Table 2.
Potential predictors of bacterial burden in longitudinal sputum and face-mask samples over 24 h
24-h face-mask IS6110 output |
24-h face-mask RD9 output |
|||
---|---|---|---|---|
Correlation coefficient (95% CI) | p value | Correlation coefficient (95% CI) | p value | |
Age | −0·26 (−0·61 to 0·17) | 0·21 | −0·13 (−0·52 to 0·30) | 0·53 |
Gender | .. | 0·89 | .. | 0·47 |
HIV status | .. | 0·48 | .. | 0·68 |
CD4 count* | −0·28 (−0·65 to 0·20) | 0·23 | −0·11 (−0·53 to 0·36) | 0·65 |
Duration of symptoms, weeks | 0·18 (−0·26 to 0·55) | 0·41 | 0·11 (−0·31 to 0·48) | 0·56 |
Chest radiography grade† | −0·22 (−0·59 to 0·23) | 0·32 | −0·07 (−0·48 to 0·36) | 0·75 |
Presence of cavitation on chest radiography† | .. | 0·52 | .. | 0·38 |
Sputum acid-fast bacilli grade‡ | −0·08 (−0·54 to 0·41) | 0·75 | −0·02 (−0·49 to 0·47) | 0·95 |
Sputum Xpert MTB/RIF grade | 0·02 (−0·40 to 0·43) | 0·92 | −0·04 (−0·45 to 0·37) | 0·84 |
Sputum culture§ | 0·14 (−0·38 to 0·59) | 0·58 | 0·07 (−0·44 to 0·54) | 0·79 |
Cough frequency | 0·10 (−0·33 to 0·50) | 0·63 | 0·07 (−0·35 to 0·47) | 0·73 |
Patient's perception of cough severity, visual analogue scale¶ | 0·28 (−0·18 to 0·63) | 0·21 | 0·31 (−0·15 to 0·65) | 0·53 |
Started tuberculosis treatment | .. | 0·68 | .. | 0·91 |
24-h sputum PCR signal content | 0·17 (−0·30 to 0·57) | 0·47 | 0·15 (−0·31 to 0·56) | 0·49 |
Data represent the ability of variables to predict levels of IS6110 and RD9 signals detected in face-mask samples over 24 h, analysed by Spearman's correlation for continuous variables and the Mann-Whitney U test for categorical data.
CD4 count was recorded for all 20 HIV-positive patients.
Chest radiography grade and presence of cavitation was recorded for 23 patients.
Sputum acid-fast bacilli grade available for 17 patients.
Sputum culture results available for 16 patients.
Visual analogue scale recorded for 22 patients.