-significantly higher in those without metformin use (prevalence difference, 1.4, 95%CI, −3.7

−6.4%) compared to those self-reporting any metformin use (OR, 1.1, 95%CI, 0.7 to 1.9).

NA

Singhal et al. 2014 [4] (Singapore)

Two retrospective cohort studies

Cohort 1: Patients suffered from both DM and TB

(n = 273)

Cohort 2: Patients with DM (n = 220)

Metformin treatment

Cohort 1: Pulmonary cavities at diagnosis and mortality during the first year after diagnosis.

Cohort 2: A diagnosis of LTBI tested by T-Spot TB array

Patients using alternative drugs for DM.

Cohort 1:

273

Cohort 2:62

Cohort 1: those receiving metformin had fewer pulmonary cavities (OR, 0.6; 95%CI, 0.36 to 0.97). The Mortality was 3% among patients who received metformin compared to 10% among patients in the non-metformin group (OR, 0.29; 95%CI, 0.14 to 0.95).

Cohort 2: metformin therapy was associated with reduced T-SPOT reactivity when compared with controls (OR, 0.44; 95%CI, 0.20 to 0.95).

NA

Marupuru et al. 2017 [19] (India)

Case-control study

Diabetics (≥ 40 years old) identified on the basis of ICD-10 coding for disease classification (E11.9, n = 448)

Metformin usage

The diagnosis of TB following ICD-10 code A15-A19

Patients without metformin usage

149

The protective effect of metformin against TB was 3.9-fold in diabetics (OR = 0.256, 0.16 to 0.40). No difference was found between those on 1000 mg metformin (27.3%), and on 500 mg (25.7%) for development of TB.

NA

Lee et al. 2017 [11] (South Korea)

Retrospective cohort study

Culture-proven pulmonary TB in patients diagnosed with DM; follow-up sputum Mtb cultures after 2 months of treatment; completion of World Health Organization-recommended TB treatment (n = 105)

Metformin treatment

(1) Sputum culture conversion after 2 months of treatment;

(2) Recurrence of TB proven by isolation of Mtb or clinical radiological evidence

Participants without metformin treatment

105

The OR of sputum culture conversion at 2 months for patients with metformin use was 2.69 (95%CI, 0.92 to 7.95); A statistical difference in the recurrence rate with the use of metformin (OR, 1.92; 95%CI, 0.42 to 8.76) was observed. Metformin improved the sputum culture conversion rate in patients with cavitary pulmonary TB (OR, 10.8; 95%CI, 1.22 to 95.63)

Sex, statin use, insulin, cancer, AFB smear grade, and drug resistance

Lin et al. 2017 [20]

(Tai Wan)

Retrospective cohort study

Patients aged 20 and older newly diagnosed with diabetes (ICD-9250.XX and A-code A181) and without a past history of pulmonary TB (n = 22,256)

Metformin treatment

Incidence of pulmonary TB identified in follow-up

Patients without metformin treatment

682

Unadjusted crude HR:

0.42 (95%CI, 0.35 to 0.50)

Adjusted HR:

0.52 (95%CI, 0.43 to 0.62)

Sex, age, alcoholism, chronic obstructive pulmonary disease, cirrhosis, alcoholic liver damage, hepatitis C, chronic kidney diseases, and malignancies

Degner et al. 2018 [12]

(Tai Wan)

Retrospective cohort study

Patients aged ≥13 years with culture-confirmed, drug-susceptible pulmonary TB undergoing treatment and diabetes (n = 634)

Metformin medication within 30 days of starting TB treatment

Effect on mortality among patients with DM undergoing TB treatment

Patients without metformin prescriptions

634

Adjusted HR: 0.56 (95%CI, 0.39 to 0.82)

Unadjusted HR: 0.50 (95%CI, 0.35 to 0.72)

Age, sex, chronic kidney disease, cancer, cavitary diseases, TB treatment adherence

Lee et al. 2018 [9]

(Tai Wan)

Retrospective cohort study

Patients had at least one hospital admission or at least three outpatient visits with a DM diagnostic code (ICD-9) within 365 calendar days (n = 177,732)

With total prescriptions of metformin for > 90 cumulative defined daily doses within 1 year after the onset of DM treatment

A newly diagnosed TB (ICD-9-CM code: 010–018) after the index date.

Diabetics do not satisfy the exposure criteria

1514

Adjusted HR: 0.84 (95%CI, 0.74 to 0.96)

TB risk was lower in high-dose metformin users than in low-dose users (HR, 0.83; 95%CI, 0.72 to 0.97)

Sex, type 1 diabetes mellitus, age, low income, chronic obstructive pulmonary disease, liver cirrhosis, pulmonary cancer, extra-pulmonary cancer, bronchiectasis, and so on

Lin et al. 2018 [10]

(Tai Wan)

Retrospective cohort study

Patients who were 20–100 years old and who were newly diagnosed with type 2 DM (ICD-9-CM, 250.X0 and 250.X2) (n = 10,052)

Metformin usage

A diagnosis of active TB (ICD-9-CM, 010–018) during the follow-up period and the prescription of more than two anti-TB medications for more than 90 days.

Participants without metformin usage

329

Adjusted RR: 0.24 (95%CI, 0.18 to 0.32)

Unadjusted RR: 0.37 (95%CI, 0.29 to 0.47)

The duration of DM diagnosis, comorbidities (chronic pulmonary disease/ renal disease), oral anti-diabetic therapy, and insulin injection therapy

Ma et al. 2018 [21]

(China)

Retrospective cohort study

Culture-positive retreatment pulmonary TB patients with type 2 DM

Multidrug-resistant TB, extensively drug-resistant TB, and extra-pulmonary TB were excluded.(n = 58)

Metformin treatment in regimens for diabetes

(1) Success treatment of TB

(2) Sputum culture conversion by the end of 2 months

(3) Relapse rates of patients

Patients without metformin medication

58

There were a higher proportion of treatment success (OR, 6.00; 95%CI, 0.71 to 50.59) and sputum culture conversions by the end of 2 months (OR, 2.80; 95%CI, 0.55 to 14.23) among metformin group. The relapse rates of patients in the metformin and non-metformin group were 6.3 and 35.7% (OR, 0.12; 95%CI, 0.01 to 1.20).

NA

Pan et al. 2018 [13]

(Taiwan)

Retrospective cohort study

Patients with a diagnosis of type 2 DM (ICD-9250 × 0, 250 × 2).

Patients aged < 20 years or had a diagnosis of TB were excluded.(n = 9475)

Participants received ≥60 cumulative defined daily dose of metformin and < 15 cumulative defined daily dose of sulfonylurea

TB occurrence

(ICD-9, 010–018)

Participants received ≥60 cumulative defined daily dose of sulfonylurea and < 15 cumulative defined daily dose of metformin

263

Adjusted RR: 0.337 (95%CI, 0.169 to 0.673)

Unadjusted RR:

0.477 (95%CI, 0.268 to 0.850)

Age, sex, adapted diabetes complication severity index score, index year, income level, and comorbidities

Tseng et al. 2018 [7]

(Taiwan)

Retrospective cohort study

Newly diagnosed diabetes patients (ICD-9) who had been followed up in the outpatient clinics with a prescription of antidiabetic drugs for two or more times

(n = 164,267)

Patients had been prescribed metformin as the first antidiabetic drug

Incidence density of TB infection

Patients without metformin prescriptions

2336

Adjusted HR:

0.552 (95%CI, 0.493–0.617)

Unadjusted HR: NA

Age, diabetes duration, sex, occupation, living region, hypertension, dyslipidemia, obesity, diabetes-related complications, antidiabetic drugs and so on