Impact of emphysema on sputum culture conversion in male patients with pulmonary tuberculosis: a retrospective analysis

In this retrospective study, we analysed data from hospitalized patients with positive acid-fast bacillus (AFB) smears at Jikei University Daisan Hospital between January 2015 and December 2018. During hospitalization, patients maintained smoking cessation. Exclusion criteria were: (1) age < 40 years; (2) isolation of MTB resistant to isoniazid (INH) or rifampicin (RMP); (3) patients not receiving standard anti-TB regimens with INH, RMP, ethambutol (EMB) and pyrazinamide, or INH, RMP and EMB; (4) patients required desensitization to any adverse effects or comorbidities; (5) co-infection with non-TB mycobacteria; (6) death within 28 days after anti-TB treatment initiation because sputum MTB conversion could not be confirmed; and (7) female sex because only three female PTB patients had emphysema. Sputum smear and culture examinations were performed at the start of treatment and then every 2 weeks, with a margin of error of a few days, during hospitalization. The sputum AFB smear was examined using a fluorochrome procedure according to the American Thoracic Society guidelines [15]. Sputum specimens were cultured in liquid (Mycobacterial Growth Indicator Tubes, MGIT; BD, Franklin Lakes, NJ, USA). Time to sputum culture conversion (TCC) was defined as the number of days between treatment initiation and the first negative sputum culture and no positive culture thereafter.⁷ The delayed conversion may be explained by the high baseline bacillary load. Baseline time to TB detection (TTD) was defined as time (in days) until MGIT culture showed a positive result at initial sputum culture [6, 16]. Atwine et al. reported a twofold increase in risk of culture conversion at 2 months in patients with baseline TTD < 14 days compared with ≥14 days [17]. This study was approved by the Ethics Committee of Jikei University School of Medicine [No. 30–003(9024)].

Baseline chest high-resolution CT (HRCT) was performed on all patients. HRCT was performed using a helical CT scanner (SOMATOM Definition AS+; Siemens, Erlangen, Germany) with the following parameters: 1.0–5.0-mm slice thickness, 0.3-mm slice separation, 120-kV tube voltage, 140 mAs, 0.6-mm collimation, 0.5-s rotation time, and 0.85 beam pitch. Images reconstructed with the B60f kernel were used for image analysis. CT scans were requested to establish diagnosis and evaluate active PTB lesions before treatment.

TB-related CT findings were evaluated by considering consolidation, nodule/mass, bronchiolitis and cavities in both lungs [18, 19]. Consolidation was defined as homogeneous increased lung attenuation that obscured the margin of vessels and airways. A nodule/mass was defined as a well-demarcated spherical area of soft tissue attenuation, ≤3 cm in diameter. Bronchiolitis was defined as the presence of centrilobular micronodules (< 10 mm in diameter) and branching nodular structures (including tree-in-bud appearance). Cavities were defined as round, gas- or fluid-filled spaces within a nodule/mass or consolidation. CT-detected emphysema was defined as low attenuation area (LAA) < − 950 HU, (LAA_<-950) [11, 20]. Emphysema severity was assessed according to the Goddard Scoring System [21, 22].: 0, normal; 1, LAA_<-950 ≤ 25% affected; 2, > 25% and ≤ 50% affected; 3, > 50% and ≤ 75% affected; and 4, > 75% affected. Six images were analyzed in three slices, which were obtained from the aortic arch level, carina level, and 1 cm above the right diaphragm, and a total score for six images was calculated for each patient. Two pulmonologists with 11 and 16 years’ experience interpreted chest CT results. When there were discrepancies in their readings, final decisions were reached by consensus. Goddard score was determined as the average value scored by the pulmonologists. Total Goddard score ≥ 1 was defined as emphysema, as reported previously [22], because 95% of nonsmokers had lungs with < 5% emphysematous involvement [23].

Statistical analysis was performed using IBM SPSS statistics version 24.0. (IBM Corp., Armonk, NY, USA). Student’s t-test or Fischer’s exact test was used to compare the distributions of characteristics between PTB patients with and without emphysema. Cox proportional hazards regression analysis was used to estimate the hazard ratio (HR) of sputum culture conversion for the following prespecified baseline variables: age, emphysema, smoking history, presence of consolidation, cavity lesions, hypoalbuminemia, and baseline TTD within 2 weeks. HRs were calculated with 95% confidence intervals (CIs). A p value < 0.05 indicated statistical significance in all analyses. Kaplan–Meier curves for TB sputum conversion probability were plotted for patients with emphysema versus those without emphysema. The log-rank test was used to test the significance in subgroups of patients with emphysema.

Between January 2015 and December 2018, 171 patients were diagnosed and hospitalized with active TB with AFB smear- and culture-positive sputum at Jikei Daisan Hospital (Fig. 1). Ninety-two patients were excluded according to the exclusion criteria. The mean age was 69.01 ± 15.92 years. No patient was positive for HIV antibody. There was no significant difference in mean age, proportion of patients with chronic kidney disease or diabetes, or baseline TTD between patients with or without emphysema (Table 1). PTB patients with emphysema had a significantly higher proportion with smoking history (100% vs 41.46%, p < 0.0001), current smoking habit (52.63% vs 14.63%, p < 0.001), and hypoalbuminemia (81.58% vs 46.34%, p = 0.001) (Table 1). About a quantitative assessment of emphysema, average Goddard score of emphysema patients was 6.74 ± 4.73 (Table 1, Fig. 2).

We compared PTB-related CT findings including cavities, consolidation, bronchiolitis, and nodule/mass between PTB patients with or without emphysema. Cavities and consolidation were seen more frequently in PTB patients with than without emphysema (71.05% vs 43.90%, p = 0.015, and 84.21% vs 60.98%; p = 0.021, respectively) (Table 2).

In the Kaplan–Meier analysis, 38 PTB patients with emphysema needed a median TCC of 52.0 days [interquartile range (IQR) 29–66 days], which was significantly longer than 28.0 days (IQR 14–42 days) in 41 patients without emphysema (p < 0.001, log-rank test) (Fig. 3). Univariate Cox proportional hazards regression analysis showed that the following were associated with delayed TCC: emphysema (HR: 2.21; 95% CI: 1.36–3.60; p = 0.001); smoking history (HR: 1.81; 95% CI: 1.09–2.99; p = 0.021); consolidation (HR: 1.77; 95% CI: 1.06–2.96; p = 0.031); cavities (HR: 2.72; 95% CI: 1.65–4.49; p < 0.0001); hypoalbuminemia (HR: 1.74; 95% CI: 1.08–2.80; p = 0.024); and baseline TTD within 2 weeks (HR: 2.56; 95% CI: 1.52–4.31; p < 0.0001) (Table 3). Multivariate Cox proportional hazards regression analysis showed that the following were associated with delayed TCC after adjusting for age, emphysema, smoking history, cavities, consolidation, hypoalbuminemia, and baseline TTD within 2 weeks: emphysema (HR: 2.43; 95% CI: 1.18–4.97; p = 0.015); cavities (HR: 2.15; 95% CI: 1.18–3.89; p = 0.012); and baseline TTD within 2 weeks (HR: 2.95; 95% CI: 1.64–5.31; p < 0.0001) (Table 3).