Introduction

Morganella morganii is an emerging Gram-negative bacterium in the Enterobacteriaceae family. It is a crucial opportunistic pathogen that causes many clinical and community-acquired infections, including wound, bloodstream, and urinary tract infections. M. morganii infections can be life threatening, particularly in immunocompromised individuals, such as people with diabetes, renal disorders, or chronic liver disease1,2. The global distribution of M. morganii infections reveals a concerning trend. In Israel, a study documented 136 adult cases of M. morganii bacteraemia admitted to a large university hospital3. In Canada, the bacterium was associated with an overall incidence rate of 77 cases per million population4. Similarly, in Australia, 709 cases of M. morganii bloodstream infections (BSIs) were identified between 2000 and 2019, corresponding to an annual incidence of 9.2 cases per million population. M. morganii infections have been reported, such as Taiwan, Japan, the United States, and Spain have higher numbers of reported cases5.

M. morganii exhibits intrinsic resistance to multiple antibiotic classes including ampicillin, first- and second-generation cephalosporins, macrolides, and clindamycin due to its natural AmpC β-lactamase production and other resistance mechanisms6. Although, most strains are naturally susceptible to third-generation and fourth-generation cephalosporins, carbapenems, aztreonam, fluoroquinolones, aminoglycosides, and chloramphenicol. The clinical significance of M. morganii is increasingly recognized, largely due to its emerging acquired resistance to additional antibiotic classes, including extended-spectrum β-lactams, fluoroquinolones, and aminoglycosides.1,7,8. Fluoroquinolones represent one of the few remaining oral antibiotic options with reliable activity against M. morganii. The resistance to fluoroquinolones is an emerging global concern, as it complicates treatment strategies and leads to higher mortality rates, particularly in vulnerable populations, such as elderly patients with comorbidities such as cancer, dementia, and heart failure5. The main mechanisms underlying fluoroquinolone resistance in M. morganii are chromosomal mutations and plasmid-mediated resistance mechanisms that cause target enzyme mutations and efflux pump overexpression9,10. Plasmid-mediated quinolone resistance (PMQR) has emerged as a significant mechanism of antimicrobial resistance involves three main mechanisms: The first mechanism is target protection by Qnr proteins that protect target enzymes (DNA gyrase and type IV topoisomerase) from quinolone inhibition. The discovery of ciprofloxacin resistance led to the identification of a second PMQR mechanism: The modification of certain quinolones by a specific aminoglycoside acetyltransferase, AAC(6′)-Ib-cr. A third mechanism was later identified with the discovery of plasmid-mediated quinolone efflux pumps, namely QepA and OqxAB. Over the past decade, PMQR genes have been detected in bacterial isolates worldwide, contributing to the global spread of quinolone resistance11.

Sheng et al.12 reported that M. morganii isolates were susceptible to fluoroquinolones until 1996, and that two years later, 20% of M. morganii isolates were resistant to fluoroquinolones in Taiwan. More recent studies have shown that the prevalence of fluoroquinolone resistance in M. morganii varies significantly across healthcare settings and geographic regions. For instance, a study conducted in China from 2014 to 2020 reported that the rate of ciprofloxacin resistance ranged from 14 to 19% across three hospitals13, while data from the USA indicated that the rate of ciprofloxacin resistance was approximately 31.7% in 201914.

Given that antibiotic resistance in M. morganii is increasing and has negative impacts on clinical outcomes, we conducted a systematic review and meta-analysis to compile and analyze data on the prevalence of fluoroquinolone resistance in M. morganii. We considered various factors that may influence the severity of antibiotic resistance (e.g., geographic location and healthcare setting). The findings provide critical information on the prevalence of fluoroquinolone resistance in M. morganii, aiding the effective selection and development of antibiotics for treating M. morganii infections and potentially reducing antibiotic resistance rates in the future.

Methods

Search strategy and selection criteria

The protocol of this systematic review is registered with PROSPERO (CRD42024562013). This study followed the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines15. A comprehensive search was conducted across the PubMed, Embase, ScienceDirect, and Scopus databases using various keyword combinations related to M. morganii and antimicrobial resistance. The search strategy included the following terms: (Morganella morganii OR M. morganii) AND (antibiotic resistance OR disk diffusion OR microdilution OR drug sensitivity OR susceptibility OR multidrug resistant) AND (fluoroquinolones OR ciprofloxacin OR ofloxacin OR levofloxacin OR norfloxacin OR nalidixic acid) (Supplementary Data 1). Articles published up to July 10, 2024 were included, without any restriction on the earliest publication date. Citation management and duplicate removal were performed using EndNote X9.0 software. Two independent reviewers conducted the title and abstract screening using the predefined inclusion and exclusion criteria. This was followed by a full-text assessment by two separate investigators. Discrepancies in the study selection process were resolved by a third investigator.

Inclusion and exclusion criteria

Studies were included if they reported the prevalence of M. morganii isolates among patients and the antibiotic resistance rates of the isolates, or if they provided only antibiotic resistance data. Studies that focused on M. morganii clinical isolates were considered, while those that investigated environmental isolates were excluded. Conference proceedings were omitted due to insufficient data for quality assessment, and dissertations and theses were not considered. Duplicate articles were excluded, as were reviews, meta-analyses, systematic reviews, case reports, brief reports, notes, editorials, correspondence, short communications, and letters to the editor. Studies published in languages other than English or for which the full text was unavailable were also excluded. Studies that examined bacterial species other than M. morganii or that included < 10 isolates were not assessed16. Studies that reported antibiotic resistance as MIC90 or that evaluated the synergistic effects of antibiotics were excluded. Furthermore, studies that categorized M. morganii as part of a broader Gram-negative bacterial group and reported only overall antibiotic resistance rates in Gram-negative bacteria were not considered. Finally, studies that tested only resistant isolates or reported only the prevalence of M. morganii infection were removed. The included articles comprised prevalence studies, retrospective studies, cross-sectional studies, and studies reporting the resistance rates of M. morganii to fluoroquinolones.

Study selection and data extraction

Two independent researchers thoroughly reviewed the full-text articles and extracted key data, including the first author’s name, study period, year of publication, geographic location (country and region), sample size, resistant isolate, and specimen type. In addition, the antibiotic susceptibility testing (AST) method was recorded (i.e., agar dilution, broth microdilution, E-test, disk agar diffusion, or the use of minimum inhibitory concentration (MIC) test strips, a VITEK system, a Phoenix system, or a MicroScan system). Additionally, data on the antibiotic resistance rates of the studied M. morganii isolates were collected. Any discrepancies between the two reviewers during the data extraction process were resolved through consensus.

Quality assessment

Two independent reviewers assessed the quality of the included studies using the Joanna Briggs Institute (JBI) critical appraisal checklist for studies reporting prevalence data17. Each criterion was evaluated on a scale of 0–1, with a maximum possible score of 10. Studies with a score ≥ 5 were categorized as high quality.

Meta-analysis

The meta-analysis was conducted using the STATA 18 software package (Stata Corporation, College Station, TX, USA). A forest plot was generated to proportion of fluoroquinolone-resistant M. morganii; the estimated prevalence and 95% confidence interval (CI) are shown. The Freeman–Tukey double arcsine transformation helps stabilize variance and prevent negative proportions, especially in cases involving extremely small or large proportions18,19. This transformation was used to calculate the weighted pooled estimate, which was then reverse-transformed. The heterogeneity among the studies was assessed using I2 statistics, with 0–25% indicating low heterogeneity, 25–50% indicating moderate heterogeneity, 50–75% indicating substantial heterogeneity, and 75–100% indicating considerable heterogeneity. A random-effects model was used when the I2 statistic indicated substantial heterogeneity (> 50%) and was statistically significant (p < 0.10) based on the χ2 test (Cochran’s Q test). To assess the risk of publication bias, Egger’s test was performed, and Begg’s funnel plots were generated. A p-value < 0.05 indicated the presence of statistically significant publication bias. A meta-regression analysis was conducted using a significance level of p < 0.05 to assess whether covariates (including publication year, sample size, geographic location, AST method, antibiotic, and quality score) could account for the observed between-study heterogeneity.

Subgroup meta-analysis

A subgroup comparative analysis was conducted to examine the prevalence of M. morganii based on its resistance to specific antibiotics, including nalidixic acid, ciprofloxacin, norfloxacin, ofloxacin, fleroxacin, pefloxacin, lomefloxacin, levofloxacin, garenoxacin, gemifloxacin, and trovafloxacin. Additional subgroup comparative analyses were performed based on the geographic region, publication year (five-year intervals), testing method, sample size, study design, and guideline for AST interpretation, quality score of the studies. We also analyzed the antibiotic resistance of the M. morganii isolates found specifically in Asia. To evaluate the risk of publication bias, Egger’s weighted regression method was applied, and funnel plots were generated. A p-value < 0.05 was considered indicative of statistically significant publication bias20.

Results

Characteristics of the included studies

A total of 2730 studies were initially identified by systematically searching four electronic databases. After 523 duplicates were removed, the remaining 2207 articles were screened based on their titles and abstracts. This led to the exclusion of 1978 articles. Among the remaining 229 articles, 26 could not be retrieved, leaving 203 for full-text assessment. Most (n = 149) of these studies were excluded because they did not meet the eligibility criteria. Hence, 54 studies comprising 45,440 clinical isolates were included in the systematic review and meta-analysis. (Fig. 1).

Fig. 1
figure 1

Study flow diagram. The protocol for selecting potentially relevant studies is shown.

Full size image

The characteristics of the studies were systematically documented to capture key information, including the study period, geographic location, and institutional sources of samples. The specimen sources and diagnoses were also recorded to provide context for the isolates analyzed, and the AST methods used and range of antibiotics tested were noted. Additionally, the quality assessment scores assigned to each study were calculated. A summary of the publication year, geographic region, AST method, sample size, and study design data is presented in Table 1, and a comprehensive overview of all variables is presented in Table 2 and Supplementary Table 2.

Table 1 Overall characteristics of the studies.
Full size table
Table 2 Characteristics of the studies.
Full size table

As shown in Table 1, the included studies were published between 1993 and 2024. The majority were reported during 2018–2022, followed by 2008–2012 and 1998–2002, respectively. Most studies were conducted in Asia, followed by Europe and North America, with only one study conducted in Australia. Regarding the antimicrobial susceptibility testing (AST) methods, broth dilution was the most commonly used, followed by disk diffusion, while a few studies employed alternative methods. Most studies included fewer than 100 isolates. In terms of study design, the majority were retrospective or cross-sectional. A few studies employed other designs, such as cohort, prospective cohort, or surveillance studies. With respect to study quality, four studies were classified as low quality (score: 1–4), while 50 studies were rated as moderate to high quality (score: 5–9) (see Supplementary Table 1).

Overall prevalence of fluoroquinolone resistance among M. morganii clinical isolates

The pooled prevalence of fluoroquinolone-resistant M. morganii was estimated to be 21% (95% CI [16, 27]; p < 0.01) (Fig. 2 and Table 3). The Egger’s test results revealed that there was no publication bias among the 54 studies. As depicted in Supplementary Fig. 1, the Begg’s funnel plot analysis showed that the data were distributed in a qualitatively symmetrical manner. Additionally, there was high heterogeneity (I2 = 99.48%; p < 0.01) among the studies (Table 3). To determine the factors that may have influenced the reported prevalence of fluoroquinolone resistance in the clinical isolates of M. morganii and the heterogeneity, we performed a subgroup analysis based on the antibiotics tested, geographic regions, study periods, AST methods, sample sizes, and quality assessment scores.

Fig. 2
figure 2

Forest plot showing the prevalence of resistance among M. morganii clinical isolates identified in the included studies. Blue squares: Effect estimates, green diamond: Pooled effect estimate, red vertical line: Pooled effect estimate, and gray vertical line: No effect line. CI Confidence interval.

Full size image
Table 3 Subgroup analysis of the prevalence of fluoroquinolone resistance in M. morganii clinical isolates.
Full size table

Resistance of M. morganii clinical isolates to different generations of quinolones

Among the included studies, ciprofloxacin was the most frequently reported antibiotic, appearing in 49 articles, followed by levofloxacin in 23 articles, ofloxacin in 5 articles, norfloxacin in 4 articles, nalidixic acid in 3 articles, and trovafloxacin in 2 articles. The remaining quinolones—fleroxacin, garenoxacin, gemifloxacin, lomefloxacin, and pefloxacin—were each reported in only 1 article.

In the first subgroup analysis, we examined the resistance of the M. morganii clinical isolates to different generations of quinolones. Among the first-generation quinolones, resistance to nalidixic acid was tested in 197 isolates, and a 27% resistance rate was found (95% CI [13, 43]; p < 0.01), with moderate heterogeneity (I2 = 51.92%). Resistance to norfloxacin was tested in 2,036 isolates, and a resistance rate of 11% was observed (95% CI [0, 95]; p < 0.01), with heterogeneity (I2 = 99.70%). For the second-generation fluoroquinolones, the resistance rate for ciprofloxacin (n = 38,520) was 22% (95% CI [16, 28]; p < 0.01), with heterogeneity (I2 = 99.40%). As depicted in Supplementary Fig. 2, qualitatively symmetrical funnel plots were observed. The rate of ofloxacin resistance (n = 866) was 9% (95% CI [1, 22]), with heterogeneity (I2 = 94.89%). The resistance rates for lomefloxacin and fleroxacin (each n = 390) were 3% (95% CI [2, 5]) and 2% (95% CI [1, 4]), respectively. The highest resistance rate (88%) was associated with pefloxacin (n = 17; 95% CI [68, 100]). Among the third-generation fluoroquinolones tested, the rate of resistance to levofloxacin (n = 9,166) was 25% (95% CI [13, 38]; p < 0.01), with heterogeneity (I2 = 99.31%). Finally, the rates of resistance to the fourth-generation fluoroquinolones were as follows: 11% for gemifloxacin (n = 94; 95% CI [5, 18]), 8% for trovafloxacin (n = 104; 95% CI [2, 18]), and 0% for garenoxacin (n = 10; 95% CI [0, 17]), with low heterogeneity (I2 = 23.27%) (Fig. 2).

Prevalence of fluoroquinolone-resistant M. morganii by geographic region

Fluoroquinolone resistance varied notably across the geographic regions included in the studies (Fig. 3 and Supplementary Figs. 3 and 4). In Asia, 2464 isolates were analyzed, the overall resistance rate was 35% (95% CI [22, 45]; p < 0.01), with high heterogeneity (I2 = 99.00%). When we performed a subregional analysis of the data obtained in Asia, the results showed that the resistance rate was 22% in East Asia (95% CI [13, 34]; I2 = 93.28%; p = 0.66), 1% in Southeast Asia (95% CI [0, 10]; I2 = 0.00%; p = 1.00), and 62% in West Asia (95% CI [45, 65]; I2 = 96.66%; p < 0.01). In Africa, 89 isolates were analyzed, the resistance rate was 55% (95% CI [45, 65]), with no heterogeneity (I2 = 0.00%). The resistance rate was 36% in South America, 151 isolates were found, (95% CI [17, 56]), 11% in North America (30,233 isolates) (95% CI [7, 16]; p = 0.63), 9% in Europe (10,209 isolates) (95% CI [7, 13]; p = 0.42), and 4% in Australia (680 isolates) (95% CI [3, 6]). Across the studies with global sampling (1,614 isolates), the resistance rate was 43% (95% CI [27, 59]; p = 0.36).

Fig. 3
figure 3

Global prevalence of fluoroquinolone-resistant M. morganii based on data extracted from the included studies.

Full size image

The studies conducted in Asia provided a considerable volume of data, and this allowed us to analyze the resistance to specific fluoroquinolones in the isolates in this dataset (Supplementary Fig. 5). The overall resistance rate was 35% (95% CI [22, 48]; p < 0.01), with high heterogeneity (I2 = 99.00%). The isolates showed similar resistance to ciprofloxacin and levofloxacin, with resistance rates of 34% (95% CI [12, 61]; I2 = 98.80%) and 33% (95% CI [8, 62]; I2 = 96.99%), respectively. The resistance rate for norfloxacin was 19% (95% CI [0, 100]; p = 0.57), with heterogeneity (I2 = 99.75%), which reflected variable data. The isolates showed less resistance to ofloxacin, with a resistance rate of 8% (95% CI [5, 12]; p < 0.01), while they showed the highest resistance to pefloxacin at 88% (95% CI [68, 100]; p < 0.01), which was likely influenced by selective use or limited data.

Prevalence of fluoroquinolone-resistant M. morganii by study period, AST method, sample size, quality score, study design, and guideline for AST interpretation

Temporal trends in fluoroquinolone resistance among M. morganii clinical isolates were determined by categorizing the studies by publication year (five-year intervals) and analyzing the resistance recorded in each period. For the five studies published between 1993 and 1997, the resistance rate was 2% (95% CI [0, 10]; p = 0.15; I2 = 99.39%). For the seven studies published between 1998 and 2002, the rate increased to 11% (95% CI [8, 15]; p < 0.01; I2 = 84.89%), and a similar rate was observed for the four studies published between 2003 and 2007 (I2 = 96.90%). The resistance rate increased to 21% (95% CI [10, 33]; p < 0.01; I2 = 96.49%) for the studies published between 2008 and 2012, and further rose to 30% (95% CI [16, 47]; p < 0.01; I2 = 87.03%) for those published between 2013 and 2017. The peak resistance rate of 37% (95% CI [22, 52]; p < 0.01; I2 = 99.34%) was associated with the studies published between 2018 and 2022, with evidence of publication bias (p = 0.03). For the four studies published in 2023–2024, the resistance rate was 33% (95% CI [22, 45]; p < 0.01; I2 = 90.26%) (Supplementary Fig. 6).

The level of fluoroquinolone resistance also varied according to the AST method used. When the disk diffusion method was used, the rate of resistance was 15% (95% CI [9, 23]; I2 = 84.98%), whereas the microdilution method was associated with a resistance rate of 24% (95% CI [16, 34]; I2 = 99.67%). When both methods were used, the resistance rate was 11% (95% CI [7, 16]; I2 = 97.54%). The use of alternative methods was associated with a resistance rate of 18% (95% CI [7, 33]; I2 = 97.66%) or 22% (95% CI [0, 61]; I2 = 93.21%), depending on the method, and no publication bias was observed (Supplementary Fig. 7).

Studies that included ≤ 100 isolates reported a 21% resistance rate (95% CI [13, 30]; I2 = 99.82%), and those with > 100 isolates reported a 22% resistance rate (95% CI [16, 28]; I2 = 88.93%) (Supplementary Fig. 8). Resistance was found to vary with the study quality: Low-quality studies (score of 1–4) reported a resistance rate of 10% (95% CI [0, 28]; I2 = 92.64%), and higher-quality studies (score of 5–9) reported a resistance rate of 22% (95% CI [17, 28]; I2 = 99.42%) (Supplementary Fig. 9). Based on the guidelines used in the included studies, resistance rates varied accordingly. Studies using CLSI reported a resistance rate of 18% (95% CI [15, 21]; I2 = 98.21%), while those using CLSI combined with EUCAST and EUCAST alone reported rates of 23% (95% CI [1, 56]; I2 = 98.16%) and 21% (95% CI [7, 39]; I2 = 95.40%), respectively (Supplementary Fig. 10). Regarding study design, cross-sectional studies reported a resistance rate of 10% (95% CI [6, 15]; I2 = 82.98%), whereas retrospective studies showed a higher rate of 28% (95% CI [21, 36]; I2 = 99.59%) (Supplementary Fig. 11).

Determination of the sources of heterogeneity

We performed a meta-regression to identify the potential sources of heterogeneity in the resistance rates among the included studies (Supplementary Table 4). The model explained 68.92% of the between-study variance, with a residual I2 of 97.42%. Among the covariates analyzed, the publication year showed a significant positive association with the resistance rate (β = 0.038, p < 0.001), suggesting an increase in fluoroquinolone resistance over time. The sample size also showed a significant positive association with the resistance rate (β = 0.001, p = 0.022), indicating a slight increase in resistance with the sample size. In addition, countries in Asia tended to have higher resistance rates than those in other regions (β = 0.255, p = 0.007), which indicated that the Asian region was a significant predictor of resistance. The use of the MIC method (β = 0.300, p < 0.001) was also significantly associated with resistance compared to other methods, highlighting the influence of methodological variability. In contrast, the type of fluoroquinolone (β =  − 0.053, p = 0.511) and the study quality (β = 0.024, p = 0.449) were not significantly associated with the heterogeneity in the resistance rates.

Discussion

M. morganii is a Gram-negative bacterium and opportunistic pathogen associated with various nosocomial infections70. The increasing resistance of M. morganii to fluoroquinolones is a significant concern, particularly in hospital settings5,71. In this systematic review and meta-analysis, we aimed to determine the global prevalence of fluoroquinolone-resistant M. morganii and to identify associated trends and contributing factors by analyzing data spanning multiple periods and regions.

When we analyzed M. morganii resistance to specific antibiotics, the results revealed that the clinical isolates were most resistant to pefloxacin (88%), followed by nalidixic acid (27%), levofloxacin (25%), and ciprofloxacin (22%). Lower resistance rates were observed for gemifloxacin (11%), ofloxacin (9%), trovafloxacin (8%), and garenoxacin (0%). Pefloxacin is a broad-spectrum second-generation fluoroquinolone that can be administered both orally and intravenously, making it a versatile option in clinical settings72. Although the isolates showed the highest resistance to pefloxacin, the limited sample size (17 isolates) affects the generalizability of this result (i.e., the extent to which this result can be applied to broader populations). There was moderate resistance among the clinical isolates to nalidixic acid (first generation), ciprofloxacin (second generation), and levofloxacin (third generation). No publication bias was observed for levofloxacin or ciprofloxacin. Ciprofloxacin is often considered to be more potent and to have broader activity against Gram-negative organisms than levofloxacin73. In some statistical findings, small differences in resistance rates between antibiotics—such as 1–2%—may reach statistical significance due to the large sample size. However, these differences may have limited clinical relevance. It is less susceptible to certain resistance mechanisms; for example, the AAC(6′)-Ib-cr enzyme can inactivate levofloxacin but not ciprofloxacin74. In addition, ciprofloxacin has been used to successfully treat several types of M. morganii infection72,74, while levofloxacin use is less frequently reported in the context of M. morganii infections. The relatively lower resistance to the newer fluoroquinolones (i.e., garenoxacin, trovafloxacin, and gemifloxacin) suggests that these may still be viable treatment options, although their use should be guided by local resistance patterns.

Geographically, the prevalence of fluoroquinolone-resistant M. morganii was highest in Africa (55%), followed by South America (36%), Asia (35%), North America (11%), Europe (9%), and Australia (4%). The global prevalence was 21%. The high level of resistant isolates in Africa may be attributed to lower public hygiene standards, varying attitudes toward antimicrobial treatment, high population densities facilitating microbial transmission, and widespread fluoroquinolone use75. In Asia, the prevalence of resistant isolates was higher in West Asia (62%) than in East Asia (22%) and Southeast Asia (1 with the studies conducted in Saudi Arabia and Iraq recording the highest prevalences of resistant isolates. These results indicate that the data obtained from the Asian region are biased toward two countries and that the emergence of fluoroquinolone-resistant M. morganii isolates in Saudi Arabia and Iraq warrants active microbiological surveillance. The geographic differences may be explained by the laboratory methods used to detect and identify fluoroquinolone-resistant M. morganii isolates. In addition, the trends found in the levels of resistant isolates align with antibiotic consumption patterns and regulatory policies76. In West Asia, high fluoroquinolone use in both human and veterinary medicine has been shown to contribute to increased resistance77, whereas East Asia’s relatively lower rates of resistant isolates may reflect the stricter antimicrobial stewardship programs employed in China, Japan, South Korea, and other countries78. Similarly, the lower rates of resistant isolates recorded in Europe (9%) and Australia (4%) were likely due to the antibiotic stewardship programs and regulatory frameworks enforced by agencies such as the European Medicines Agency (EMA)79 and Australia’s Therapeutic Goods Administration. In North America (11%), the Centers for Disease Control and Prevention and the Food and Drug Administration have implemented fluoroquinolone restriction policies, which have led to lower prescribing rates and reduced resistance80. However, in South America (36%), resistance remains high due to inadequate antimicrobial stewardship, widespread over-the-counter antibiotic sales, and inconsistent regulatory enforcement81.

Our temporal analysis of fluoroquinolone resistance in M. morganii revealed a significant increase over time. In the studies published between 1993 and 1997, the prevalence of fluoroquinolone-resistant isolates was low (2%), likely due to limited exposure to these antibiotics and their selective use. During this period, fluoroquinolones were introduced into clinical practice and used primarily as targeted therapeutic agents for treating severe infections rather than as an empirical treatment option82. The limited exposure to fluoroquinolones likely resulted in lower selection pressure, slowing the development of resistance83. Between 1998 and 2007, as fluoroquinolones became widely used for treating urinary tract and intra-abdominal infections, the rate of resistant isolates rose to 11%84. A more substantial increase (to 21%) occurred between 2008 and 2012, coinciding with the widespread adoption of fluoroquinolones in human and veterinary medicine, as well as the emergence of plasmid-mediated quinolone resistance (PMQR) genes, such as qnrD, in M. morganii11. Plasmid-mediated quinolone resistance (PMQR) in M. morganii is primarily associated with mechanisms such as Qnr proteins, which protect DNA gyrase and topoisomerase IV; the fluoroquinolone-modifying enzyme AAC(6′)-Ib-cr; and plasmid-encoded efflux pumps like QepA and OqxAB. The presence of these determinants in M. morganii and other Gram-negative bacteria underscores the risk of horizontal gene transfer and regional spread of resistance11. Between 2013 and 2017, the rate of fluoroquinolone resistance reached 30%, and between 2018 and 2022, it peaked at 37%. The most recent data, obtained in 2023–2024, indicated that the prevalence of resistant isolates was 33%. This prevalence aligns with the global rise in multidrug-resistant Gram-negative bacteria. The increased use of fluoroquinolones for empirical therapy, along with the horizontal transfer of PMQR genes among M. morganii and other Enterobacterales species, has further contributed to the surge in resistance85. These findings suggest that despite ongoing efforts to curb resistance, fluoroquinolone resistance in M. morganii is a persistent challenge that necessitates continued surveillance and the development of alternative treatment strategies86.

This study has several limitations. The asymmetrical distribution of data observed in some of the funnel plots in the subgroup analysis indicates potential publication bias. As a result, the subgroup analysis results should be interpreted with caution. Significant heterogeneity was observed across the included studies. To address this, we conducted a meta-regression to explore potential sources of heterogeneity. The variability was likely attributed to differences in study design. Additionally, variations in the types of antibiotics used across different regions in the treatment of M. morganii infections may have contributed to resistance development over time, thereby influencing the results. Another contributing factor was the wide range of sample sizes, which likely affected the overall estimates. Furthermore, as the included studies spanned from 1993 onwards, changes in resistance interpretation guidelines—such as updates from CLSI and EUCAST—along with differences in the antibiotic discs used by various manufacturers87, may have influenced the AST results, as supported by previous literature. Taken together, these factors likely contributed to the heterogeneity, which is to some extent unavoidable. Therefore, the pooled estimates should be interpreted with caution. Additionally, data availability was limited for certain regions, which may have introduced geographic bias. Furthermore, differences in research priorities and reporting practices among countries may have led to the underrepresentation of the prevalence of resistant isolates in some countries and, thus, potential publication bias. Variability in the laboratory techniques used to detect fluoroquinolone-resistant M. morganii isolates may also have contributed to inconsistencies in reported resistance rates.

Conclusion

In this comprehensive global assessment of fluoroquinolone-resistant M. morganii, we determined that there are significant geographic and temporal variations. Our findings support the use of ciprofloxacin over levofloxacin in the treatment of M. morganii infections and suggest that later-generation fluoroquinolones are more effective, possibly due to structural modifications that reduce their susceptibility to common resistance mechanisms, such as efflux pumps and target site mutations.

The trend of rising resistance, particularly in high-burden regions such as Africa, Asia, and South America, underscores the urgent need for more stringent antimicrobial stewardship programs, stricter regulatory policies, and enhanced surveillance efforts. Future research should focus on elucidating the molecular mechanisms underlying fluoroquinolone resistance, evaluating alternative therapeutic strategies, and assessing the impact of targeted interventions to mitigate the spread of resistance. Clinicians should consider local resistance patterns when selecting empirical therapeutic measures for M. morganii infections to ensure optimal treatment outcomes.