Role of MIC levels of resistance to clarithromycin and metronidazole in Helicobacter pylori eradication

Abstract

Objectives

Antimicrobial resistance to clarithromycin and metronidazole significantly affects the cure rate of standard therapies for Helicobacter pylori infection. We tested whether different MIC levels of resistance to these antibiotics play a role in therapeutic efficacy.

Methods

This was a post hoc analysis of data from a therapeutic trial in which patients with antibiotic susceptibility testing (Etest) received first-line sequential therapy. The level of antibiotic resistance was classified according to MIC values into low (MIC from >0.5 to ≤8 for clarithromycin, and from >8 to ≤32 for metronidazole) and high (MIC from >8 to 256 mg/L for clarithromycin, and from >32 to 256 mg/L for metronidazole).

Results

Data from 1006 patients were included. There were 520 (51.7%) patients with susceptible strains, 136 (13.5%) with clarithromycin-resistant strains, 144 (14.3%) with metronidazole-resistant strains and 206 (20.5%) with clarithromycin-resistant and metronidazole-resistant strains. In the presence of double resistance, the cure rate was still high (38/41, 92.7%) when MIC levels were low and it was reduced (94/112, 83.9%) only when MIC levels of both antibiotics were high. The cure rates did not significantly differ between patients with single antibiotic-resistant strains, irrespective of MIC values, and those with susceptible strains.

Conclusions

We found that MIC levels of resistance to either clarithromycin or metronidazole play a role in H. pylori therapy outcome and that bacterial resistance becomes relevant in vivo when clarithromycin-resistant and metronidazole-resistant strains have high MIC values for at least one of these antibiotics.

Introduction

Primary antibiotic resistance to clarithromycin and metronidazole significantly affects the cure rate of standard eradication regimens for Helicobacter pylori infection.¹ However, there is evidence that bacterial eradication is still achieved in a definite percentage of patients, despite the presence of single or combined resistance to these antimicrobial agents.² In clinical practice, antibiotic resistance of H. pylori is generally assessed by using Etest, a culture-based method performed on gastric biopsies. According to the updated EUCAST recommendations, MIC values of 0.5 and 8 mg/L are the cut-offs above which H. pylori is deemed resistant to clarithromycin and metronidazole, respectively.³ However, in resistant H. pylori strains, MICs range widely from >0.5 to 256 mg/L for clarithromycin and from >8 to 256 mg/L for metronidazole.³ On these bases, it could be clinically valuable to test whether efficacy of eradication therapy depends on different MIC levels of resistance to these antibiotics rather than with an on/off effect. Indeed, administration of an antibiotic combination, together with a high-dose proton pump inhibitor such as in therapies used for H. pylori eradication, may exert a synergistic effect overcoming in vivo some levels of the bacterial resistance tested in vitro.

Therefore, we designed this study to test whether the efficacy of H. pylori eradication therapy may be affected by different MIC values of clarithromycin and/or metronidazole resistance.

Patients and methods

Patients

This study was a post hoc evaluation of a therapeutic trial in which consecutive patients with dyspeptic symptoms underwent upper endoscopy in a single centre (Bologna).⁴ Only data from patients with both successful bacterial culture/antibiogram and with a [¹³C]urea breath test 6–8 weeks following eradication therapy were included. All patients received sequential therapy with 40 mg of esomeprazole for 10 days, together with 1 g of amoxicillin for the first 5 days and with 500 mg of clarithromycin and 500 mg of tinidazole for the successive 5 days. All drugs were given twice daily.

Ethics

The initial study was approved by the Ethics Committee of the University of Bologna, Italy (47/2012/U/OSS).⁴ All patients gave informed consent to participate.

Bacterial culture and antibiotic susceptibility test

Biopsy sampling for histological and microbiological assessment (Etest; AB Biodisk, Solna, Sweden) has been reported elsewhere.⁴ According to the current EUCAST recommendations, bacteria were considered resistant when the the MIC of clarithromycin was >0.5 mg/L and when the MIC of metronidazole was >8 mg/L.³ For the purpose of the study, we arbitrarily classified the level of antibiotic resistance according to MIC values into low (MIC from >0.5 to ≤8 for clarithromycin, and from >8 to ≤32 for metronidazole) and high (MIC from >8 to 256 for clarithromycin, and from >32 to 256 mg/L for metronidazole).

Statistical analysis

Eradication rates and their 95% CIs at PP analysis were computed for each subgroup. Comparison of cure rates was performed using the χ² test and Fisher’s exact test, as appropriate. Differences were considered significant at the 5% probability level. Analyses were performed by using the StatSoft 7.1 program for Windows 10.

Results

We considered data from the first 1006 patients (mean ± SD age = 50.4 ± 14.6 years; male/female = 396/610; non-ulcer dyspepsia/peptic ulcer = 909/97) consecutively enrolled in the original study.⁴ Antibiotic resistance patterns for clarithromycin and metronidazole, the overall eradication rate and the demographic characteristics did not differ between the selected sample and the entire study population.⁴ A total of 520 (51.7%) patients were infected with strains susceptible to both clarithromycin and metronidazole. When considering clarithromycin-resistant strains (single or combined resistance; n = 342, 34.0%) a high level of resistance was detected in 181 (52.9%) and a low level in the remaining 161 (47.1%). When considering metronidazole-resistant strains (single or combined resistance; n = 350, 34.8%) a high level of resistance was detected in 254 (72.6%) and a low level in the remaining 96 (27.4%).

In patients with H. pylori strains resistant to either clarithromycin or metronidazole the cure rates following sequential therapy did not significantly differ among subgroups with different MIC levels of resistance as compared with those with susceptible strains; a success rate of >90% was achieved in any subgroup (Table 1). By considering patients with H. pylori strains having double antibiotic resistance, we found that the eradication rate (70/77, 90.9%; 95% CI = 84.5%–97.3%) in patients with H. pylori strains having a low level of clarithromycin resistance was not significantly (P = 0.064) reduced as compared with susceptible strains (498/520, 95.8%; 95% CI = 94%–97.4%). On the contrary, the cure rate was significantly lower (P = 0.0001) when clarithromycin MIC levels were high (109/129, 84.5%; 95% CI = 78.2%–90.7%) as compared with the cure rate for those with susceptible strains. As shown in Table 2, in the presence of double resistance, the highest cure rate (38/41, 92.7%; 95% CI = 84.7%–100.6%) was achieved in the subgroup infected with strains having the lowest level of resistance to both antibiotics, and the lowest eradication rate (94/112, 83.9%; 95% CI = 77.1%–90.7%) in those with the highest MIC levels.

Table 1.

Eradication rates according to MIC levels of resistance to either clarithromycin or metronidazole

	Low resistance	High resistance	Overall
Clarithromycin resistant/metronidazole susceptible, n/N (%)	77/84 (91.7)	48/52 (92.3)	125/136 (91.9)
Clarithromycin susceptible/metronidazole resistant, n/N (%)	37/38 (97.4)	105/106 (99.1)	142/144 (98.6)
Clarithromycin susceptible/metronidazole susceptible, n/N (%)	—	—	498/520 (95.8)

No statistically significant difference emerged among subgroups.

Table 2.

Eradication rates according to MIC levels of resistance to both clarithromycin and metronidazole

	Clarithromycin resistant, low	Clarithromycin resistant, high	Overall
Metronidazole resistant, low, n/N (%)	38/41 (92.7)	15/17 (88.2)	53/58 (91.4)
Metronidazole resistant, high, n/N (%)	32/36 (88.9)	94/112 (83.9)a	126/148 (85.1)a
Overall, n/N (%)	70/77 (90.9)	109/129 (84.5)a	179/206 (86.9)

a

P = 0.0001 when compared with susceptible strains.

Discussion

Although different factors are involved,⁵ antibiotic resistance is considered to be among the main factors affecting the efficacy of H. pylori therapy. This study aimed to evaluate not only the overall success of eradication therapy in patients with susceptible and resistant strains of H. pylori, but, only in resistant strains, to also detect any possible correlation between MICs of clarithromycin and/or metronidazole and eradication rates. For the first time, we found that MIC values of these antimicrobial agents play a relevant role in therapy outcome. Indeed, we observed that in the presence of dual clarithromycin and metronidazole resistance the overall cure rate is significantly reduced only in those H. pylori strains having high MIC values of either clarithromycin or metronidazole, whilst it remained distinctly high (92.7%) for strains with low MIC values. The lowest cure rate was observed in H. pylori strains with high MIC values of both antibiotics, even though the infection was still cured in 83.9% of these cases, at least following sequential therapy. Notably, we found that resistance to either clarithromycin or metronidazole—irrespective of MIC values—does not significantly affect the cure rate. Therefore, it can be hypothesized that the pre-treatment with amoxicillin and the simultaneous administration of clarithromycin and tinidazole, as in the sequential therapy, is able to overcome in vivo the single antibiotic resistance detected in vitro, suggesting a potential synergistic effect.⁶

The discovered role of MIC levels of resistance in the success rate of eradication therapy for H. pylori infection prompts at least two clinically relevant considerations. First, the usefulness of PCR-based tools for clarithromycin resistance evaluation is deeply undermined, when considering that levels of resistance are not assessed by these methods. Second, it is assumed there is a complete lack of effect of an antibiotic on H. pylori when a specific bacterial resistance is present in vitro.⁷ However, such a hypothesis is not corroborated by our clinical results in vivo, at least for the sequential therapy. Indeed, we found that the eradication rate is still >80% even in the presence of dual clarithromycin and metronidazole resistance with high MIC levels of resistance, which is distinctly higher than the 50% expected following only 5 days of amoxicillin therapy.⁸ Conversely, tailoring first-line therapy according to antibiotic susceptibility (as advised by some experts⁷) does not ensure bacterial eradication in all cases. Indeed, a recent meta-analysis found that susceptibility-guided antibiotic therapy achieved a cure rate of 89.2% for 860 patients.⁹ These observations would suggest that in vitro susceptibility testing does not accurately predict in vivo efficacy of associated antibiotics, at least when bacterial strains were categorized as ‘susceptible’ or ‘resistant’ based on a fixed cut-off. Therefore, clinicians could keep in mind the MIC values—besides the resistant status—of clarithromycin/metronidazole resistance when choosing an eradication therapy.

In conclusion, our data suggest that MIC levels of resistance to either clarithromycin or metronidazole are a matter for concern in H. pylori therapy and that bacterial resistance becomes relevant in vivo when strains with dual clarithromycin and metronidazole resistance have high MIC values of at least one of these antibiotics. These results would encourage further research that might lead to a re-evaluation of current MIC breakpoints.

Funding

This study was carried out as part of our routine work.

Transparency declarations

None to declare.