Recurrence of Helicobacter Pylori after Successful Eradication in Northwestern China: Rates, Risk Factors, and Resistance Profiles of Recurrence Isolates Authors: Min Niu, Xiaoming Su, Zhen Ma, Lulu Li Published: 2026 ABSTRACT PurposeThis study aimed to investigate the recurrence rate of H. pylori after successful eradication in Ningxia, to identify independent risk factors associated with recurrence, and to characterize the antibiotic resistance profiles of isolates obtained from recurrent patients.MethodsThis study is a retrospective cohort study. A total of 891 H. pylori-infected patients who received initial eradication treatment and completed 1-year follow-up from June 2021 to October 2024 were included. Demographic characteristics, lifestyle, family hygiene habits, and antibiotic exposure history were collected through questionnaires. A total of 27 strains successfully isolated and cultured from patients with recurrence were tested for antibiotic susceptibility using the Kirby-Bauer disk diffusion method.ResultsThe H. pylori recurrence rate was 4.1% (37/891). Multivariate logistic regression identified six independent risk factors: rural residence (OR = 3.52, P = 0.003), smoking (OR = 6.06, P < 0.001), frequent shared tableware (OR = 3.46, P = 0.013), frequent dining out (OR = 7.41, P = 0.008), poor dining-out hygiene (OR = 4.50, P = 0.007), and having an H. pylori-infected household member (OR = 5.68, P < 0.001). Of the 37 patients with recurrence, 27 had successful H. pylori culture and obtained susceptibility results. The resistance rates to metronidazole (MET), clarithromycin (CLA), levofloxacin (LEV), and amoxicillin (AMO) were 96.3%, 77.8%, 62.9%, and 3.7%, respectively; all isolates were susceptible to furazolidone (FUR) and tetracycline (TET). Multidrug-resistance was observed in 51.8%, mainly as MET+CLA+LEV pattern (48.1%).ConclusionThe one-year recurrence rate of H. pylori in Ningxia is 4.1%. Factors significantly associated with an increased risk of recurrence include rural residence, smoking, shared tableware, regular dining out, poor dining-out hygiene, and household infection. Isolates from patients with recurrence exhibited high resistance rates to MET, CLA, and LEV, predominantly in multidrug-resistant patterns, whereas resistance to AMO, FUR, and TET remains low. Introduction Helicobacter pylori (H. pylori) is a Gram-negative, spiral-shaped, microaerophilic pathogen that plays a critical role in the pathogenesis of chronic gastritis, peptic ulcer disease, mucosa-associated lymphoid tissue (MALT) lymphoma, and gastric adenocarcinoma.1 Classified as a Group 1 carcinogen by the World Health Organization, H. pylori remains the only prokaryotic organism unequivocally recognized as a human carcinogen.2 Current international and domestic consensus guidelines recommend eradication therapy for all infected individuals without contraindications, irrespective of symptomatology.1,3 Successful eradication not only promotes healing of active gastritis and peptic ulcers but also attenuates the progression of gastric atrophy and reduces the long-term risk of gastric cancer.3 Currently, standard quadruple therapy with bismuth or dual therapy combining vonoprazan and amoxicillin achieves eradication rates approaching or exceeding 90% in first-line treatment.4 Nevertheless, recurrence following successful eradication remains a clinical concern, as it is closely associated with the relapse of peptic ulcers and MALT lymphoma.5 Consequently, elucidation of the risk factors and resistance profiles that contribute to post-eradication recurrence is of paramount importance for refining therapeutic strategies and minimizing the long-term clinical and economic burden of H. pylori infection. Recurrence of H. pylori following successful eradication can be classified into two distinct types based on its origin: recrudescence of the original strain and reinfection with a different strain. Recrudescence occurs when the initial treatment fails to completely eliminate the original strain, resulting in a positive test result during follow-up; this essentially represents treatment failure, whereas reinfection refers to the acquisition of a genetically distinct strain after the original organism has been completely cleared.5,6 Several studies have demonstrated that genotyping methods, including whole-genome sequencing (WGS) and DNA fingerprinting, can reliably distinguish recrudescence from reinfection. However, their translation into routine clinical practice is constrained by high costs, technical complexity, and stringent laboratory requirements, which currently preclude large-scale implementation.5,7,8 Furthermore, serological IgG antibody testing has also been shown to be useful for indicating H. pylori recurrence.9,10 Lee et al demonstrated that following successful H. pylori eradication, serum IgG antibody titers typically decline gradually over a period of 12 to 18 months. In their 18-month follow-up of 185 patients with confirmed eradication, the mean IgG titer fell from a baseline of 2.06 to 0.63. In contrast, patients who experienced reinfection showed a marked and abrupt elevation in antibody titers, which correlated significantly with the results of invasive diagnostic tests. These findings suggest that serological IgG testing may serve as a useful tool for monitoring H. pylori recurrence.9 Nevertheless, this approach has several limitations. It does not distinguish between recrudescence and reinfection, and it requires continuous longitudinal monitoring to confirm eradication status, which restricts its feasibility in routine clinical practice. Given its operational simplicity, however, a time-based classification has been widely adopted in clinical practice: recurrence occurring within one year after successful eradication is generally considered recrudescence, while recurrence beyond one year is more likely to represent true reinfection.5,6,11 Therefore, all recurrence events in this study were regarded as recrudescence. It should be noted that this 1-year time threshold is merely an empirical convention in clinical practice, and its accuracy has not been verified by molecular typing. The classification of “recrudescence” and “reinfection” based on this time threshold may be subject to misclassification bias. Nevertheless, the recurrence rate after H. pylori eradication is negatively correlated with the Human Development Index (HDI). In countries with very high, high, and medium/low HDI, the recurrence rates are 3.1%, 6.2%, and 10.9%, respectively.12 A study in the southern coastal area of China reported a one-year recurrence rate of 3.5% and identified contact with other infected individuals, frequent dining out, and irregular diet as independent risk factors for recurrence.6 These behavioral factors are closely related to socioeconomic development levels, further supporting the impact of HDI on recurrence rates. In addition, recurrence rates and their associated factors vary considerably across regions, depending on the survey area, population characteristics, sample size, socioeconomic status, and study period.12,13 Furthermore, antibiotic resistance is considered a major cause of eradication failure,3 and eradication failure is an important component of recurrence. Previous studies in the Ningxia region have shown that both the infection rate and resistance rate of H. pylori are at a relatively high level;14,15 the baseline resistance rates of clarithromycin (CLA) and levofloxacin (LEV) were close to 40%, while that of metronidazole (MET) exceeded 90%.14 This may compromise the efficacy of empirical regimens containing these agents, thereby increasing the risk of both eradication failure and subsequent recurrence. However, Ningxia is a multi-ethnic region with distinct dietary and hygienic practices. Whether and how these modifiable behavioral factors contribute to recurrence remains unclear, and identifying them would have direct public health implications for community-based health education and behavioral interventions. Furthermore, large-scale studies on recurrence after eradication in this area remain lacking, and the resistance profiles of recurrent isolates have yet to be characterized. Therefore, the aim of the study is to investigate the recurrence rate of H. pylori and its associated risk factors, and to characterize the resistance profiles of isolates from patients with recurrence. Materials and Methods Patients and Design This retrospective cohort study consecutively enrolled 1176 patients with H. pylori infection who received their first eradication treatment at the Department of Gastroenterology, People’s Hospital of Ningxia Hui Autonomous Region, or the Fifth People’s Hospital of Ningxia, between June 2021 and October 2024, using uniform inclusion and exclusion criteria to minimize selection and information bias. Part of the research data was derived from our previously published studies.16,17 All participants provided written informed consent prior to undergoing eradication therapy and H. pylori culture. The study protocol was approved by the Ethics Committee of Ningxia People’s Hospital (Approval Nos. 2021LL017 and 2022LW006) and was conducted in accordance with the principles of the Declaration of Helsinki. To minimize information bias, questionnaires were administered face-to-face by professional researchers using standardized questions. A structured questionnaire was administered to collect information on demographic characteristics, personal behaviors and lifestyle factors, family hygiene habits, household H. pylori infection status, and previous antibiotic exposure. Demographic variables included gender, age, education level, and monthly income. Lifestyle and behavioral factors comprised living area, smoking, alcohol consumption, pickled food intake, dairy foods consumption, and drinking tea history. Family-related variables included sharing of tableware and cups, washing tableware with detergent, frequency of dining-out, hygiene conditions of dining-out, number of family members, and H. pylori infection status of household members. Frequency of lifestyle exposures was categorized as occasional or regular, where occasional was defined as less than 1–2 times per week or 1–3 times per month, and regular as at least 3 times per week or almost daily. Inclusion and Exclusion Criteria Patients were eligible for inclusion if they met all of the following criteria: (1) confirmed H. pylori infection with initial eradication treatment; (2) age between 18 and 75 years; and (3) no use of proton pump inhibitors, histamine receptor antagonists, potassium-competitive acid blockers, bismuth-containing agents, antibiotics, or non-steroidal anti-inflammatory drugs within 4 weeks prior to the 14C-urea breath test (14C-UBT). Exclusion criteria were: (1) age <18 or >75 years; (2) refusal to undergo the 14C-UBT; (3) confirmed gastrointestinal malignancy or history of digestive tract surgery; (4) previous or ongoing radiotherapy or chemotherapy; (5) known allergy to relevant drugs; (6) pregnancy or lactation; and (7) participation in other clinical trials. Successful eradication was defined as 14C-UBT reexamination 6–8 weeks after the completion of eradication therapy, and if the test result was negative, it was considered as successful eradication. Of the 1012 patients who achieved successful eradication, 891 completed the 1-year follow-up and provided complete data, and were thus included in the subsequent analysis. The remaining 121 patients (11.9%) were lost to follow-up because of refusal to retest 14C-UBT (n = 48) and loss of contact (n = 73). Since patients who were lost to follow-up did not provide complete follow-up data, we were unable to directly compare their baseline characteristics with those of the population who completed follow-up, so the study could not provide a quantitative assessment of potential bias introduced by loss to follow-up. It should be noted that the rate of loss to follow-up was about 11.9%, which was within the acceptable range, and the reasons for attrition were primarily objective, including refusal to undergo retesting and loss of contact. The 14C-UBT was used to monitor H. pylori status during follow-up. The annual recurrence rate was calculated, and comparisons were made between patients with and without recurrence in terms of baseline demographics, socioeconomic characteristics, lifestyle factors, and prior antibiotic exposure. H. pylori culture and antimicrobial susceptibility testing were performed on isolates from 37 patients who experienced recurrence, of whom 27 yielded valid resistance results. In vitro Culture of H. Pylori and Drug Sensitivity Test Antibiotic susceptibility testing was performed using the disk diffusion method (Kirby-Bauer). Pure colonies were selected and suspended in sterile physiological saline to a density of 0.5 McFarland standard. The bacterial suspension was evenly inoculated onto Columbia blood agar plates without antibiotics, and antimicrobial disks (Oxoid, UK) were placed on the agar surface. The plates were incubated under microaerobic conditions at 37°C for 48 hours. After incubation, the diameters of the inhibition zones around each disk were measured, and the susceptibility of each isolate was determined according to the following criteria: resistance was defined as zone diameters ≤ 13 mm for CLA, ≤ 13 mm for LEV, ≤ 14 mm for AMO, ≤ 14 mm for FUR, ≤ 14 mm for TET, and ≤ 16 mm for MET; sensitivity was defined as zone diameters ≥ 18 mm for CLA, ≥ 17 mm for LEV, ≥ 17 mm for AMO, ≥ 17 mm for FUR, ≥ 19 mm for TET, and ≥ 21 mm for MET.18 The reference strain H. pylori ATCC 43504 was used as a quality control strain, in accordance with the CLSI M45 guidelines. All in vitro culture and antimicrobial susceptibility testing procedures were performed at Shanghai Xinchao Medical Laboratory, China. Statistical Analysis This was a retrospective cohort study, and no a priori sample size calculation was performed. Statistical analyses were performed using SPSS version 27.0 (IBM Corp., Armonk, NY, USA). Continuous variables were expressed as mean±standard deviation (SD) and compared using t-test or one-way analysis of variance, as appropriate. Categorical variables were presented as frequencies (n) and percentages (%) and compared using the Pearson chi-square test or Fisher’s exact test, as appropriate. Variables with P value less than 0.05 in the univariate analysis were entered into a multivariate logistic regression model to identify independent risk factors, including residential area, smoking history, shared tableware use, dining-out frequency, eating-out hygiene, and household H. pylori infection. The events-per-variable (EPV) ratio was approximately 6.2 (37 events/6 variables), which falls at the lower bound of the recommended range (≥ 5–10). Results of the multivariate analysis were reported as odds ratios (OR) with 95% confidence intervals (CI). A two-sided P < 0.05 was considered statistically significant. Results A total of 1,012 patients who received first-line H. pylori eradication therapy and tested negative on the 14C-UBT at follow-up were initially enrolled. Among them, 891 patients completed the 1-year follow-up assessment, including both the 14C-UBT and the questionnaire survey. Of these 891 patients, 854 tested negative for H. pylori, yielding a non-recurrence rate of 95.8%, while 37 tested positive, corresponding to a recurrence rate of 4.1%. All 37 patients with recurrence underwent antimicrobial susceptibility testing, with valid results obtained for 27 of them. Figure 1.Figure 1Flowchart of this study.A flowchart of H. pylori eradication study with patient outcomes and follow-up results. Flowchart of this study. Baseline Characteristics of Patients Completed Follow-Up A total of 891 patients who completed the follow-up were included in the analysis. Among them, 37 patients (4.1%) were in the recurrence group and 854 patients (95.8%) were in the non-recurrence group. There were no statistically significant differences in gender, age, educational level, and monthly income between the recurrence group and the non-recurrence group (all P > 0.05). Table 1.Table 1Comparison of Baseline Characteristic Between Patients with Recurrence and Non-RecurrenceTotal (n=891)Recurrence Group (n=37)Non-Recurrence Group (n=854))PGender0.308 Male481 (54.0)23 (62.2)458 (53.6) Female410 (46.0)14 (37.8)396 (46.4)Age (years)0.435 <40212 (23.8)12 (32.4)200 (23.4) 40–60471 (52.9)18 (48.6)453 (53.0) >60208 (23.3)7 (18.9)201 (23.6)Education level Junior375 (42.1)15 (40.5)350 (42.2)0.586 Senior397 (44.6)19 (51.4)378 (44.3) Undergraduate119 (13.4)3 (8.1)116 (13.6)Monthly income (yuan)0.322 <5000634 (71.2)29 (78.4)605 (70.8) >5000257 (28.8)8 (21.6)249 (29.2)Antibiotics regimen AMO+CLA156 (17.5)13 (35.1)143 (16.7) AMO+LEV156 (17.5)12 (32.4)144 (16.9) AMO+FUR157 (17.6)4 (10.8)153 (17.9) AMO+TET145 (16.3)0 (0.0)145 (17.0) TNZ+CLA128 (14.4)7 (18.9)121 (14.2) AMO(HDDT)149 (16.7)1 (2.7)148 (17.3)Abbreviations: MET, metronidazole; LEV, levofloxacin; CLA, clarithromycin; AMO, amoxicillin; FUR, furazoldone; TET, tetracycline; TNZ, tinidazole; HDDT, high-dose dual therapy; P, P-value. Table 1Comparison of Baseline Characteristic Between Patients with Recurrence and Non-RecurrenceTotal (n=891)Recurrence Group (n=37)Non-Recurrence Group (n=854))PGender0.308 Male481 (54.0)23 (62.2)458 (53.6) Female410 (46.0)14 (37.8)396 (46.4)Age (years)0.435 <40212 (23.8)12 (32.4)200 (23.4) 40–60471 (52.9)18 (48.6)453 (53.0) >60208 (23.3)7 (18.9)201 (23.6)Education level Junior375 (42.1)15 (40.5)350 (42.2)0.586 Senior397 (44.6)19 (51.4)378 (44.3) Undergraduate119 (13.4)3 (8.1)116 (13.6)Monthly income (yuan)0.322 <5000634 (71.2)29 (78.4)605 (70.8) >5000257 (28.8)8 (21.6)249 (29.2)Antibiotics regimen AMO+CLA156 (17.5)13 (35.1)143 (16.7) AMO+LEV156 (17.5)12 (32.4)144 (16.9) AMO+FUR157 (17.6)4 (10.8)153 (17.9) AMO+TET145 (16.3)0 (0.0)145 (17.0) TNZ+CLA128 (14.4)7 (18.9)121 (14.2) AMO(HDDT)149 (16.7)1 (2.7)148 (17.3)Abbreviations: MET, metronidazole; LEV, levofloxacin; CLA, clarithromycin; AMO, amoxicillin; FUR, furazoldone; TET, tetracycline; TNZ, tinidazole; HDDT, high-dose dual therapy; P, P-value. Comparison of Baseline Characteristic Between Patients with Recurrence and Non-Recurrence Abbreviations: MET, metronidazole; LEV, levofloxacin; CLA, clarithromycin; AMO, amoxicillin; FUR, furazoldone; TET, tetracycline; TNZ, tinidazole; HDDT, high-dose dual therapy; P, P-value. Analyses of Clinical Factors Influencing the Recurrence of H. Pylori Eradication To identify factors associated with H. pylori recurrence, we performed a univariate analysis comparing patients with and without recurrence. In terms of demographic and lifestyle factors, patients living in rural areas accounted for a significantly higher proportion in the recurrence group than in the non-recurrence group (70.3% vs 44.7%). Regarding smoking, the proportion of regular smokers was higher in the recurrence group than in the non-recurrence group (67.6% vs 28.6%), and the difference was statistically significant (P < 0.001). Regular sharing of tableware was more frequent among patients with recurrence than among those without (83.8% vs 65.1%, P = 0.019). Regarding patients who frequently dining-out were overrepresented in the recurrence group compared with the non-recurrence group (94.6% vs 69.0%, P < 0.001). Furthermore, the recurrence group had higher proportions of patients who perceived the hygiene conditions of dining-out establishments as “poor” (48.6% vs 18.3%) or “average” (37.8% vs 34.1%), whereas a lower proportion rated them as “good” (13.5% vs 47.7%). All these differences were statistically significant (all P < 0.001). In terms of family members, the presence of H. pylori-infected household members was more common in the recurrence group than in the non-recurrence group (73.0% vs 32.4%, P < 0.001). No statistically significant differences were observed between the two groups with respect to alcohol consumption, sharing cups, consumption of pickled foods, intake of dairy products, tea drinking, washing tableware with detergent, or family size (all P > 0.05). Table 2.Table 2Univariate Analysis of Factors Influencing the Recurrence of H. Pylori After EradicationTotal (n=891)Recurrence (n=37)Non-Recurrence (n=854))PLive region0.002 Rural408 (45.8)26 (70.3)382 (44.7) City483 (54.2)11 (29.7)472 (55.3)Smoking<0.001 Sometimes622 (69.8)12 (32.4)610 (71.4) Regular269 (30.2)25 (67.6)244 (28.6)Alcohol0.380 Sometimes395 (44.3)19 (51.4)376 (44.0) Regular496 (55.7)18 (48.6)478 (56.0)Shared tableware0.019 Sometimes304 (34.1)6 (16.2)298 (34.9) Regular587 (65.9)31 (83.8)556 (65.1)Shared water cup0.121 Sometimes372 (41.8)20 (54.1)352 (41.2) Regular519 (58.2)17 (45.9)502 (58.8)Eat pickled foods0.416 Sometimes562 (63.1)21 (56.8)541 (63.3) Regular329 (36.9)16 (43.2)313 (36.7)Consume dairy and dairy foods0.089 Sometimes385 (43.2)21 (56.8)364 (42.6) Regular506 (56.8)16 (43.2)490 (57.4)Drink tea0.282 Sometimes681 (76.4)31 (83.8)650 (76.1) Regular210 (23.6)6 (16.2)204 (23.9)Wash tableware with detergent0.308 Sometimes61 (6.8)1 (2.7)60 (7.0) Regular830 (93.2)36 (97.3)794 (93.0)Dining-out frequency<0.001 Sometimes267 (30.0)2 (5.4)265 (31.0) Regular624 (70.0)35 (94.6)589 (69.0)Dining-out hygiene<0.001 Poor174 (19.5)18 (48.6)156 (18.3) Average305 (34.2)14 (37.8)291 (34.1) Good412 (46.2)5 (13.5)407 (47.7)Family size0.554 154 (6.1)3 (8.0)51 (6.0) 2-4774 (86.9)30 (81.1)744 (87.1) >563 (7.1)4 (10.8)59 (6.9)Household H. pylori infection<0.001 No test215 (24.1)1 (2.7)214 (25.1) No infection372 (41.8)9 (24.3)363 (42.5) Infection304 (34.1)27 (73.0)277 (32.4)Abbreviation: P, P-value. Table 2Univariate Analysis of Factors Influencing the Recurrence of H. Pylori After EradicationTotal (n=891)Recurrence (n=37)Non-Recurrence (n=854))PLive region0.002 Rural408 (45.8)26 (70.3)382 (44.7) City483 (54.2)11 (29.7)472 (55.3)Smoking<0.001 Sometimes622 (69.8)12 (32.4)610 (71.4) Regular269 (30.2)25 (67.6)244 (28.6)Alcohol0.380 Sometimes395 (44.3)19 (51.4)376 (44.0) Regular496 (55.7)18 (48.6)478 (56.0)Shared tableware0.019 Sometimes304 (34.1)6 (16.2)298 (34.9) Regular587 (65.9)31 (83.8)556 (65.1)Shared water cup0.121 Sometimes372 (41.8)20 (54.1)352 (41.2) Regular519 (58.2)17 (45.9)502 (58.8)Eat pickled foods0.416 Sometimes562 (63.1)21 (56.8)541 (63.3) Regular329 (36.9)16 (43.2)313 (36.7)Consume dairy and dairy foods0.089 Sometimes385 (43.2)21 (56.8)364 (42.6) Regular506 (56.8)16 (43.2)490 (57.4)Drink tea0.282 Sometimes681 (76.4)31 (83.8)650 (76.1) Regular210 (23.6)6 (16.2)204 (23.9)Wash tableware with detergent0.308 Sometimes61 (6.8)1 (2.7)60 (7.0) Regular830 (93.2)36 (97.3)794 (93.0)Dining-out frequency<0.001 Sometimes267 (30.0)2 (5.4)265 (31.0) Regular624 (70.0)35 (94.6)589 (69.0)Dining-out hygiene<0.001 Poor174 (19.5)18 (48.6)156 (18.3) Average305 (34.2)14 (37.8)291 (34.1) Good412 (46.2)5 (13.5)407 (47.7)Family size0.554 154 (6.1)3 (8.0)51 (6.0) 2-4774 (86.9)30 (81.1)744 (87.1) >563 (7.1)4 (10.8)59 (6.9)Household H. pylori infection<0.001 No test215 (24.1)1 (2.7)214 (25.1) No infection372 (41.8)9 (24.3)363 (42.5) Infection304 (34.1)27 (73.0)277 (32.4)Abbreviation: P, P-value. Univariate Analysis of Factors Influencing the Recurrence of H. Pylori After Eradication Abbreviation: P, P-value. In order to identify the independent risk factors for H. pylori recurrence, we included the variables with statistical significance from the univariate analysis into the multivariate logistic regression model for analysis. The results demonstrated that living in rural (OR = 3.52, 95% CI: 1.55–8.00, P = 0.003), frequent smoking (OR = 6.06, 95% CI: 2.75–13.33, P < 0.001), sharing tableware (OR = 3.46, 95% CI: 1.30–9.26, P = 0.013), regular dining-out (OR = 7.41, 95% CI: 1.69–32.26, P = 0.008), poor dining-out hygiene (OR = 4.50, 95% CI: 1.51–13.43, P = 0.007), and the presence of H. pylori-infected household members (OR = 5.68, 95% CI: 2.37–13.61, P < 0.001) were independently associated with an increased risk of recurrence. In addition, patients who rated the hygiene of dining-out establishments as “average” also had a significantly higher risk of recurrence compared with those who rated it as “good” (OR = 3.34, 95% CI: 1.29–6.94, P = 0.005). Table 3.Table 3Multivariate Analysis of Factors Associated with Recurrence of H. Pylori After EradicationOR (95% CI)PLive region3.52 (1.55–8.00)0.003Smoking6.06 (2.75–13.33)<0.001Shared tableware3.46 (1.30–9.26)0.013Eating out frequency7.41 (1.69–32.26)0.008Eating-out hygiene Average vs good3.34 (1.29–6.94)0.005 Poor vs good4.50 (1.51–13.43)0.007Household H. pylori infection Infection v. no infection5.68 (2.37–13.61)<0.001Abbreviations: OR, odds ratio; CI, confidence interval; P, P-value. Table 3Multivariate Analysis of Factors Associated with Recurrence of H. Pylori After EradicationOR (95% CI)PLive region3.52 (1.55–8.00)0.003Smoking6.06 (2.75–13.33)<0.001Shared tableware3.46 (1.30–9.26)0.013Eating out frequency7.41 (1.69–32.26)0.008Eating-out hygiene Average vs good3.34 (1.29–6.94)0.005 Poor vs good4.50 (1.51–13.43)0.007Household H. pylori infection Infection v. no infection5.68 (2.37–13.61)<0.001Abbreviations: OR, odds ratio; CI, confidence interval; P, P-value. Multivariate Analysis of Factors Associated with Recurrence of H. Pylori After Eradication Abbreviations: OR, odds ratio; CI, confidence interval; P, P-value. Antibiotics Resistance Profiles of H. Pylori in Patients with Recurrence To further characterize the antimicrobial resistance profiles of H. pylori in patients with recurrence after eradication therapy, we performed in vitro susceptibility testing against six antibiotics on isolates obtained from 37 patients, of whom 27 yielded valid results. The highest resistance rate was observed for MET (96.3%, 95% CI: 79.1–99.8), followed by CLA (77.8%, 95% CI: 57.3–91.4) and LEV (62.9%, 95% CI: 42.4–80.6), while the resistance rate to amoxicillin was low (3.7%, 95% CI: 0.1–19.0). Notably, all isolates remained susceptible to FUR and TET. Regarding resistance patterns, single-drug resistance to MET accounted for 11.1% (95% CI: 2.4–29.2) of cases. Among double-resistance patterns, the combination of MET+CLA accounted for 25.9% (95% CI: 11.1–46.3), followed by MET+LEV (11.1%, 95% CI: 2.4–29.2). In terms of multidrug resistance (resistance to three or more antibiotics), the predominant pattern was MET+CLA+LEV, which accounted for 48.1% (95% CI: 28.7–68.1) of cases, while the combination of AMO+CLA+LEV accounted for 3.7% (95% CI: 0.1–19.0). Table 4.Table 4Antibiotics Resistance Profile of H. Pylori Isolates From Recurrent Patients (n=27)Total (n=27)Resistance (95% CI)MET2696.3 (79.1–99.8)LEV1762.9 (42.4–80.6)CLA2177.8 (57.3–91.4)AMO13.7 (0.1–19.0)FUR00.0 (0–13.0)TET00.0 (0–13.0)Single resistance MET311.1 (2.4–29.2)Double resistance MET+CLA725.9 (11.1–46.3) MET+LEV311.1 (2.4–29.2)Multiple resistance MET+CLA+LEV1348.1 (28.7–68.1) AMO+CLA+LEV13.7 (0.1–19.0)Abbreviations: MET, metronidazole; LEV, levofloxacin; CLA, clarithromycin; AMO, amoxicillin; FUR, furazoldone; TET, tetracycline; CI, confidence interval. Table 4Antibiotics Resistance Profile of H. Pylori Isolates From Recurrent Patients (n=27)Total (n=27)Resistance (95% CI)MET2696.3 (79.1–99.8)LEV1762.9 (42.4–80.6)CLA2177.8 (57.3–91.4)AMO13.7 (0.1–19.0)FUR00.0 (0–13.0)TET00.0 (0–13.0)Single resistance MET311.1 (2.4–29.2)Double resistance MET+CLA725.9 (11.1–46.3) MET+LEV311.1 (2.4–29.2)Multiple resistance MET+CLA+LEV1348.1 (28.7–68.1) AMO+CLA+LEV13.7 (0.1–19.0)Abbreviations: MET, metronidazole; LEV, levofloxacin; CLA, clarithromycin; AMO, amoxicillin; FUR, furazoldone; TET, tetracycline; CI, confidence interval. Antibiotics Resistance Profile of H. Pylori Isolates From Recurrent Patients (n=27) Abbreviations: MET, metronidazole; LEV, levofloxacin; CLA, clarithromycin; AMO, amoxicillin; FUR, furazoldone; TET, tetracycline; CI, confidence interval. Discussion H. pylori is intricately associated with the pathogenesis and progression of chronic gastritis, peptic ulcer disease, MALT lymphoma, and gastric adenocarcinoma.1 Successful eradication of H. pylori has been shown to reverse the progression of gastric mucosal atrophy and substantially reduce the long-term risk of gastric cancer.3 Despite the availability of highly effective regimens, including bismuth-based quadruple therapy and high-dose dual therapy, which can achieve eradication rates exceeding 90%,4 a proportion of patients still experience recurrence after successful treatment. Recurrence after successful eradication remains a major clinical problem worldwide, particularly in regions characterized by high baseline infection rates, a heavy burden of chronic gastric diseases, and limited socioeconomic and healthcare resources.6 Notably, the epidemiological features of H. pylori recurrence exhibit marked geographical variation. A systematic review and meta-analysis by Hu et al demonstrated that the annual recurrence rate of H. pylori was inversely correlated with the socioeconomic development level of the region.12 In countries and areas with a “very high” Human Development Index, the annual recurrence rate was 3.1% (95% CI: 2%–4%), whereas it increased to 6.2% (95% CI: 4%–8%) in regions with a “high” HDI, and rose further to 10.9% (95% CI: 6%–18%) in those with a “medium” or “low” HDI.12 These findings suggest that H. pylori recurrence is closely intertwined with socioeconomic and sanitary conditions. In a retrospective analysis of 891 patients from a single center in our region, the one-year recurrence rate after successful eradication was 4.1%, which, while slightly higher than the 3.1% rate reported in developed countries, remains considerably lower than the average of 10.9% observed in developing countries.12 This outcome may be attributable to the distinctive regional profile of the Ningxia area. On one hand, both the prevalence and antimicrobial resistance rates of H. pylori in this region—particularly for MET, CLA, and LEV—are notably elevated, and multidrug resistance patterns are highly prevalent. On the other hand, socioeconomic and behavioral determinants, including relatively underdeveloped economic conditions, limited healthcare resources, and the traditional practice of shared meals, may collectively contribute to an unfavorable risk landscape for recurrence.14,15 Furthermore, we performed antimicrobial susceptibility testing on isolates from patients with recurrence and delineated their resistance profiles, which may offer valuable guidance for the selection of subsequent eradication regimens in this population. Our findings showed that rural residence was significantly associated with H. pylori recurrence (OR = 3.52). This finding is consistent with a prospective case–control study, which reported that rural residence was significantly associated with an elevated risk of recurrence, whereas city residence served as a protective factor against reinfection (OR = 0.155).19 In addition, a multicenter prospective study from China showed that geographic location was an independent risk factor for reinfection, with hazard ratios of 5.5 for the western region and 4.9 for the central region compared with the northern region.20 However, it is worth noting that Gómez Rodríguez et al, in a 4-year follow-up of 208 patients with successful eradication, found that rural or urban residence did not emerge as an independent predictor of reinfection in multivariate analysis.21 This discrepancy may be attributed to differences in basic sanitary conditions between urban and rural areas in developed versus developing countries. In regions with relatively balanced sanitary conditions, the predictive value of residence itself may be attenuated. The higher recurrence rate observed in rural patients in our region may result from a combination of factors, including limited healthcare resources, prevalent multi-generational household living, shared tableware practices, and low health awareness. Therefore, focusing solely on individualized eradication regimens while neglecting improvements in living and sanitary conditions may make it difficult to substantially reduce the recurrence rate. Our study also found that regular smoking was significantly associated with H. pylori recurrence (OR = 6.06). Zhang et al reported that smoking was an independent risk factor for H. pylori recurrence, with univariate analysis showing a significant association between smoking and recurrence; the smoking rate was 46.2% in the recurrence group versus 22.9% in the non-recurrence group (P = 0.006).6 This is consistent with our findings. However, a study from South Korea and a multicenter prospective cohort study both reported no significant association between smoking and H. pylori recurrence, and smoking was not identified as an independent risk factor,20,22 which contrasts with our results. This discrepancy may be related to several factors. First, our study was a retrospective cohort study, and the presence of follow-up or loss‑to‑follow-up bias may have influenced the comparability of the results to some extent. Second, differences in sample size, population characteristics, geographical settings, and control of confounding factors across studies may also contribute to the observed heterogeneity. Of note, smoking may affect H. pylori eradication efficacy and recurrence risk through multiple mechanisms. A meta-analysis showed that smoking can reduce gastric mucosal blood flow and affect the local concentration of antibiotics in the stomach, thereby compromising treatment efficacy.23 In addition, smoking has been confirmed to be closely associated with eradication failure, with smokers having a significantly higher risk of failed eradication than non-smokers.23,24 It should be noted that although our study focused on recurrence after successful eradication rather than initial eradication failure, the unfavorable impact of smoking on eradication efficacy may indirectly increase the risk of recurrence. Our findings showed that regular shared tableware and the presence of H. pylori-infected household members were both significantly associated with recurrence, and both factors were identified as independent risk factors associated with H. pylori recurrence. It is well established that H. pylori is primarily transmitted from person to person via the oral–oral route25 and sharing tableware has been recognized as an important route of intrafamilial transmission.26 Regarding the mechanisms of household transmission, studies have shown that patients with persistent H. pylori in their oral dental plaque after successful eradication had a significantly higher risk of recurrence (OR = 3.8), those who received periodontal therapy had their annual recurrence rate reduced from 15.14% to 2.04% (OR = 0.67, P < 0.001) and combination with periodontal treatment was associated with a 31% reduction in recurrence risk (P < 0.001). In addition, patients with poor oral hygiene had a significantly higher recurrence risk than those with good oral hygiene.25 These findings suggest that sharing tableware is essentially a behavioral manifestation of oral-salivary transmission, and that dental plaque may serve as a potential reservoir for H. pylori,21 and through shared tableware within households, family members remain continuously exposed to infection, which may constitute an important source of reinfection after successful eradication. Further analysis indicated that the presence of an H. pylori-infected household member was also an independent risk factor for recurrence (OR = 5.68), with an effect size slightly larger than that of sharing tableware (OR = 3.46). This result is consistent with previous study.6 H. pylori infection exhibits obvious familial clustering,26 and both sharing tableware and having infected household members appear to point to intrafamilial transmission as an important pathway for H. pylori recurrence. This is supported by studies using RAPD fingerprinting analysis, which have confirmed a high degree of homology among H. pylori strains infecting members of the same household,27 providing molecular evidence for why the presence of an infected household member significantly increases the recurrence risk in other members. However, a prospective serological study from Finland involving 329 families found no correlation in H. pylori seropositivity between spouses or between parents and children, and found no evidence supporting intrafamilial transmission.28 This finding appears to contradict our results, but we believe that this discrepancy reflects significant regional heterogeneity in H. pylori transmission patterns. Finland, as a high-HDI country with an H. pylori infection rate of only 5%,28 has well-established practices of meal sharing, utensil disinfection, personal hygiene awareness, and basic sanitation infrastructure, which may render the risk of transmission through shared tableware and intrafamilial contact negligible. In contrast, in our region, where shared meal culture is longstanding and the H. pylori infection rate remains relatively high, the risk of intrafamilial transmission is further increased. Given the coexistence of these two independent factors, the 2021 Chinese consensus on household-based H. pylori infection control and management explicitly recommends that synchronized eradication therapy should be considered for H. pylori-infected family members.26 We believe that, in addition to individualized eradication regimens, behavioral interventions such as implementing separate meal service and dedicated tableware may help reduce the recurrence rate of H. pylori in our region. However, due to limitations in our study design, we did not test for H. pylori in dental plaque, and thus we could not validate at the microbiological level the role of dental plaque as a reservoir in our recurrent patients. This mechanism warrants further investigation in future studies. Our findings showed that frequent dining out and poor hygiene at eating establishments were significantly associated with H. pylori recurrence. Previous studies have shown that dining out increases the risk of cross-infection with H. pylori, and poor restaurant hygiene has been identified as an independent risk factor for recurrence.29,30 Other study has also demonstrated that patients who frequently dine out and consume food in unsanitary conditions have a significantly higher recurrence rate.19 These findings are consistent with our observations.19,29,30 Notably, a follow-up study by Zhou et al further revealed a dose–response relationship between dining-out frequency and recurrence risk: the recurrence rates among patients who rarely, occasionally, or frequently dined out were 0.63%, 2.92%, and 13.46%, respectively (P < 0.001), indicating a rising trend in recurrence risk with increasing frequency of dining out.31 From the perspective of transmission mechanisms, H. pylori is primarily transmitted through the oral–oral route.25 Studies have shown that H. pylori can be detected in food samples such as milk, vegetables, and meat, and that contaminated food may serve as a vehicle for infection.32 Frequent dining out therefore increases an individual’s exposure to diverse food environments and shared dining settings, thereby raising the likelihood of contact with H. pylori-infected individuals. Concurrently, poor hygiene at dining venues may facilitate cross-contamination. Based on our findings and previous evidence, we recommend that patients reduce the frequency of dining out after successful H. pylori eradication, choose eating establishments with good hygiene standards when dining out, and adopt healthy practices such as sharing meal servings. These simple behavioral modifications may be of considerable importance in reducing the risk of H. pylori recurrence. In this study, the resistance profile of H. pylori in patients with recurrence was complex. Isolates from these patients showed high resistance rates to MET, CLA, and LEV, with multidrug resistance being the predominant feature, whereas resistance to AMO, FUR, and TET remained extremely low. This finding is consistent with previously reported patterns of secondary resistance and resistance profiles in refractory H. pylori infections in the Ningxia region.33,34 Local data have shown that the primary and secondary resistance rates to three antibiotics were as follows: MET (94.3% vs 97.5%), CLA (39.8% vs 74.6%), and LEV (40.5% vs 68.2%).33 In our study, the resistance profile of recurrent isolates was closer to the secondary resistance data than to the primary resistance data. In addition, a study of patients with refractory H. pylori infection reported resistance rates of 96.55% for MET, 77.59% for CLA, and 56.90% for LEV, with the most common resistance pattern being the triple resistance to MET+CLA+LEV (48.28%),34 which is highly similar to the resistance profile observed in our study. Meanwhile, analysis of prior antibiotic regimens (Table 1) revealed that a high proportion of patients in the recurrence group had received CLA-containing (AMO+CLA) or LEV-containing (AMO+LEV) regimens, whereas those who received TET, FUR, or high-dose AMO-containing regimens had substantially lower recurrence rates. This distribution pattern was consistent with the resistance profile observed in our recurrent patients. These findings suggest that the resistance characteristics of recurrent isolates may be more closely related to secondary resistance patterns, possibly reflecting either incomplete eradication (ie, recrudescence) in patients who were reported as successfully treated, or acquisition of highly resistant strains from the environment. However, due to the limitations of our study design, we were unable to determine whether these recurrent isolates originated from recrudescence of the original strains or from reinfection with new strains. Nevertheless, our findings provide valuable information for selecting subsequent treatment regimens for patients with recurrence. In our study, the multidrug resistance rate among recurrent isolates was as high as 51.8% (14/27), with the MET+CLA+LEV pattern being the most common (48.1%). This high rate of multidrug resistance has important clinical implications. First, for patients with previous treatment failure or recurrence, multidrug resistance greatly limits the availability of effective antibiotic combinations, and continued empirical use of regimens containing these drugs carries a high risk of failure in subsequent eradication attempts. Second, the emergence of multidrug‑resistant strains is often associated with prior multiple antibiotic exposures,33 suggesting that in patients with recurrence, simply repeating previously used regimens is not advisable, and prior antibiotic history should be carefully considered. The preferred strategy is to select individualized treatment regimens guided by antimicrobial susceptibility testing to avoid the blind reuse of drugs with high resistance rates, such as CLA and LEV. When susceptibility testing is not readily available, empirical regimens containing AMO, TET, or FUR should be prioritized. Our study has several limitations. First, this was a retrospective cohort study, and we did not implement a regular and rigorous follow-up schedule for patients after successful eradication. Some recurrent cases may have been missed because patients did not actively report them, which may have affected the accuracy of the recurrence rate estimate. Second, some patients who achieved initial eradication success did not return for follow-up testing at the scheduled time, either because they lacked typical dyspeptic symptoms or had limited awareness of the clinical significance of H. pylori recurrence. This may have introduced loss-to-follow-up bias. Although we took measures to minimize bias, including consecutive patient enrollment and questionnaire administration by trained investigators, selection bias and information bias may still be present. In addition, some information relied on patient recall, which may have introduced recall bias. Because patients lost to follow-up did not provide complete follow-up data, we were unable to directly compare them with those who completed the study, and thus could not quantitatively assess the impact of attrition on the recurrence‑rate estimate. Third, we were unable to include several potential confounders in our analysis, such as treatment adherence, proton pump inhibitor or P-CAB use, adverse events, oral hygiene or periodontal disease status, household cotherapy, socioeconomic factors beyond monthly household income, and comorbidities. These factors may influence the risk of H. pylori recurrence to some extent. However, due to the retrospective design and the limited number of recurrent cases, the final model included only six variables, and the above-mentioned factors were not incorporated, which may affect the robustness of our findings. Fourth, antimicrobial susceptibility testing was performed using the disk diffusion method rather than the standard agar dilution method, and the reliability of the results requires further validation. Fifth, we did not collect baseline susceptibility data for the initial H. pylori isolates before eradication therapy. As a result, we could not directly compare the resistance profiles of recurrent isolates with those of the original infecting strains, nor could we determine whether the resistance in recurrent isolates originated from recrudescence of a primary resistant strain or from newly acquired resistance in a reinfecting strain. Although the resistance profile of recurrent patients was similar to the secondary resistance pattern observed in our region, the lack of baseline data limited our ability to further interpret the resistance-related recurrence patterns. Sixth, the sample size of H. pylori isolates from recurrent patients was small, and 10 of the 37 recurrent patients failed to yield cultivable isolates, leaving their resistance profiles unknown. Thus, the resistance data we obtained may have limited representativeness. In addition, although we adopted a parsimonious modeling strategy with six independent variables (EPV=6.2) and the overall model was statistically significant (P < 0.001), the Hosmer–Lemeshow test indicated limited model fit (P < 0.05), which may be attributable to data sparsity due to the small number of events. Larger sample sizes are needed in future studies to validate our findings. Seventh, we did not perform molecular typing, such as genome sequencing, on H. pylori isolates from patients with recurrence. Therefore, we could not determine at the molecular level whether recurrent infections represented recrudescence of the original strain or reinfection with a new strain. Although we adopted the 1-year time threshold as a classification criterion for recurrence types based on previous literature, this empirical criterion has not been validated by molecular methods in our study, and classification based on this threshold may be subject to misclassification bias. It should be clearly stated that all discussions of “recrudescence” and “reinfection” in this study are based on time-based classification rather than molecularly confirmed conclusions. Future prospective multicenter studies, incorporating molecular typing, are needed to further validate our findings. Conclusion In summary, this study revealed a one-year H. pylori recurrence rate of 4.1% in the Ningxia region, which is comparable to rates reported in other developing countries. Factors significantly associated with an increased risk of recurrence include rural residence, smoking, sharing tableware, regular dining out, poor hygiene conditions at dining establishments, and the presence of an H. pylori-infected household member. With regard to antimicrobial resistance, isolates from patients with recurrence exhibited high resistance rates to MET, CLA, and LEV, with multidrug resistance being the predominant pattern. In contrast, resistance rates to AMO, FUR, and TET remained relatively low. These findings suggest that effective prevention of H. pylori recurrence may require integrated interventions that target both high-risk behaviors and local resistance patterns.