Antimicrobial resistance prevalence in clinical and aquatic environmental ESKAPE: a systematic review with meta-analysis

This systematic review and meta-analysis reveals that while antimicrobial resistance in ESKAPE pathogens is significantly more prevalent in clinical isolates than in aquatic environments, wastewater and effluent-impacted waters serve as critical reservoirs of resistance, highlighting the need for standardized methodologies within a One Health framework.

The Gist

Imagine a global game of "Hide and Seek," but instead of children, the players are tiny, invisible bacteria, and the seekers are the medicines (antibiotics) we use to cure infections. For a long time, we thought the bacteria were mostly hiding in hospitals, wearing their "super-villain" armor. But this new study asks a big question: Are they also hiding in our rivers, lakes, and wastewater, waiting to jump back into humans?

Here is a simple breakdown of what the researchers found, using some everyday analogies.

1. The "ESKAPE" Gang

First, the study focuses on a specific group of six bacteria known as ESKAPE. Think of them as the "Bad Boys" of the bacterial world. They are the ones most likely to break out of our medical treatments and make people very sick in hospitals.

• The Goal: The researchers wanted to see how many of these "Bad Boys" are resistant to medicine in two places: Hospitals (Clinical) and Water (Environmental).

2. The Big Reveal: Hospitals are the Main Fortress

The study looked at 18 different research papers from around the world. When they compared the two worlds, the results were clear:

• Hospitals: The bacteria here are like heavily armored tanks. About 67% of them are resistant to antibiotics. This makes sense because hospitals are where we use the most medicine, so the bacteria there have been forced to evolve super-fast defenses.
• Water (Rivers, Lakes, Wastewater): The bacteria here are more like lightly armed soldiers. Only about 24% are resistant.
• The Takeaway: The "Bad Boys" are definitely strongest in the hospital. That's where the biggest threat to patients right now is.

3. The "Polluted Water" Effect

However, the water isn't innocent. The study found that water near wastewater treatment plants (where sewage goes) had much higher resistance than clean, natural rivers.

• The Analogy: Imagine a river as a highway. If the highway is clean, cars (bacteria) drive normally. But if you dump a pile of "super-weapon" trash (antibiotics from hospitals and farms) into the river, the cars start building armor to survive the trash.
• The Finding: Water impacted by human waste (effluent) had higher resistance than pristine water. This means our wastewater is acting as a training ground, helping bacteria learn how to fight back against our medicines.

4. The "Fake Out" (Why some numbers looked weird)

Here is the tricky part. In a few specific categories of medicine (like Rifamycins and Polymyxins), the water bacteria seemed more resistant than the hospital bacteria.

• The Reality Check: The researchers realized this wasn't a real super-power for the water bacteria. It was a statistical illusion caused by how the studies were done.
• The Analogy: Imagine you are trying to count how many people in a city can lift 100 pounds.
  • In the hospital study, you ask everyone in the gym.
  • In the water study, you only ask people who already said, "I can lift heavy weights," because you used a special filter to catch only the strong ones.
  • Of course, your water study will look like it has stronger people!
• The Conclusion: Some studies used special "filters" (selective media) to catch only the super-resistant bacteria in the water, which made the numbers look scary high. When you look at the whole picture, the hospital bacteria are still the tougher opponents.

5. Why is the Data So Messy?

The researchers noted that comparing these studies was like trying to compare apples, oranges, and bananas.

• The Problem: Every study used different methods, tested different antibiotics, and sampled different types of water. It's like trying to measure the speed of cars when one study uses a stopwatch, another uses a radar gun, and a third just guesses.
• The Result: The data was extremely "noisy" (messy), making it hard to get a perfect number.

The Final Verdict

The "One Health" Lesson:
This study is a wake-up call for the One Health approach, which means we need to treat human health, animal health, and environmental health as one connected system.

• Good News: The bacteria in our rivers aren't quite as tough as the ones in our hospitals yet.
• Bad News: Our wastewater is acting as a mixing bowl where bacteria can swap genes and learn new tricks. If we don't clean up our water and stop dumping antibiotics into it, those "training grounds" might eventually produce bacteria that are just as dangerous as the hospital super-villains.

In short: The hospital is still the main battlefield, but the water is the enemy's secret training camp. We need to clean up the camp before the soldiers get too strong.

Technical Summary

1. Problem Statement

Antimicrobial resistance (AMR) in ESKAPE pathogens (Enterococcus faecium, Staphylococcus aureus, Klebsiella pneumoniae, Acinetobacter baumannii, Pseudomonas aeruginosa, and Enterobacter spp.) poses a critical global health threat. While these organisms are well-documented causes of healthcare-associated infections, their role as reservoirs in aquatic environments (rivers, wastewater, drinking water) remains under-characterized.

• The Gap: There is a lack of comparative data integrating clinical and environmental isolates within a "One Health" framework.
• The Challenge: Aquatic environments receive inputs from wastewater and agricultural runoff, potentially acting as transmission pathways for resistance genes. However, existing literature suffers from methodological heterogeneity, making it difficult to quantify the true prevalence of AMR in environmental matrices compared to clinical settings.

2. Methodology

The study followed PRISMA guidelines and was registered in PROSPERO (CRD420251020930).

• Search Strategy: A comprehensive search of PubMed, Embase, and the Cochrane Library was conducted up to January 14, 2025. No language or date restrictions were applied.
• Eligibility Criteria:
  • Inclusion: Original research reporting culture-based antimicrobial susceptibility data for ESKAPE pathogens isolated from both aquatic environmental matrices (water/wastewater) and human clinical samples.
  • Exclusion: Reviews, case reports, genomic-only studies (without phenotypic data), studies lacking comparative components, or those focusing exclusively on veterinary or ICU-only samples.
• Data Extraction: Two independent reviewers extracted data on pathogen type, geographic location, sample source, number of isolates, and resistance profiles for specific antibiotics.
• Statistical Analysis:
  • Model: Pooled resistance prevalence was estimated using binomial-normal generalized linear mixed models (GLMMs) with a logit link function to account for within- and between-study variability.
  • Heterogeneity: Assessed using Cochran's Q test, $\tau^2$, $I^2$, and $H^2$ statistics.
  • Bias Assessment: Small-study effects were evaluated using contour-enhanced funnel plots and Egger's regression test.
  • Software: Analyses were performed in R (v4.2.2) using metafor and meta packages.

3. Key Contributions

• Comparative Framework: This is one of the few meta-analyses to directly compare AMR prevalence in ESKAPE pathogens across clinical and aquatic environmental compartments using a unified statistical model.
• Methodological Critique: The study explicitly identifies and quantifies the impact of methodological artifacts (e.g., use of antibiotic-supplemented media for isolation) on resistance estimates, distinguishing between true ecological signals and statistical noise.
• One Health Insight: It provides evidence that while clinical settings remain the primary reservoir of high-level resistance, specific environmental niches (particularly wastewater) act as significant reservoirs, necessitating standardized monitoring protocols.

4. Key Results

Study Characteristics

• Inclusion: 18 studies met the inclusion criteria (out of 304 identified records).
• Geography: Predominantly European and Asian studies; Australia was the most frequent single-country location.
• Pathogens: The majority of data focused on E. faecium, P. aeruginosa, and K. pneumoniae. S. aureus and A. baumannii were less represented, and no eligible studies reported Enterobacter spp.

Overall Prevalence and Heterogeneity

• Overall Pooled Resistance: 0.46 (95% CI: 0.36–0.57).
• Heterogeneity: Extremely high ($I^2 = 98.8\%$), indicating significant variability across studies due to ecological and methodological differences.
• Clinical vs. Environmental:
  • Clinical Isolates: Significantly higher resistance (0.67; 95% CI: 0.55–0.77).
  • Environmental Isolates: Lower resistance (0.24; 95% CI: 0.14–0.39).
  • Effluent Impact: Environmental resistance was higher in effluent-impacted waters (0.28) compared to non-effluent sources (0.15).

Antibiotic Class Analysis

• General Trend: Clinical isolates showed higher resistance for most classes, including Macrolides (90.6% vs. 86.4%), $\beta$-lactams (74.1% vs. 46.6%), and Fluoroquinolones (60.9% vs. 17.8%).
• Anomalous Findings: Four antibiotic classes showed higher pooled resistance in environmental isolates:
  • Rifamycins (34.3% env vs. 3.3% clinical)
  • Nitrofurans (14.5% vs. 7.5%)
  • Polymyxins (9.5% vs. 3.3%)
  • Streptogramins (3.4% vs. 2.1%)
  • Interpretation: The authors attribute these reversals to methodological artifacts (e.g., selective media enrichment in wastewater studies) rather than true ecological dominance.

Small-Study Effects

• Significant funnel plot asymmetry was detected for aminoglycosides and fluoroquinolones (environmental) and $\beta$-lactams (clinical).
• Cause: The authors concluded this was not publication bias but rather a result of uneven representation of specific antibiotics within aggregated classes (e.g., a few studies testing highly resistant specific drugs skewing the class average).

5. Significance and Conclusion

• Clinical Dominance: AMR remains most prevalent in clinical settings due to intense selective pressures (high antibiotic use, vulnerable populations).
• Environmental Reservoirs: Aquatic environments, particularly wastewater, are confirmed reservoirs of resistance, with effluent-impacted waters showing elevated resistance levels.
• Methodological Imperative: The extreme heterogeneity ($I^2 > 98\%$) and anomalous findings highlight a critical lack of standardized protocols in environmental AMR surveillance. The use of selective media in environmental studies artificially inflates resistance estimates, complicating direct comparisons with clinical data.
• Future Directions: The authors call for a One Health approach utilizing harmonized sampling and analytical frameworks to accurately quantify transmission dynamics between environmental and clinical compartments. Without standardization, the true contribution of environmental reservoirs to the global AMR burden cannot be accurately assessed.