Introduction: Tuberculosis is an infectious disease that continues to plague the world today, causing concerns due to its high mortality rate. The therapy regimens used for the treatment of tuberculosis today have demonstrated high efficacy and safety, potentially reducing the disease’s burden, but the use of some standardized medications has caused many resistances to emerge. Over the last decade, researchers have been looking for suitable alternatives, with quinolones emerging as the most promising candidate due to their efficacy, safety, and availability. However, their efficacy as a first-line treatment remains debatable.

Aim: This study is focused on assessing the efficacy and safety of quinolone therapy combined with other regimens for pulmonary TB first-line treatment.

Materials and methods: Literature search was conducted in the databases of Cochrane, PubMed, ScienceDirect, and Google Scholar using the Boolean keywords of “tuberculosis”, “treatment”, and “quinolone”. Statistical analyses were performed using ReviewManager 5.4.1 and shown as forest plots of odds ratios.

Results: We analyzed 13 studies in this study. There were no significant differences between the study group and controls in the conversion rate ≤8 weeks (p=0.07, 95% CI 0.9–13.01), the success rate (p=0.80, 95% CI 0.57–1.54), adverse events (p=0.13, 95% CI 0.91–2.13), mortality rate (p=0.25, 95% CI 0.68–1.11), and recurrence rate (p=0.39, 95% CI [0.74–2.12]) .

Conclusions: According to the forest plot analysis, each of the results is non-significant. This concludes that the efficacy and safety of quinolone tended to be used as the second-line of choice of pulmonary tuberculosis treatment compared to the control group.

efficacy, first-line treatment, quinolone, safety, tuberculosis

Tuberculosis (TB) is an infectious disease that keeps negatively affecting global health due to its high fatality rate.‌^[1] According to the World Health Organization (WHO), there were an estimated 10.6 million people suffering from tuberculosis (TB) in 2021, accounting for 1.6 million deaths due to the disease’s progression. According to the statistics, TB care is still inadequate around the world, both in terms of facilities and public awareness.^[2]

Nowadays, the therapeutic regimens used in the treatment of tuberculosis have been proven to be so efficacious and safe that they might prevent the disease’s burden; however, when the standardized drugs were introduced in the management of the disease, they caused many resistances to emerge and became a challenge. Therefore, there was an extensive search for an applicable alternative in the past decade, with quinolone as the most potential candidate due to its efficacy, safety, and availability. Gatifloxacin, moxifloxacin, and levofloxacin are the most numerous quinolones combined with the standard regimen of rifampin (R), isoniazid (H), pyrazinamide (Z), and ethambutol (E).^[3] There were many preceding meta-analyses showing the promising efficacy and safety of the use of these drugs in TB treatment. However, its potency as the first-line treatment is still debatable. As an instance to this, a study has shown that the quinolone-containing regimen had a higher rate of sputum culture conversion in 2 months of treatment but had less favorable outcome.^[4]

Due to its vagueness, we performed a meta-analysis to unveil the efficacy and safety of quinolone in TB treatment, especially if the drugs were combined with first-line standardized regimen.

A total of 717 papers were obtained in the literature search, 706 of which were excluded due to various reasons referred to in Fig. 1. Thus, a total of 11 eligible studies were included in the meta-analysis.

Figure 1.

PRISMA chart of the study.

All data in these studies are associated with the effects of any of the quinolones used as first-line therapy, whether it is moxifloxacin, gatifloxacin, or even a standardized levofloxacin usually used in second-line treatment of pulmonary tuberculosis in nowadays guidelines. All studies were reported as RCT, and the outcomes of adverse events, mortality rate, and recurrence rate or treatment failure were also recorded. The characteristics of each study are presented in Table 2.

Conversion rate ≤8 weeks

The sputum conversion is defined as a negative status of the mycobacterial content in the sputum at the time it was examined. The analysis was assessed within the 2 months or 8 weeks span after treatment initiation. Nine studies were included in this analysis with random-effects model used to calculate the overall OR in 95% CI, since there was a significant heterogeneity (I²: 98%). The overall effect OR in this outcome was 3.43 (95% CI 0.90–13.1, p=0.07), showing that the use of quinolones combined with other regimens improve the time of sputum conversion rate to negative. However, this was nonsignificant, concluding that there were no notable differences in sputum conversion time compared to the control group. The forest plot is shown in Fig. 2.

Figure 2.

Forest plot of conversion rate analysis.

Success rate

Success rate was considered as completing treatment or stated as being cured by physicians. Eight studies were included in this analysis with random-effects model used to calculate the overall OR in 95% CI, since there was a significant heterogeneity (I²: 74%). An OR of 0.94 was obtained (95% CI 0.57–1.54, p=0.80), suggesting that the use of quinolone addition to other drugs combination decrease the success of treatment compared to the standardized fixed-dose combination control. Nevertheless, this was insignificant, concluding that there were no notable differences in treatment completion or outcome compared to the control group. The forest plot analysis of success rate is shown in Fig. 3.

Figure 3.

Forest plot of success rate.

Adverse events

Adverse events were defined as any of unexpectable outcomes found during the treatment period. This may be skin rashes, elevated liver enzymes, QT interval prolongation, and many other clinical manifestations or laboratory anomalies. Eleven studies included in this analysis with random-effects model used to calculate the overall OR in 95% CI, since there was a significant heterogeneity (I²: 85%). The analysis’s OR was 1.39 (95% CI 0.91–2.13, p=0.13), suggesting that using quinolones in combination causes more adverse events to occur. The p-value in the analysis, however, inferred that there were no significant differences in the outcome of adverse events in both groups. The forest plot is shown in Fig. 4.

Figure 4.

Forest plot of adverse events analysis.

Mortality rate

Mortality rate was considered as death related to treatment. Nine studies were analyzed with fixed-effects model to calculate the overall OR in 95% CI due to insignificant heterogeneity (I²: 14%). An OR of 0.87 (95% CI 0.68–1.11, p=0.25) simply showed that the mortality rate of subjects using quinolones as part of their regimens was lower compared to that of the control group but still failed to reach statistical significance. Thus, the mortality rates in both groups were about just the same. The forest plot analysis is shown in Fig. 5.

Figure 5.

Forest plot of mortality rate.

Recurrence rate

Tuberculosis treatment is prone to recurrence after completion. In this study, we analyzed a total of 5 studies with fixed-effects model to calculate the overall OR in 95% CI due to insignificant heterogeneity (I²: 10%). The OR was 1.26 (95% CI 0.74–2.12, p=0.39), suggesting that the experimental group tended to have recurrences of the disease more often than the control group. Nonetheless, there were no significant differences in the recurrence rate between the two groups. The forest plot analysis is shown in Fig. 6.

Figure 6.

Forest plot of recurrence rate.

This meta-analysis, which includes 11 eligible studies, was conducted to assess the efficacy and safety of quinolone combined with other regimens administered primarily in pulmonary tuberculosis. This therapy was meant to be administered as medications given without any preceding therapy after diagnosis had been made. The primary outcome of efficacy was measured with the conversion rate of sputum in less than 8 weeks with OR of 3.43 with non-significant difference (p=0.07), suggesting that addition of quinolones, such as moxifloxacin, gatifloxacin, or levofloxacin, may enhance the activity of bacterial elimination in comparison with the standardized FDC. However, this may not be an important finding because it was non-significant, suggesting that there was no difference in the bacterial elimination activity in both groups.

For secondary outcomes to assess safety, the success rate, adverse events, mortality rate, and recurrence rate were taken into account. Success rate showed OR about 0.94 (p=0.80) suggesting that the addition of quinolone into other first-line regimens lowering patients’ adherence to therapy compared to the nowadays recommended regimens. This was non-significant, too, so there was no difference between the experimental group and control group in terms of patients’ adherence or outcome. The OR was 1.39 (p=0.13) for adverse events implying that the use of quinolone added to other regimens may add other adverse events compared to the RHZE regimens alone, but this was non-significant for both groups. The mortality rate statistical analysis shown OR of 0.87 (p=0.25), meaning that the use of additional quinolone to other regimens may be safer or decrease the rate of death related to therapy compared to the standardized regimen group. However, this was also insignificant so the safety of both compared regimens were no different. In recurrence rate, the OR was 1.26 which shows that regimens including quinolones tend to recur after the therapy has been completed compared to the FDC-treated group. This suggests that quinolone may be more prone to develop resistance compared to RHZE. Nonetheless, this too, was also non-significant so there was no difference in resistance tendency between both groups.

Some previous similar studies of meta-analysis also compared groups of quinolone-combined regimens and RHZE. The results vary between studies. In a previous meta-analysis^[16] assessing similar outcomes, with 9 eligible studies included, revealed that the use of moxifloxacin in the recommended regimen improved the rate of sputum culture conversion compared to the control group within 2 months of treatment, indicating that addition of moxifloxacin could enhance the antibacterial effect. The findings are inconsistent in the present study. The discrepancies in the outcome of this present meta-analysis may be due to the varying first-line treatment combined with varying quinolone regimens, where the experimental group drugs including other drugs of choices, such as bedaquiline or pretomanid, while the earlier study only used H, R, Z, or E with one of the drugs switched with moxifloxacin or the regimen alone was added with moxifloxacin directly. The difference in exclusion and inclusion criteria along with the sample size of eligible studies (nine in the present study versus six in the previous study) might also result in inconsistent findings.

An older review of literature made by Chen et al.^[17], with 6 eligible studies, demonstrated that the addition of moxifloxacin did not significantly improve the rate of sputum culture conversion compared to the control group within 2 months of treatment. These findings are consistent with the current study, which found that adding moxifloxacin did not increase the rate of sputum culture conversion compared to the control group within two months of treatment.‌^[17] This implies that the addition of any quinolone regimens to first-line agents is not as effective due to the high potency of the first-line drugs alone in bactericidal activity.

Regarding adverse events, no significant differences were found in these studies, showing that the addition of any quinolone might not add adverse events during the launch of treatment. The recurrence rate in present study is also showing similar trend - no significant differences were found between the experimental group and the controls. On the other hand, the two previous studies have shown that the use of quinolone might reduce relapse cases. The most likely cause of this inconsistent finding, like in the outcome of efficacy, was the use of varying first-line treatments combined with varying quinolone regimens, regardless of the drug’s potency in specific populations or settings where certain drugs are most potent. Furthermore, the difference in the study methodology may also yield this inconsistent finding.

Quinolones, known also as fluoroquinolones, are synthesized drugs made from structural modification of the 4-oxo-1,4-dihydroquinoline or 1,8-naphthyridine nucleus. The fluorination of these molecules at position 6 will usually omit the product of fluoroquinolone, such as levofloxacin, moxifloxacin, and gatifloxacin which have antituberculosis activity due to its C-8 methoxy moiety. In general, these drugs are used to treat MDR-TB. It has been approved that the use of quinolone (ofloxacin and ciprofloxacin) is effective with success rates of about 70% for MDR-TB cases. Furthermore, levofloxacin may substitute ofloxacin in some cases, showing superior efficacy. Also, moxifloxacin and gatifloxacin are effective in the MDR-TB treatment due to its lower mutant prevention concentrations for tuberculosis and should have greater potential to prevent the development of drugs resistances. In DS-TB, the most common reason to include quinolone in first-line treatment is due to standardized regimens-induced hepatotoxicity. In case of definitive treatment, ofloxacin or levofloxacin may be used in combination with R or H, or perhaps even pyrazinamide, depending on the liver function status. A study also stated a benefit of using quinolone, showing that there is no additional hepatic insult if levofloxacin and moxifloxacin are used. Aside from this, the lack of pharmacological interactions between drugs proved to be advantageous. However, random regimens could not be used.^[18]

There are some limitations in this meta-analysis. Firstly, this study includes all drugs used primarily in tuberculosis treatment, which may distort the potency of the most effective combination regimen between quinolone and nonquinolone drugs. Secondly, this study does not differentiate the subjects with DS or MDR status. Lastly, this study is a retrospective study which means that this meta-analysis has methodological limitations.

In conclusion, the efficacy and safety of quinolone tended to be used as the second-line choice of pulmonary tuberculosis treatment compared to the control group. However, additional studies should be conducted in the future to define the clear capabilities of quinolone in tuberculosis therapy.