Evolution of Susceptibilities of Helicobacter pylori Strains Circulating in Cameroon to Usual Antibiotics: A Three-year Study

This study was conducted to estimate the evolution of antimicrobial susceptibilities of H. pylori strains isolated in Cameroon from 2014 to 2016 in relation to sociodemographic factors and clinical outcomes. A total of 278 H. pylori strains were isolated from patients with gastroduodenal disorders and tested for their susceptibility to nine antibiotics using the disc diffusion method. With time, a significant decreased of resistance was observed against clarithromycin (31.11 to 3.20%, p<0.0001), erythromycin (66.69 to 9.6%, p<0.0001) and metronidazole (86.67 to 69.6%, p<0.0001). Resistance to tetracycline (0 to 2.4%) and doxycycline (0 to 1.6%) slightly increase with time. Resistance to amoxicillin (95.5 to 94.4%), ciprofloxacin (4.44 to 4%) and levofloxacin (0 to 0.8%) was relatively constant. No resistance was detected against rifampicin with the time. A significant decrease of resistance was detected against metronidazole / clarithromycin (22.22 to 1.6%, p<0.0001) and amoxicillin / clarithromycin (26.67 to 3.2%, p<0.0001). Among the tested isolates, 0.92 and 2.4% elicited triple resistance to metronidazole / tetracycline / amoxicillin in 2015 and 2016, and 0.8% quadruple resistance to metronidazole / tetracycline / clarithromycin / amoxicillin in 2016. Higher resistance rate was noticed as age increase and among female compared to men. Our data showed evolution in the antimicrobial susceptibilities of H pylori strains circulating in our milieu with time. This finding highlights the need of monitoring periodically H. pylori resistance profile to antibiotics in order to determine the adapted treatment for this infection.


Introduction
Peptic ulcer disease is a major public health problem because of its high prevalence, its complications and its cost. Since the discovery of Helicobacter pylori by Warren and Marshall in 1982 [1], peptic ulcer disease is mostly curable with a week's course of antibiotics and antisecretory drugs [2,3]. Regimen used in H. pylori eradication are triple and quadruple therapy [4,5]. Triple-therapy regimen consists of proton pump inhibitor and two antibiotics: amoxicillin and clarithromycin, or metronidazole and clarithromycin while quadruple therapy consists of proton pump inhibitor, bismuth and two antibiotics: amoxicillin plus clarithromycin, or metronidazole plus tetracycline [4,5]. Other treatments have also been proposed, including metronidazole, as well as tetracycline, fluoroquinolones, and rifamycins [4,5]. However, these treatments, mainly because of resistance have become less effective. A recent study conducted throughout Europe shown that H. pylori resistance rates among adults were 17.5% for clarithromycin and 34.9% for metronidazole, and were significantly higher for clarithromycin and levofloxacin in Western/Central and Southern Europe (>20%) than those in Northern European countries (<10%) [6]. On the other hand, H. pylori resistance pattern within a particular population may varies with time. Resistance to metronidazole has increased from 26.7 to 47.2% in 2010 and 2015 respectively, while tetracycline resistance increased from 5.9 to 11.7% in the same period (7,8). In the same way, resistance to clarithromycin has been found to increase from 17.2% in 2010 to more than 19.7% in 2017 [7][8][9].
In Africa, resistance rate to MET of 90% was found by Seck in Senegal [10], 100% by Sherif et al. in Egypt [11], 56% by Chaabouni in Tunisia [12]. Against macrolides, resistance rate of 4 and 17.5% were found respectively in Egypt [11] and in Tunisia [12]. In Cameroon, resistance rate of 97.85% was found against MET [13]; 97.14 and 85.6% against amoxicillin respectively in 2018 and 2006; 13.57 and 44.7 against clarithromycin respectively in 2018 and 2006 [13,14]. Taken into account this widespread emergency of MDR H. pylori strains, the Maastricht consensus meeting IV recommended a regular and periodical investigation of H. pylori resistance profile to the most commonly used antibacterial medications in different countries [15]. Epidemiological studies has reveal that the prevalence of this infection ranged from 92.2% to 64.34 respectively in 2004 and 2016 [16] in Cameroon. However, there is not data on the timely progression of H. pylori antimicrobial susceptibility in Cameroon. Therefore, the present investigation was aimed at determining the antimicrobial susceptibility progression of H. pylori strains isolated from 2014 to 2016 to nine routinely used antimicrobial agents in order to adopt an antibiotic regional Programs. The relationship between the resistance of H. pylori isolates with age, sex and endoscopic clinical outcomes was also examined.

Sample Collection
H. pylori isolates were recovered from gastric biopsies of patients with gastro-duodenal disorders undergoing endoscopy at Laquintinie Hospital and General Hospital both in Douala, Cameroon from August-2014 to December-2016. None of the patients had received neither antimicrobial therapy nor proton pump inhibitors for at least 2 weeks prior to endoscopy. The specimens that tested positive for urease test were placed in the sterile tubes containing 3-4 ml sterile normal saline for isolation within 4 h.

Isolation and Identification of H. Pylori
The biopsies were ground and about two drops of homogenates were inoculated into supplemented Columbia agar (Columbia agar + 5% (v/v) lacked horse blood and 1% (v/v) Vitox) for 48 to 72 h under microaerophilic conditions. The colonies gotten from this were identified using Gram staining and urease, oxidase, catalase reactions. Isolates that exhibited Gram negative curved rods and were positive for catalase, oxidase and urease tests were considered as H. pylori. The isolates were suspended in Eppendorf tubes containing Brain Heart Infusion broth supplemented with 5% horse serum (BHI-serum) with 30% glycerol and stored at -80°C until used.
All isolates were removed from storage at −80°C and subcultured on supplemented Columbia Agar medium for two passages. 200 µl of inoculum from 48 hour H. pylori colonies was prepared at Mac Farlands standard turbidity 3 (6 × 10 8 CFU/ml) and used to seed supplemented Columbia agar plate. The discs of antibiotics were then placed and pressed on the inoculated agar surface. The place were then incubated at 37°C for 48 hours under microaerophilic conditions. Zones of lysis around the antibiotic discs were measured and the isolates were classified as sensitive or resistant according to the Microbiologist French Society guidelines [18]. The experiment was performed in triplicate and the mean diameters of inhibition zone recorded for each antibiotic was calculated.

Statistical Analysis
SPSS software, version 13 was used to perform statistical analysis. The significance of antibiotic resistance patterns between groups of patients according to socio demographic factors, clinical outcome was determined using Chi-square and Fisher's exact tests. Difference was considered statistically significant at p < 0.05.

Patients
Among the 842 patients enrolled, 616 were H. pylori positive. From these positive patients, 278 H. pylori clinical isolates was recovered, 152 from women and 126 from men. Their mean age was 43±17 years (range 15 to 60 years). Approximately thirty percent (29.14%) of these subjects were with superficial gastritis, 25.54% with duodenal ulcers, 12.59% with erosive gastritis and 4.32% with gastric cancer. The characteristics of patient are summarized in Table 1.
No resistance was detected against RIF with the time ( Figure  1 and Table 2). A significant linear decrease of dual resistance was detected against MET-CLR and AMX-CLR with resistance rate decreasing from 22.22 to 1.6% and from 26.67 to 3.2% respectively (p<0.0001 and p<0.0001 respectively) ( Table 2 and 3, Figure 1). In 2014, no case of triple or quadruple resistance were detected, whereas in 2015 and 2016, 0.92 and 2.4% cases of resistance to MET-TET-AMX was detected and 0.8% to MET-TET-CLR-AMX in 2016 (Table  2).

Factors Associated with Antibiotic Resistance
Increasing age was significantly associated with ERY (p=0.038) and DOX (p=0.0001) resistance ( Table 4). The resistance frequencies to LEV (p=0.0001), CLR (p=0.013) and ERY (p=0.043) was significantly associated with endoscopic outcome with a high resistance rate among patients with bulbar ulcer and gastric cancer for CLR, gastric ulcer and chronic gastritis for ERY. CLR (p=0.0001), MET (p=0.018) and ERY (p=0.0001) resistance were all significantly higher in subjects enrolled at HLD compared to those from HGD (Table 4).
A significantly higher resistance rate among female compared to men was detected against MET-AMX (p=0.001). Also, significant increase of resistance rate to MET-AMX was noticed as age increase (p=0.026). Again, subjects enrolled at HLD compared to those from HGD were significantly more resistant to AMX-MET (p=0.002) ( Table  4).

Discussion
The emergence of MDR H. pylori strains has become a serious challenge all over the world, as it is the major reason for the low rate of H. pylori eradication and the failure of the therapies both in developing [18,20] and developed countries [21,22]. This highlights the need for knowledge on the resistance pattern and their timely progression in order to adopt an antibiotic regional Programs. In the present investigation, we evaluated the evolution of antimicrobial susceptibilities of 278 H. pylori clinical isolates from Cameroon during a 3-years period (2014 to 2016).
In this study, 93.9% of the tested isolates were resistant to amoxicillin with the resistance rate fluctuation of 95.5 to 94.4% during the study period, indicating a constant high resistance to this drug. This observation is in line with previous studies in Cameroon, showing H. pylori high resistance rate of 85.6 and 97.14% respectively in 2006 [14] and 2018 [13]. This high resistance rate to amoxicillin could be the consequence of the wide use of this antibiotic in Cameroon due to its cheapness. Similar reports were noticed in Japan where the authors detected a stable trend of resistance to amoxicillin in their sample isolates with time [23]. However, it is in contrast with some studies revealing an increase trends of primary resistance to amoxicillin in South Korea [24] and a decrease rate from 7.9 to 0% among Bulgarian children in 1996-1999 and 2005-2008 [25].
Resistance rate of 10.75% and 34.5% respectively against clarithromycin and erythromycin was detected among our sample isolates (Table 2). We also found a significant decrease of resistance against CLR (p<0.0001) and ERY (p<0.0001) during the 3 year study. Similar resistance trend to clarithromycin is observed when putting together previous reports regarding resistance to this antibiotic in Cameroon. In fact, resistance rate to CLR of 44.7% and 13.57 was found respectively in 2006 and 2018 in Cameroon [13,14]. High cost of clarithromycin limits its use in our milieu [26], this may partially explained such decrease in the resistant rate of the tested isolates to this drug. Our results showed a simultaneous decrease of resistance to both clarithromycin and erythromycin with the time, this may be due to crossresistance in this class of antimicrobials as previously described [13]. Some data on the timely variability of the susceptibility of this pathogen to macrolides are available in the literature. A stable resistance trend to clarithromycin has been reported in UK and a decrease one in Netherlands, where antibiotic sailing is lower than in others EU countries [27,28]. Also in Sweden where national antibiotic use is lower, low H. pylori resistance rate of 1.0-1.5% to clarithromycin had been noticed [29]. Owing to the growing consumption of macrolides, H. pylori resistance has increased in many countries [30]. Increasing macrolide use for pulmonary, ear, nose and throat infections in Japan has been associated with growing clarithromycin resistance from 18.9% to 27.7% respectively between 2002-2003 and 2004-2005 [23]. Similarly, increase resistance rate has been reported against macrolide in Italy (2.1-fold over 15 years) [25], in Bulgaria (1.8-fold over 10 years) [25], in South Korea (4.9-fold over 10 years) [24] and in Japan (1.5-fold over 2 years) [23]. In South Korea, the primary clarithromycin resistance rate increasing from 0 to 13.8% in 1987 and 2003 has been reported [31]. In Bulgaria, the primary clarithromycin resistance increase of 10% from 1996 to 1999 and 18.4% from 2005 to 2008 was observed, probably owing to an increase in the national consumption of macrolides from 0.8 daily doses per 1000 inhabitants in 2004 to 1.8 daily doses per 1000 inhabitants in 2006 [32,33].
Our findings showed resistance rate of 74.5% against metronidazole and a significant decrease of resistance to the said antibiotic with time (p<0.0001). Previous reports in Cameroon regarding resistance to MET did not shown a similar resistance trend to this antibiotic. In fact, resistance rate of 93.2 and 97.85% respectively was revealed in 2006 and 2018 [13,14], suggesting little or no variation of the resistance against this drug. This high resistance is thought to be linked to the frequency of use of nitroimidazole derivatives in the treatment of digestive and genital parasitic infections [34]. Primary H. pylori resistance to metronidazole are reported to be stable in countries such as Bulgaria [24], Japan [23] and The Netherlands [27], unlike in South Korea where an increase resistance rate due to growing use or abuse of metronidazole for parasitic or female genital infections has been reported [31].
Resistance to tetracycline is not widespread. Low rates of 0.3, 1.7 and 2% were detected respectively in Sweden [35], Estonia [36] and Lebanon [37], which are slightly in accordance with the resistance rate of 1.4% detected in this study. Although tetracycline were slightly affected by resistance, low increasing of TET and DOX resistance from 0 to 2.4% and 0 to 1.6% respectively was detected in our milieu. In Cameroon, resistance rate of 43.9 and 2.86% had been reported in 2006 and 2018 [13,14], suggesting a drastic decrease of resistance to tetracycline from this time elapsed. However, our result as regard resistance rate to TET is similar to that reported in 2018, thus the differences between these resistance trends probably reflect the variation in tetracycline usage between our sample populations. Increasing primary H. pylori resistance to tetracycline has been described in South Korea and a stable one in Bulgaria due to decreasing tetracycline consumption from more than 4.2 daily doses per 1000 inhabitants in 1994-1999 to 2.4 daily doses per 1000 inhabitants in 2006 [24,31].
Although no resistant to fluoroquinolones has been early observed in Cameroon [13], this study showed low resistance rate of 3.85% (11/278) and 2.4% (3/278) respectively against ciprofloxacin and levofloxacin. Our results also showed that resistance to levofloxacin (0 to 0.8%) and ciprofloxacin (4.44 to 4%) was relatively constant from 2014 to 2016, which is seem to be similar to that observed in Bulgaria [24]. However, increasing resistance trend to levofloxacin of 5.1-fold in children and 4.2-fold in adults since 2004 was observed in Taiwan [38,39], whereas in South Korea it was ranged from 0 to 33.3% from 1987 to 2003 [31].
Regarding multi-drug resistance pattern, our finding showed that 75% of the isolates were resistant to at least two antibiotics, which is in accordance with the 70% previously reported in Cameroon [13]. This high prevalence of MDR phenotype among H. pylori isolates circulating in Cameroon may be attributed to the exhaustive use of antibiotics across the country. Dual resistance to MET-AMX, MET-CLR, MET-TET, AMX-CLR, AMX-TET and CLR-TET was detected among the isolates with a significant decrease against MET-CLR and AMX-CLR. In contrast, increasing double resistance rates to MET-CLR from 2.8 to 3.1% and from 5.0 to 11.2% had been observed at two centers in the UK, and a stable resistance pattern in Bulgaria [40]. Also, Pereira et al. in their study detected two clinical isolates with resistance to MET-CLR and three to metronidazole-clarithromycinciprofloxacin simultaneously [41]. Resistance rate range from 0 to 0.92% and to 2.4% was detected against MET-TET-AMX respectively in 2015 and 2016 and 0.8% against MET-TET-CLR-AMX in 2016, indicating an increase of simultaneous resistance with time. Few studies reporting similar low quadruple-drug resistance rate from India (2.5%), Bulgaria (0.7%), Vietnam (1.9%), and Indonesia (2.6%) has been documented [24,42,43]. However, these resistance rate were lower than the 15.71 and 5.71% triple and quadruple drugs resistance reported in Cameroon [25] and the 37.9% reported in Chile [42].
Our data also showed a significantly higher resistance rate against AMX-MET among female compared to men (p=0.001). Study conducted in Costa Rica [44] and Mexico [45] also reported similar distribution of resistance to MET according to sex. The widespread use of this antibiotic by women for gynecological infections may explained such observation [34]. Our results also showed a high resistant rate among older compare with younger (p=0.026), may be due to the fact that increase in age enhance the chance of exposure to resistance strains.

Conclusion
The present study showed that the resistance patterns of H. pylori isolates from Cameroon has changed within the 3 years study. Our data showed significant decrease in resistance of H pylori isolates against CLR, ERY and MET, a slightly increase of resistance to TET and DOX, stable state of resistance to amoxicillin, ciprofloxacin and levofloxacin, and a significant decrease of dual resistance against MET-CLR and AMX-CLR from 2014 to 2016 in our milieu. Women and older were the groups showing higher resistance rates to antibiotics. Increasing resistance rates to TET and DOX should be considered as an indicator of abuse and overuse of these antibiotics that ultimately limit their effectiveness for future generations. Even though the resistance to CLR and MET is decreasing, it continues to be worryingly high. Therefore, the frequency of H. pylori resistance to current antibiotics should be monitored periodically to determine the adapted treatment for this infection.

Author Contributions
Kouitcheu Mabeku LB designed the work; Fotso CS, Malongue A and Ateba MR examined the patients, collected patient's clinical data and biopsies sample; Eyoum BB and Tali NLD carried out the analysis; Kouitcheu Mabeku LB and Kuiate JR analyzed the data and wrote the paper.

Declarations
All the authors do not have any possible conflicts of interest.