Author: admin

Introduction

Acute Kidney Injury (AKI) is a rapidly progressive decline in glomerular filtration rate (GFR) indicated by a rise in serum creatinine (SCr) of 0.3mg/dl or more within 48 hours or to 1.5 times the baseline in 7 days and/or reduction in urine output (UOP) to less than 0.5 mL/kg/hour for at least 6 hours.¹ Globally, AKI affects approximately 13.3 million people, causing 1.7 million deaths each year.² AKI is not only a very common condition, but also a predictor of morbidity, and can cause chronic kidney disease (CKD) or progress to kidney failure, which further complicate patient management and worsen prognosis.³ However, AKI may be reversible if detected early enough.⁴ Critically ill patients are more likely to develop AKI. AKI in critically ill adults arises from hemodynamic, inflammatory, and nephrotoxic factors. Ischemic injury due to reduced renal perfusion is a common pathway, particularly in conditions like sepsis and shock. The resulting hypoxia and oxidative stress can cause direct tubular damage and apoptosis.⁵ Inflammatory cytokines and mediators released during systemic infections and critical illnesses contribute to endothelial dysfunction, increased vascular permeability, and further renal injury.³

Moreover, the administration of nephrotoxic agents, either as part of therapeutic regimens or inadvertently, exacerbates renal injury. Drugs such as aminoglycosides, contrast agents used in diagnostic imaging, and certain chemotherapeutic agents have well-documented nephrotoxic effects.⁶ The cumulative impact of these factors leads to the clinical manifestation of AKI.

Generally, AKI affects about 10–15% of in-hospital patients, and more than 60% of patients admitted in ICUs.^7–9 Severe AKI occurs in about 4–5% of critically ill patients^9–11 due to resistant volume overload, uncontrollable electrolyte disorders, uremic complications, and drug toxicity. The exact mechanism of how AKI influences the clinical outcomes in severely ill patients remains unclear. However, it’s thought that it induces multi-system inflammatory responses.¹² Variations in reported incidences across studies on AKI among critically ill patients are attributed to differences in the study population, geographical area of study, patient baseline characteristics, length of observation period, and the criteria used to determine AKI.^13–18 Patients who develop AKI have an 8.8 times higher risk of developing CKD, posing far greater long-term health and cost consequences.^7,19–21 Early recognition and management of AKI can prevent its major complications.^22,23 Because critically ill patients may develop multi-organ dysfunction, the development of AKI in this population may have an impact on the outcome. Despite the overwhelming morbidity caused by AKI in critically ill patients, data on the incidence and predictors of AKI in this patient population in Uganda is limited. Bagasha et al (2015) studied the prevalence of AKI among adult patients with sepsis on the medical ward of a national referral hospital in Uganda and found a prevalence of 16.3% and in-hospital mortality of 21%.²⁴ Similarly, Kimweri et al (2021) studied the incidence and risk predictors of AKI among HIV-positive patients with sepsis at Mbarara Regional Referral Hospital. In their study, the incidence of AKI in 48 hours was 19.2%.²⁵ However, the incidence and predictors of AKI among other critically ill patient populations in Uganda have not been studied. This study intended to assess the incidence and independent predictors of AKI among critically ill patients at Mbarara Regional Referral Hospital (MRRH) in southwestern Uganda. The study also aimed to describe the management of AKI and evaluate the treatment outcomes among patients with incident AKI. The study, ultimately, aimed to guide health workers and policy makers to innovate and implement strategies towards mitigating the burden of AKI and reducing the associated complications in critically ill patients.

Methods

Study Setting

The site of the study was Mbarara Regional Referral Hospital (MRRH) in Mbarara city of southwestern Uganda. This is a tertiary 600-bed health facility serving a population of at least 4 million people from Mbarara and neighboring districts, including those from the Masaka Health Region and neighboring countries in the south western area. It is also the teaching hospital for Mbarara University of Science and Technology, Mayanja Memorial Training Institute and Bishop Stuart University.

The major specialized services provided are Emergency medicine (surgical and medical), Community Health, Internal Medicine, Obstetrics & Gynaecology, Paediatrics, and Surgery. The hospital’s emergency ward accommodates both medical and surgical patients. Patients will typically be admitted to the emergency ward for stabilization before they are transferred to the general medical and surgical wards for continued care. The hospital also has an 8-bed capacity ICU which serves critically ill patients. It also has 2 nephrologists and can provide hemodialysis to patients who require the intervention.

Study Design

This was a prospective cohort study among critically ill in-patients between 1^st February and 30^th May 2024.

Study Population

All critically ill adult patients who provided informed consent by themselves or through their next of kin were recruited into the study. The critically ill patients who were diagnosed with AKI at admission, or were dialysis-dependent, got discharged, or died within the first 48 hours of admission were excluded. Eligible participants were recruited consecutively over the study period without predefined sample size calculation, as recruitment was based on availability of critically ill patients within the study timeframe. Post-hoc power analysis was done to assess the sufficiency of the sample size to answer the study objectives.

Data Collection and Procedure

The team of research assistants who collected data were taken through a one-week’s training on the study protocol and tools such as the National Early Warning Score 2 (NEWS-2). They also completed the Responsible Conduct of Research (RCR) and Ethics in Research training, to enhance safety of the participants and integrity of data.

We used the “National Early Warning Score 2 (NEWS-2)” to assess and determine the degree of illness of a patient thus prompting critical care intervention.²⁶ A patient with a NEWS score of 5 or more was considered to be critically ill and requiring prompt emergency assessment. Critical illness in this study was defined as a potentially reversible life-threatening condition in which there is a decline in the function of vital organs, and death is imminent in case of absence of appropriate care.²⁷ Accordingly, critically ill patients were screened and consecutively enrolled from the intensive care unit (ICU), resuscitation bays at the medical emergency and surgical emergency wards as well as the adult general medical and surgical wards. Upon admission, patients were assessed using the “National Early Warning Score 2 (NEWS-2)”.

After obtaining informed consent, the relevant data including socio-demographic characteristics, medical history, admission vitals, NEW 2 score, baseline serum creatinine (SCr), blood urea nitrogen and serum electrolytes, complete blood count, random blood sugar, AKI management, exposure to nephrotoxic drugs, current drug and alcohol use were collected using a structured data abstraction form. Follow-up serum creatinine measurements were done after 48 hours to assess for incident AKI. Blood samples were tested at the Mbarara Regional Referral Hospital laboratory which does routine external quality assurance.

The “Kidney Disease Improving Global Outcomes (KDIGO)” AKI definition as “an increase in serum creatinine by greater or equal to 0.3mg/dl within 48 hours”²² was used for this study. Whenever CKD was suspected, the diagnosis was made using evidence from past records when available, history including the duration of symptoms, urinalysis and hematological indices, and the use of kidney ultrasound to determine the kidney sizes. The ultimate decision was made in consultation with the nephrologist.

Participants were followed until discharge, death or day 7 from enrolment, whichever would come first.

Data Statistical Analysis

Data cleaning was done using Epi-Info, after which data was imported into STATA version 13 for analysis. Numeric variables were summarized using means or medians with their respective measures of dispersion according to their distribution. Categorical variables were summarized using frequencies and percentages. The Fisher exact test or χ2 test was used to compare the participants baseline characteristics. The incidence of AKI was computed by dividing the number of participants who incurred AKI by the total number of study participants. The total person-days was computed as a sum of the days each of the participants stayed in the hospital. We computed the incidence rate by dividing the total number of participants who developed AKI by the total person-days, then multiplied the quotient by 1000. Univariate and multivariate logistic regression analysis was used to identify the independent predictors of AKI. A p-value ≤0.05 was considered statistically significant.

Ethical Considerations

This study was carried out in compliance with the Declaration of Helsinki. The MUST Research and Ethics Committee approved the study [Reference No: MUST-2023-1235] and site clearance was given by the hospital administration. The patient, or caretaker for those who were too sick, gave written informed consent before enrollment of any participant.

Results

Over 4 months, 420 patients were screened. Out of these, 220 were excluded because they were not critically ill. Out of 200 eligible patients, 28 were admitted to the ICU while 172 were admitted to both the surgical and medical/emergency wards. Overall, 161 participants who satisfied the eligibility criteria were involved in the final analysis. This data is shown in the study flow diagram in Figure 1.

Figure 1 Study flow diagram showing the enrolment of patients.

Baseline Characteristics of the Study Participants

The study sample comprised 161 participants, of whom the majority were male (59.6%), without a statistically significant difference in gender distribution between those who developed AKI and those who did not (64.6% vs 35.4%, P=0.432). The overall median age was 48 years (IQR: 31–65). The highest number of participants (52.8%) were admitted in the general medical or medical emergency wards, and only 11.2% of participants were admitted in the ICU.

There was a statistically significant increase in AKI incidence among participants who had been recently hospitalized within the past 3 months. About, 76.1% of the participants who developed AKI had reported a history of previous hospitalization compared to 23.9% of those without AKI (P=0.021). This data is elaborated in Table 1.

Table 1 Baseline Characteristics of Study Participants

Clinical Characteristics of Study Participants

Among the notable co-morbidities, hypertension was the most prevalent, affecting 26.1% of participants overall. However, the prevalence of comorbidities between the participants who incurred AKI and those who did not was comparable. The overall mean NEWS-2 (National Early Warning Score 2) was 11.9 (SD: 2.4), suggesting a high risk of clinical deterioration among participants. However, the comparative analysis did not show a statistically significant difference in the NEWS 2 scores of the “AKI” and “No AKI” groups (P = 0.102).

Among the laboratory parameters, the median WBC count was significantly higher in AKI patients (12.7 × 10⁹/L) compared to non-AKI patients (10.1 × 10⁹/L) (P=0.006). The median creatinine level at 0 hours was also significantly higher in AKI patients (1.0 mg/dl) compared to non-AKI patients (0.9 mg/dl) (P=0.028). This data is summarized in Table 2.

Table 2 Clinical Characteristics of the Study Participants (N=161)

Incidence of Acute Kidney Injury Among Critically Ill Patients

Among all the 161 participants who were followed up for a median duration of 6 days (IQR: 4–10), the incidence rate of AKI was 70 (95% CI 55–90) per 1000 person days of observation (Table 3). Out of the 100 participants who developed AKI, 84% (84/100) had stage 1 AKI, while those with stage 2 were 8/100 (8%) and stage 3 were 8/100 (8%).

Table 3 Incidence of Acute Kidney Injury Among Critically Ill Adults

Medications Used Among Critically Ill Patients Admitted at Mbarara Regional Referral Hospital

Out of 161 study participants, about 96 (60%) were exposed to at least one (mean, 0.8) potentially nephrotoxic drug during hospitalization (Figure 2). A total of 127 (39.1%) drugs used by the study participants during hospitalization were deemed potentially nephrotoxic. Penicillins, Angiotensin receptor blockers (ARBs), loop diuretics, proton pump inhibitors, Angiotensin converting enzyme inhibitors (ACEIs), and Tenofovir Disoproxil Fumarate, and Phenytoin being the most commonly used. See Figures 3 and Box 1.

Box 1 Other Medications Used by Critically Ill Patients During Hospitalization

Figure 2 Prevalence of potentially nephrotoxic drug exposure among study participants.

Figure 3 Commonest drugs used by study participants.

Predictors of AKI Among Critically Ill Patients

Patients who developed AKI had higher odds of having been previously hospitalized within the last 3 months, with a significant association observed in both univariable (cOR 2.44, 95% CI: 1.13–5.29, P=0.023) and multivariable analyses (aOR 2.56, 95% CI: 1.08–6.06, P=0.032). Admission to the surgical ward was another significant predictor of AKI (aOR 4.32, 95% CI: 1.22–15.24, P=0.023). Further, elevated serum creatinine at 0 hours (≥1.2 mg/dl) significantly increased the odds of developing AKI (aOR 2.44, 95% CI: 1.13–5.27, P=0.023). Similarly, a baseline WBC count ≥12 x10⁹/L was a significant independent predictor of AKI (aOR 2.57, 95% CI: 1.21–5.46, P=0.014) (Table 4).

Table 4 Predictors of Incident AKI Among Critically Ill Adults

Management and Outcomes of Incident AKI Among Critically Ill Adult Patients

Out of 100 participants who developed AKI, only 2% (n=2) underwent hemodialysis. The rest of the patients were managed conservatively. The mortality rate among the patients who developed AKI was 25% (25/100) compared to 13.1% (8/61) among the non-AKI group. Of the patients who survived, 19% (n=14) had complications secondary to the underlying illnesses. The commonest complication was hemiplegia (n=12). Long-term effects of AKI could not be assessed during the follow-up period of 7 days. Despite not having shown statistical significance in multivariable analysis, the patients who developed AKI had up to 2.2 more odds of dying than the ones who did not (OR 2.20, 95% CI: 0.92–5.27, P=0.074). The median overall time of hospitalization was 6 days (IQR: 4–10) with no significant difference between those who had AKI (6 days, IQR: 4–10) and those without AKI (7 days, IQR: 5–9.5) (P=0.307).

Discussion

Incidence of Acute Kidney Injury

We found an overall incidence rate of AKI of 70 (95% CI 55–90) per 1000 person days of observation among critically ill adult patients. This high incidence indicates the susceptibility of critically ill patients to AKI. Patients in low- and middle-income countries have a higher incidence of AKI than those in high-income countries. For example, Ashine et al conducted a retrospective follow-up study in Central Ethiopia, which revealed an incidence rate of AKI of 30.1 per 1000 person-days of observation.²⁸ We found the incidence rate of AKI in our population to be twice as high as that reported by that study. Conversely, a study by Susantitaphong et al (2013) in the United States reported incidence rates of AKI between 5% and 7% in critically ill patients. This is due to differences in patient demographics, healthcare practices, and underlying comorbidities.²⁹ Furthermore, majority of the patients had stage 1 AKI. Therefore, optimum patient care may be achieved by early recognition of AKI and initiating timely interventions which may prevent further kidney damage.³⁰

Predictors of Incident AKI Among Critically Ill Adult Patients

Previous hospitalization in the last 3 months was a strong predictor of AKI. This finding aligns with previous studies that showed a history of hospitalization to increase the risk of AKI, likely due to pre-existing comorbidities and exposure to nephrotoxic agents during previous hospital stays.³¹

The odds of developing AKI also significantly increased among patients admitted to the surgical ward. This could be resulting from postoperative complications and exposure to nephrotoxic medications among these patients. Hobson et al (2015) showed that patients with surgical conditions are at a risk for AKI due to fluid shifts and blood loss.³² This highlights the importance of monitoring and correcting fluid imbalances among surgical patients.

Patients with higher creatinine levels at admission also had higher odds of developing AKI. Coca et al (2009) had similar findings in their research which established baseline renal function as a critical determinant of AKI.³³ Patients with underlying renal impairment are likely to be more vulnerable to further insults during critical illness.

Having higher baseline white blood cell (WBC) count also significantly increased the odds of developing AKI. A raised WBC may reflect an underlying inflammatory state or infection contributing to AKI. Inflammatory processes contribute to the pathogenesis of AKI, and elevated WBC counts have been revealed to be a risk factor in previous studies as well.³⁴

Potentially Nephrotoxic Medications Use

In our study, more than half (60%) of study participants were exposed to potentially nephrotoxic medications during hospitalization. Avoidance or dose-adjustment of potentially nephrotoxic drugs is recommended in the acute care settings. However, lack of clear evidence for choice of equally effective but less nephrotoxic alternative drugs leads to inevitable use of nephrotoxic drugs like proton pump inhibitors, loop diuretics, NSAIDs, and some beta-lactam antibiotics like piperacillin/tazobactam. The high prevalence of nephrotoxic drug use and the diverse mechanisms of renal injury caused by these drugs might explain the high incidence of AKI in critically ill patients in resource-limited settings.

Management and Outcomes of Patients with Incident AKI

Only 2% of the patients who developed incident AKI were managed with hemodialysis. This is because hemodialysis bills at our hospital are met by the patients and their families, most of whom have no health insurance and therefore, cannot afford the service. On the contrary, higher rates of specialized treatments like hemodialysis are seen in high-income countries. This difference can be explained by the readily available resources and the different thresholds for initiating dialysis.²

Death occurred in 25% (25/100) of those who developed AKI compared to 13.1% (8/61) in the No-AKI group. Similarly, studies by Bellomo et al (2017) and Coca et al (2015) have shown high mortality rates among patients with AKI. The pooled mortality rate due to AKI in a meta-analysis by Susantitaphong, P.et al (2013) was 23.9%.^29,35,36

We found no difference in length of hospital stay between the AKI and No-AKI groups. However, other studies have demonstrated longer durations of hospital stay among patients with AKI. This has been attributed to the development of complications and the need for closer monitoring and management among these patients.^29,37

Limitations

Only one study site was used for this study, limiting the generalizability of the findings. We only measured serum creatinine at 2 time points; at admission and 48 hours later. This limited our follow-up of patients for the onset of AKI beyond 48 hours of admission. Serum creatinine takes time to increase, therefore, using its levels to detect AKI might lead to missing the detection in some cases. The short duration of follow-up for a maximum of 7 days did not allow for assessment of reduction in urine output during the study period and the long-term effects of AKI. Additionally, we did not assess the dosage and duration of exposure to individual nephrotoxic drugs. This limited our assessment of the causative relationship between nephrotoxic drug exposure and AKI incidence. Subsequent studies should focus on addressing these limitations to provide more insight into this topic.

Conclusions

There is a high incidence of AKI among critically ill patients. We found an incidence rate of 70 (95% CI 55–90) per 1000 person days of observation in this study.

A history of previous hospitalization within the past 3 months, having a baseline serum creatinine above 1.2 mg/dl, a white blood cell counts above 12 × 10⁹/L, and being admitted to the surgical ward were independently associated with incident AKI.

The majority of the patients with incident AKI received conservative management while only 2% underwent hemodialysis. A quarter of the participants with incident AKI died in hospital.

Recommendations

These findings highlight the importance of considering previous hospitalization, surgical admission, elevated baseline creatinine, and WBC counts as key predictors of AKI in critically ill patients. Prioritization of critically ill patients according to the number of these risk factors and subsequent closer monitoring might serve in the prevention, early diagnosis, and timely management of AKI, thus mitigating the burden of AKI in this vulnerable population.

We also recommend interventions at all levels to make RRT, particularly hemodialysis, more accessible and affordable in centers that manage critically ill patients to provide higher standard care and improve patient outcomes, especially among those with stage 3 AKI. For instance, one priority should be setting up a dialysis center at every regional referral hospital, training and employing dialysis nurses and equipping these centers with reagents and maintenance to minimize out-of-pocket expenditure for RRT services.

Furthermore, we recommend the implementation of standardized protocols for the management of AKI, and multidisciplinary care to optimize clinical outcomes and reduce mortality rates.

Further research on this topic should focus on studying the long-term effects of AKI in critically ill patients both during and after admission.

Abbreviations

AKI, Acute Kidney Injury; BMI, Body Mass Index; BP, Blood Pressure; CBC, Complete Blood Count; CKD, Chronic Kidney Disease; DM, Diabetes Mellitus; eGFR, Estimated Glomerular Filtration Rate; GCS, Glasgow Coma Score; GFR, Glomerular Filtration Rate; HIV, Human Immunodeficiency Virus; HT, Hypertension; ICU, Intensive Care Unit; KDIGO, Kidney Disease Improving Global Outcome; MAP, Mean Arterial Pressure; MRRH, Mbarara Regional Referral Hospital; MUST, Mbarara University of Science and Technology; RLS, Resource Limited Setting; RR, Respiratory Rate; RRT, Renal Replacement Therapy; SCr, Serum Creatinine; SSA, Sub-Saharan Africa; TDF, Tenofovir Disoproxil Fumarate; UOP, Urine Output; WHO, World Health Organization.

Data Sharing Statement

The complete datasets for this study will be availed by the corresponding author on request.

Acknowledgments

We acknowledge and appreciate all the study team members, hospital administration, and the participants for their consent to participate in the study.

Author Contributions

All authors made a significant contribution to the work reported, whether that is in the conception, study design, execution, acquisition of data, analysis and interpretation, or in all these areas; took part in drafting, revising or critically reviewing the article; gave final approval of the version to be published; have agreed on the journal to which the article has been submitted; and agree to be accountable for all aspects of the work.

Funding

This study was funded by the Internal Research Funds from the Mbarara University of Science and Technology Directorate of Research and Graduate Training (Grant number DRGT/SG/FY23-24/R4/T2P15).

Disclosure

The authors have no competing interests for this work.

References

1. KDIGO. KDIGO clinical practice guideline for acute kidney injury. 2012:1–141.

2. Hoste EA, Kellum JA, Selby NM, et al. Global epidemiology and outcomes of acute kidney injury. Nature Reviews Nephrology. 2018;14(10):607–625. doi:10.1038/s41581-018-0052-0

3. Chawla LS, et al. Acute kidney injury in critically ill patients. Clinical Journal of the American Society of Nephrology. 2014.

4. Abebe A, Kumela K, Belay M, et al. Mortality and predictors of acute kidney injury in adults: a hospital-based prospective observational study. Scientific Reports. 2021;11(1):15672. doi:10.1038/s41598-021-94946-3

5. Gomez H, Ince C, De Backer D, et al. A unified theory of sepsis-induced acute kidney injury: inflammation, microcirculatory dysfunction, bioenergetics, and the tubular cell adaptation to injury. Shock. 2014;41(1):3–11. doi:10.1097/SHK.0000000000000052

6. Kellum JA, Prowle JR. Paradigms of acute kidney injury in the intensive care setting. Nature Reviews Nephrology. 2018;14(4):217–230. doi:10.1038/nrneph.2017.184

7. Lameire NH, Bagga A, Cruz D, et al. Acute kidney injury: an increasing global concern. The Lancet. 2013;382(9887):170–179. doi:10.1016/S0140-6736(13)60647-9

8. Case J, Khan S, Khalid R, et al. Epidemiology of acute kidney injury in the intensive care unit. Critical care research and practice. Critical Care Res Practice. 2013;2013(1):479730. doi:10.1155/2013/479730

9. Uchino S. Beginning and Ending Supportive Therapy for the Kidney (BEST Kidney) investigators. acute renal failure in critically ill patients: a multinational, multicenter study. JAMA. 2005;294:813–818. doi:10.1001/jama.294.7.813

10. Metnitz PG, Krenn CG, Steltzer H, et al. Effect of acute renal failure requiring renal replacement therapy on outcome in critically ill patients. Critical Care Medicine. 2002;30(9):2051–2058. doi:10.1097/00003246-200209000-00016

11. Bagshaw SM, Laupland KB, Doig CJ, et al. Prognosis for long-term survival and renal recovery in critically ill patients with severe acute renal failure: a population-based study. Critical Care. 2005;9:1–10. doi:10.1186/cc3879

12. Rabb H, Griffin MD, McKay DB, et al. Inflammation in AKI: current understanding, key questions, and knowledge gaps. Journal of the American Society of Nephrology. 2016;27(2):371. doi:10.1681/ASN.2015030261

13. Chertow GM, Levy EM, Hammermeister KE, et al. Independent association between acute renal failure and mortality following cardiac surgery. The American Journal of Medicine. 1998;104(4):343–348. doi:10.1016/S0002-9343(98)00058-8

14. Escoresca Ortega A, et al. Kidney failure after heart transplantation. in Transplantation proceedings. 2010.

15. Srisawat N, Sileanu FE, Murugan R, et al. Variation in risk and mortality of acute kidney injury in critically ill patients: a multicenter study. American Journal of Nephrology. 2015;41(1):81–88. doi:10.1159/000371748

16. Hoste EA, De Corte W. Implementing the kidney disease: improving global outcomes/acute kidney injury guidelines in ICU patients. Current Opinion in Critical Care. 2013;19(6):544–553. doi:10.1097/MCC.0000000000000039

17. Zappitelli M, Parikh CR, Akcan-Arikan A, et al. Ascertainment and epidemiology of acute kidney injury varies with definition interpretation. Clinical Journal of the American Society of Nephrology. 2008;3(4):948–954. doi:10.2215/CJN.05431207

18. Horino T. Acute kidney injury: definition and epidemiology. acute kidney injury and regenerative medicine. 2020;3–20.

19. Mehta RL, Cerdá J, Burdmann EA, et al. International Society of Nephrology’s 0by25 initiative for acute kidney injury (zero preventable deaths by 2025): a human rights case for nephrology. The Lancet. 2015;385(9987):2616–2643. doi:10.1016/S0140-6736(15)60126-X

20. Horne KL, Packington R, Monaghan J, et al. Three-year outcomes after acute kidney injury: results of a prospective parallel group cohort study. BMJ Open. 2017;7(3):e015316. doi:10.1136/bmjopen-2016-015316

21. Coca SG, Singanamala S, Parikh CR. Chronic kidney disease after acute kidney injury: a systematic review and meta-analysis. Kidney International. 2012;81(5):442–448. doi:10.1038/ki.2011.379

22. Kellum JA, et al. Acute kidney injury. Nat rev Dise prime. 2021;7(1):1–17.

23. Anderson S, Eldadah B, Halter JB, et al. Acute kidney injury in older adults. Journal of the American Society of Nephrology. 2011;22(1):28–38. doi:10.1681/ASN.2010090934

24. Bagasha P, Nakwagala F, Kwizera A, et al. Acute kidney injury among adult patients with sepsis in a low-income country: clinical patterns and short-term outcomes. BMC Nephrology. 2015;16:1–7. doi:10.1186/1471-2369-16-4

25. Kimweri D, Ategeka J, Ceasor F, et al. Incidence and risk predictors of acute kidney injury among HIV-positive patients presenting with sepsis in a low resource setting. BMC Nephrology. 2021;22:1–5. doi:10.1186/s12882-021-02451-6

26. Anam AM, A.s FS, King MRG. Preventing Unrecognized Deterioration & Improving Outcomes of Critically Ill Patients Using the National Early Warning Score 2 in a High Dependency Unit in …. Tropical Doctor; 2023.

27. Kayambankadzanja RK, Schell CO, Gerdin Wärnberg M, et al. Towards definitions of critical illness and critical care using concept analysis. BMJ Open. 2022;12(9):e060972. doi:10.1136/bmjopen-2022-060972

28. Ashine TM, Mekonnen MS, Heliso AZ, et al. Incidence and predictors of acute kidney injury among adults admitted to the medical intensive care unit of a comprehensive specialized hospital in Central Ethiopia. PLoS One. 2024;19(6):e0304006. doi:10.1371/journal.pone.0304006

29. Susantitaphong P, Cruz DN, Cerda J, et al. World incidence of AKI: a meta-analysis. Clinical Journal of the American Society of Nephrology. 2013;8(9):1482–1493. doi:10.2215/CJN.00710113

30. Khwaja A. KDIGO clinical practice guidelines for acute kidney injury. Nephron Clinical Practice. 2012;120(4):c179–c184. doi:10.1159/000339789

31. Daratha KB, Short RA, Corbett CF, et al. Risks of subsequent hospitalization and death in patients with kidney disease. Clinical Journal of the American Society of Nephrology. 2012;7(3):409–416. doi:10.2215/CJN.05070511

32. Hobson C, O.-b.t K, Kuxhausen A, et al. Cost and mortality associated with postoperative acute kidney injury. Ann Surg. 2015;261(6):1207–1214. doi:10.1097/SLA.0000000000000732

33. Coca SG, Shlipak YB, Garg MG, Parikh AX, Parikh CR. Long-term risk of mortality and other adverse outcomes after acute kidney injury: a systematic review and meta-analysis. Am J Kidney Dis. 2009;53(6):961–973. doi:10.1053/j.ajkd.2008.11.034

34. Bagshaw SM, Bellomo GC, Bellomo R. Early acute kidney injury and sepsis: a multicentre evaluation. Critical Care. 2008;12(2). doi:10.1186/cc6863

35. Bellomo R, Kellum JA, Ronco C, et al. Acute kidney injury in sepsis. Intensive Care Medicine. 2017;43:816–828. doi:10.1007/s00134-017-4755-7

36. Coca SG, et al. Long-term risk of mortality and other adverse outcomes after acute kidney injury: a systematic review and meta-analysis. American Journal of Kidney Diseases. 2015.

37. Kashani K, Al-Khafaji A, Ardiles T, et al. Discovery and validation of cell cycle arrest biomarkers in human acute kidney injury. Critical Care. 2013;17. doi:10.1186/cc12503

Asrar A. Alahmadi, MBBS, is heading a first-in-human, open-label, phase 1 clinical trial (NCT06801002) investigating JBZ-001, an orally bioavailable small-molecule inhibitor of dihydroorotate dehydrogenase (DHODH) for the treatment of patients with advanced solid tumors and non-Hodgkin lymphoma.

Developed by Jabez Biosciences, JBZ-001 targets the de novo pyrimidine biosynthesis pathway by inhibiting DHODH, the rate-limiting enzyme in this pathway. Preclinical studies demonstrated JBZ-001’s broad antitumor activity across a range of cell lines and patient-derived xenograft models, particularly in aggressive cancers such as lymphoma, leukemia, and small cell lung cancer. The drug showed higher potency and lower toxicity compared with other agents in its class, suggesting its potential as both monotherapy and in combination with chemotherapy.

The phase 1 study now follows a standard 3+3 dose escalation design to determine the optimal biological dose (OBD), rather than the maximum tolerated dose. Pharmacokinetic and pharmacodynamic assessments include a novel blood-based biomarker to monitor DHODH activity in real time. Once OBD is reached, expansion cohorts will explore efficacy signals in specific tumor types.

The trial is currently enrolling patients, with future plans for combination studies and phase 2 trials in promising indications such as small cell lung cancer and hematologic malignancies, according to Alahmadi, assistant professor at The Ohio State University and lead principal investigator at The Ohio State University Comprehensive Cancer Center-James Cancer Hospital & Solove Research Institute.

In an interview with Targeted Oncology^TM, Alahmadi discussed the trial in progress and key goals of the study.

Targeted Oncology^TM: Can you discuss the mechanism of action of JBZ-001?

Alahmadi: The drug is a small molecule that inhibits a very critical enzyme involved in DNA and RNA synthesis, which is essential in rapidly dividing cells. That’s essentially what cancer is—unchecked, rapid cell proliferation. The molecule blocks DHODH, which is part of the pyrimidine biosynthesis pathway. This pathway is vital for cell survival during rapid division. The inhibition shows promising results against aggressive cancers like lymphoma and leukemia, and we’ve also seen potential in some relapsed solid malignancies like small cell lung cancer.

What preclinical data supported the initiation of this phase 1 study?

We observed efficacy in cell lines, patient-derived xenograft models, and through large-scale screening across multiple malignancies. We demonstrated broad activity, which we believe is due to the drug’s mechanism. It targets cancer cell metabolism, giving it a wide scope of action.

What is the study design, including the planned dose-escalation scheme and patient eligibility?

This is a first-in-human study. We’re testing the drug as a monotherapy using a standard 3+3 dose-escalation design. We’re collecting pharmacokinetic and pharmacodynamic data, including a blood-based marker—DHO/DHODH level—which correlates with enzyme activity. Our aim is to identify the optimal biological dose, rather than just the maximum tolerated dose. Once we reach OBD, we will proceed with expansion cohorts, especially in areas where we observe potential efficacy.

What are the key goals of the study? What are you hoping to learn about the tolerability of the drug?

We want to assess safety and tolerability, of course. Preclinical data suggest this agent may be more potent and less toxic than other DHODH inhibitors. We’re hoping this translates into the clinic—more efficacy with lower toxicity. Looking ahead, we aim to test it in combination with chemotherapy to enhance treatment outcomes in solid tumors. We also want to determine the optimal biological dose that balances efficacy and safety, which aligns with the FDA’s Project Optimus initiative.

Assuming the safety and preliminary efficacy are favorable, what are the potential next steps in the clinical development of this agent?

The next step would be testing in combination with chemotherapy. Also, if we observe efficacy signals in specific cancer types, we’ll open expansion cohorts and move into phase 2 trials. For example, we’re already seeing encouraging signs in small cell lung cancer. There’s evidence from other groups, including large CRISPR-based screens, suggesting that small cell lung cancer heavily depends on the pyrimidine biosynthesis pathway—making this an attractive target.

What are the key takeaways about this study and what should oncologists know about recruitment or the drug itself?

So far, our preclinical data suggest this drug may be more potent than other DHODH inhibitors. We’re currently enrolling patients in the dose expansion phase. I think that’s a key point.

I think we can always do better in our fight against cancer. Over the years, clinical trials have brought promising new treatments to market—treatments that genuinely improve patients’ lives, like targeted therapies and immunotherapies. With this development, we hope not only to show this drug is effective, but also to identify biomarkers of response. That way, we can deliver the right drug to the right patient at the right time.

REFERENCE:

Alahmadi A, Bennett C, Biglione S, et al. Abstract CT199: An open-label phase 1 study to investigate JBZ001 in adults with advanced solid tumors and non-Hodjkin lymphoma (JBZ001, trial in progress). Cancer Res. 2025;85 (8_Supplement_2): CT199. doi:10.1158/1538-7445.AM2025-CT199

Introduction

Antimicrobial resistance has become a global concern, posing significant challenges in the treatment of infectious diseases.¹ Beta-lactam antibiotics, including penicillins, cephalosporins, carbapenems, and aztreonam, are commonly prescribed for bacterial infections.² However, the widespread use of third generation cephalosporins to combat Gram-negative bacterial infections has contributed to the development of resistance against beta-lactam antibiotics.³ Consequently, this has led to the emergence of organisms producing Extended-Spectrum Beta-Lactamases (ESBLs) and carbapenemase, further exacerbating the issue.⁴ In 2024, the World Health Organization (WHO) classified carbapenem-resistant Pseudomonas aeruginosa, carbapenem-resistant Acinetobacter baumannii, and Carbapenem-resistant Enterobacterales (CRE) as “critical” and “high” priority pathogens.⁵

The spread of antibiotic-resistant organisms in health care facilities and in the community is a worrying epidemiological problem that could be influenced by the rate of faecal carriage of antibiotic-resistant organisms.⁶ The Enterobacterales, regardless of whether they are community or hospital-acquired, primarily originate from the digestive tract.⁷ This area is a hub for the exchange of resistance genes and antibiotic treatments, which can lead to the overgrowth of resistant bacteria.⁸ When antibiotic resistant isolates colonize the intestinal compartment, the risk of infection due to these bacteria significantly increases.⁹ Additionally, the intestinal compartment may act as a source for ESBL- and carbapenemase-producing resistance genes, which can spread to other Enterobacterales through horizontal transmission.¹⁰ Hence, being aware of the frequency of patients carrying resistant bacteria in their digestive tracts can serve as a means of forecasting the extent to which these bacteria may cause infections and spread among people.^11–13In the past decade, there has been a global rise in reported cases of hospital-acquired infections caused by Gram-negative bacilli (GNB) that are resistant to carbapenem antibiotics.^14–17 Hospitalized patients face a double burden as they not only have to deal with their health problems but also with a severe infection of this scale. Hence, this critical situation is a matter of concern for all healthcare systems.^18,19

In Africa, ESBL-PE (ESBL Producing Enterobacterales) carriage in the gastrointestinal tract ranges from 10% to 50%, highlighting the severity of the situation.²⁰ In Benin, a West African country, government has formally validated its National Action Plan (NAP); however, there are ongoing efforts to implement these measures, and as of now, the NAP has not yet been included on the WHO NAP website.²¹ More resources are required to develop laboratory testing capacity for the accurate detection of antibiotic-resistant bacteria throughout the country.²² While several studies have reported systematic data about AMR prevalence in general or focused on specific sample types, there is a lack of data regarding fecal carriage of antibiotic-resistant bacteria.^23–28 Recent studies in Benin have reported high rates of multidrug-resistant pathogens and revealed the first detection of a plasmid-encoded New-Delhi metallo-beta-lactamase-1 (NDM-1) producing Acinetobacter baumannii from surgical site infections within hospitals.^29,30 Prior to this, a study reported the first occurrence of hospital-originated Pseudomonas aeruginosa producing carbapenemase VIM-2.³¹ Additionally, there is an increased risk of infection associated with care,^24,32 and wound contamination by hospital bacteria.^24,33 Therefore, this study aims to determine the one-day prevalence of faecal carriage of ESBL and carbapenemase producing Gram-negative bacilli, along with associated risk factors and to characterize the resistant isolates.

Methods

Study Design and Sampling

This cross-sectional study, conducted in December 2022, focused on fecal samples from post-surgery patients (accidents and injury-related cases) in tertiary hospitals in Benin. A consecutive sampling approach was employed, where stool/rectal samples were collected from every consenting patient who had been hospitalized for more than 48 hours, until the target sample size of 30 was reached in each hospital. This 48-hour cutoff was used to distinguish hospital-acquired from community-acquired colonization. Four hospitals (Central Hospital and University of Mother and Child Lagoon of Cotonou, Departmental Hospital Center of Zou and the Hills, Departmental Hospital Center of Borgou-Alibori, and Hospital Saint Jean de Dieu of Tanguiéta) participated in the study, with patients with digestive pathologies excluded. The sample size was determined based on the bed capacity of the surgery wards in each hospital, ranging from 39 to 47 beds. All samples were collected on the same day, following strict aseptic protocols. A questionnaire was administered to gather sociodemographic information, clinical data, and medical history. The study received prospective ethical approval from the Ethics and Research Committee of the Institute of Applied Biomedical Sciences (CER-ISBA) under approval number 154. Written informed consent was obtained from each participant or their parent/guardian before participation, with a clear explanation of the study’s objectives. All research activities were conducted in accordance with the Declaration of Helsinki.

Species Identification, Antibiotic Susceptibility Testing, ESBL and Carbapenemase Detection

Fecal samples were screened for betalactam and carbapenem resistance using CHROMID^® ESBL and CHROMID^® CARBA SMART agar media (Bio-Rad, USA). After 24 hours of incubation at 37°C, distinct colonies were selected based on their characteristic color on the selective media. Pure cultures were obtained using again CHROMID^® ESBL and CHROMID^® CARBA SMART agar. Gram-negative bacilli, excluding Stenotrophomonas maltophilia, were identified using Matrix-Assisted Laser Desorption Ionization–Time of Flight (MALDI-TOF) mass spectrometry (Bruker Daltonics, Bremen, Germany). Antibiotic susceptibility testing was performed with inoculum from a subculture on Mueller–Hinton agar using the disc diffusion method on Mueller–Hinton agar, following recommendations from the European Committee on Antimicrobial Susceptibility Testing.³⁴ Various antibiotic discs manufactured by Bio-Rad were used for the testing. ESBL production was confirmed using the double disk synergy technique. For Enterobacterales, discs (Bio-Rad, United States) containing cephalosporins including cefotaxime 30 µg, ceftazidime 30µg, cefepime 30µg were applied next to clavulanic acid disc following recommendations from EUCAST.³⁵ For Pseudomonas spp, imipenem double-disc synergy test (DDS-IPM) with discs containing ceftazidime 30µg, cefepime 30µg and a disc with imipenem 10µg as the ESBL inhibitor was used.³⁶ Acinetobacter spp were tested for ESBL production by placing discs containing piperacillin (30 μg) and a combination of piperacilline + tazobactam (30–6µg) on agar plates.³⁷ Resistance to one of carbapenems tested was used as an indicator for the continuation of tests for detecting carbapenemase production. On these strains, the carbapenemase production test was carried out using the rapid diagnostic kit NG-TEST CARBA 5 (Hardy Diagnostics, California, United States) capable of detecting KPC, OXA-48-like, VIM, IMP, and NDM carbapenemases. Reference strain of Escherichia coli ATCC 25922 was used as quality control for the antibiotic susceptibility test.

Detection of Antibiotic Resistance Genes

Bacterial DNA was extracted from strains isolated from ESBL agar using boiling method. Multiplex PCR was performed to identify isolates carrying betalactamase encoding genes,³⁸ carbapenemase encoding genes³⁹ and PMQR genes.⁴⁰ Each reaction was prepared in a total volume of 25 µL using OneTaq^® Quick-Load^® 2X Master Mix with Standard Buffer (Biolabs, South Africa) and following the manufacturer instructions. The DNA fragments underwent electrophoresis in a 2% agarose gel, and the results were interpreted by comparing the migration of the fragments to marker sizes. The reference strain Escherichia coli ATCC 25922 was used as a negative control, while clinical strains producing ESBL and carbapenemase served as positive controls for the PCR assays.

Statistical Analysis

Data entry was performed using Excel 2019, and statistical analysis was conducted using Stata SE 11 software. Univariate analysis was initially conducted, selecting variables with a p-value less than 0.05 as potentially associated with ESBL and carbapenemase producing bacteria carriage. Results were reported as odds ratio (OR) with corresponding 95% confidence intervals, using an interpretation threshold of α = 0.05. The chi-square test was also performed to determine the degree of association between factors and independent variables.

Results

Factors Associated with the Carriage of ESBL and Carbapenemase Producing Gram Negative Bacteria

Table 1 described the factors associated with the carriage of ESBL Producing Gram negative bacteria. Out of the 120 patients enrolled in the study, 46 were identified as carriers of ESBL-producing bacteria, resulting in a prevalence of ESBL carriage at 38.33%. The study of the association of variables shows a high degree of association (p < 0.05) with ESBL carriage for the factors: independent walking and use of a wheelchair (Table 1).

Table 1 Factors Associated with the Carriage of ESBL Producing Bacteria

Table 2 described the factors associated with the carriage of carbapenem-resistant bacteria. Out of the 120 patients enrolled in the study, 59 were identified as carriers of carbapenem-resistant bacteria, resulting in a prevalence of 49.16%. Analysis of the association between variables revealed no statistically significant associations (p > 0.05) among the factors studied (Table 2).

Table 2 Factors Associated with the Carriage of Carbapenem-Resistant Bacteria

Species Identification and Percentage of Antibiotic Resistance

Table 3 shows the results of the identification of the isolated bacterial species. A total of 92 strains were isolated using ESBL agar. Among the bacterial species isolated from ESBL agar, Escherichia coli (44.56%) was the most frequently isolated, followed by Pseudomonas aeruginosa (26.08%) and Acinetobacter baumannii (10.87%). Notably, all Enterobacter cloacae strains (100%), most Escherichia coli strains (82.93%), and Klebsiella pneumoniae strains (80%) were found to be ESBL producers. A total of 83 strains were isolated using carba agar. The identification of isolates from carba agar showed a predominance of the species Escherichia coli (44.45%) followed by the species Klebsiella pneumoniae (16.06%) and Acinetobacter baumanii (13.58%). A total of 64 (77.10%) isolates were found to be resistant at least to one of the tested carbapenem antibiotics. Klebsiella pneumoniae (92.30%) and Escherichia coli (75%) were the most represented species.

Table 3 Distribution of Isolates According to the Culture Media

The resistance percentage of the isolates from ESBL agar is presented in Table 4. Notably, high rates of resistance were observed for ofloxacin (ranging from 97.11% to 100%), amoxicillin + clavulanic acid (ranging from 79.41% to 100%), and cefotaxime (ranging from 50% to 100%) among the tested antibiotics. Most Enterobacter cloacae, Escherichia coli, Klebsiella pneumoniae strains demonstrated multidrug resistance. On the other hand, Pseudomonas aeruginosa strains exhibited susceptibility to most of the tested antibiotics, except for cefsulodin. Furthermore, the lowest levels of resistance were observed with antibiotics such as imipenem, ertapenem, and temocillin (Table 4).

Table 4 Percentage of Antibiotic Resistance of Isolates Recovered from ESBL Screening Agar

The study of the resistance profile of Enterobacterales isolates from carba agar was presented in Table 5. The results showed that all isolates (100%) were resistant to amoxicillin and piperacillin. Similarly, a higher rate of resistance was observed to carbapenem, ranging from 87.50% to 100%. Various resistance rates were observed for other antibiotics.

Table 5 Percentage of Antibiotic Resistance of Enterobacterales Isolates Recovered from Carba Screening Agar

The study of the resistance profile of Pseudomonas spp was presented in Table 6. The results showed that isolate of Pseudomonas putida was resistant to all tested antibiotics excepted amikacin. Similarly, 80% (4/5) of Pseudomonas aeruginosa isolates exhibited resistance toward all tested antibiotics but some isolates (3/5) were susceptible to amikacin and aztreonam.

Table 6 Percentage of Antibiotic Resistance of Pseudomonas Spp Isolates Recovered from Carba Screening Agar

The study of the antibiotic resistance profile of Acinetobacter spp was presented in Table 7. The results showed that isolate of Acinetobacter nosocomialis was resistant to all tested antibiotics. Higher resistance rate (>80%) was also observed for Acinetobacter baumannii but 41.66% were susceptible to Amikacin.

Table 7 Percentage of Antibiotic Resistance of Acinetobacter Spp Isolates Recovered from Carba Screening Agar

Distribution of Carbapenemase Types

A total of 64 isolates resistant at least to one of the tested carbapenem antibiotics were selected for NG-TEST CARBA 5 (KPC, OXA-48-like, VIM, IMP, and NDM). All isolates selected were found to be carbapenemase producers. NDM (43.08%) was the most detected type of carbapenemase followed by strains for which carbapenemase type was not detected (30.77%) (Table 8).

Table 8 Distribution of Carbapenemase Types Among Species Isolated by Phenotypic Test

Table 9 Distribution of Detected Carbapenemase Gene Among Species Isolated

Antibiotic Resistance Genes Detection

Tables 9 and 10 present the distribution of resistance genes in carbapenemase and ESBL producing isolates. bla_NDM (54.68%) was the most detected carbapenemase genes. Among the ESBL-producing isolates, the presence of five ESBL genes was identified and presented in Table 10. The most detected genes were bla_CTXM-1 and bla_CTXM-15, accounting for 91.49% of the isolates. Additionally, the bla_TEM (36.17%) and bla_OXA-1 (29.78%) were found in the isolates. Quinolone resistance genes were prevalent among the isolates, with qnrb present in 70.21% of the isolates, followed by qnrs in 63.82% of the isolates, and aac(6′)-ib-cr in 53.19% of the isolates. Among the different species, Escherichia coli exhibited a high prevalence (97.05%) of the bla_CTXM-1 gene, while Enterobacter cloacae had a prevalence of 80% for the same gene. Electrophoresis gel images are shown in supplementary data (Figures S1–S6).

Table 10 Distribution of Resistance Genes in ESBL Producing Isolates

Discussion

The emergence and global spread of extended-spectrum β-lactamase (ESBL) and carbapenemase producing Enterobacterales pose a significant threat to public health.²⁰ The objective of this study was to determine the one-day prevalence of faecal carriage of ESBL and carbapenemase producing Gram-negative bacilli, along with associated risk factors, and to characterize the resistant isolates.

In this study, a culture-dependent approach was used to detect ESBL-producing- and carbapenem-resistant Gram-negative bacteria. The results of this study revealed a significant prevalence of carriage of ESBL-producing bacteria (38.33%), indicating a potential risk for nosocomial infections and the dissemination of antimicrobial resistance genes in the community. Lower rates of ESBL-PE carriage prevalence were reported in other African countries such as Gabon (11.8% to 16.7%), Cameroon (15% to 18%), Central African Republic (19.3%), and Nigeria (20.9%).^41–44 In Madagascar, the prevalence of ESBL-PE among individuals was found to be 10.1%.⁴⁵ On the other hand, a higher prevalence of ESBL-PE was reported in Burkina Faso.⁴⁶ Disparities in prevalence rates among countries and the increasing prevalence of ESBL-producing bacteria in Africa have been highlighted.⁴⁷ However, our findings do not align with their estimation that the overall rate of ESBL-PE in clinical samples was below 15%. Concerning carbapenem resistance detection, out of the 120 samples that were collected, a total of 81 strains were isolated. As 59 were identified as carriers of carbapenem-resistant bacteria, the prevalence of carbapenem resistance carriage is therefore 49.16%. A study conducted in Egypt revealed that 62.7% of Enterobacterales isolates were resistant to carbapenems, which is consistent with our findings.48 Similarly, high rates of carbapenem resistance were observed in South Africa (68%),⁴⁹ and Sudan (83%).⁵⁰ Respectively, 28.6% and 35% rates of resistance were observed in Uganda and Tanzania.^51,52 The prevalence of ESBL-PE values among individuals in different African countries varies significantly not only between countries, as underlined by the authors, but also between cities, sites (such as rural versus urban areas, and hospitals versus communities), and over different years. This variability should be highlighted even if the provided data remain indicative. The unrestricted use of antibiotics, which is prevalent among most of the population in low- and middle-income countries (LMICs) in Africa especially in Benin, is likely to lead to an increase in carbapenem resistance in the region.⁵³ In this study, the samples were collected within one day (24 hours), and it is not possible to estimate whether these results show an intermittent or regular situation in the study area. Sampling at different times of the year could give a more reliable profile of the situation.

In this study, no significant association was found between these risk factors and carriage of ESBL-producing bacteria. Similar investigations focusing on carbapenemase-producing isolates did not yield comparable results.^54,55 The study of the association of variables shows a high degree of association (p < 0.05) for the factors sex, independent walking, and use of a wheelchair. Patients who use wheelchairs or have limited mobility spend generally more time in healthcare facilities, potentially increasing their exposure to resistant isolates. Also, young patient age has been regarded as a risk factor for Carbapenem-resistant Enterobacterales (CRE) infection.⁵⁰ Length of hospital stay, sex, age, presence of immunosuppression, independent walking, bedridden patient, diabetic patient, presence of sign of infection, antibiotic treatment, history of hospitalization in the last 6 months, antibiotic treatment last 6 months, patient with faecal/urinary incontinence were the other risk factors examined in this survey. Neither of these risk factors was significantly associated with carbapenemase producing strains carriage. Numerous studies have shown interest in assessing the risk factors associated to carbapenemase producing isolates carriage but none of them have found similar results.^54,55

In total, 51.08% of the isolated strains were found to be ESBL producers. The identification of ESBL producing bacteria species revealed a predominance of Escherichia coli (82.93%), followed by Enterobacter cloacae (100%; n = 5) and Klebsiella pneumoniae (80%; n = 4). Notably, Enterobacter cloacae, Escherichia coli, and Klebsiella pneumoniae strains exhibited high proportions (100%, 82.93%, and 80% respectively) of ESBL production. Similar species predominance was observed in recent studies.^44,56 Other studies conducted in Cameroon and China have reported a higher diversity of ESBL-PE species, including Enterobacter spp. and Citrobacter spp.^57,58 Nevertheless, E. coli consistently emerges as the most identified species during colonization by ESBL-PE. It is worth noting that these bacteria, which pass through the human intestine, have the potential to acquire resistance through horizontal gene transfer.⁴⁶ The identification of bacteria with resistance to carbapenems showed a predominance of the species Escherichia coli (44.45%) followed by the species Klebsiella pneumoniae (16.06%) and Acinetobacter baumanii (13.58%). Same isolates were reported in various studies.^48–52 Patients with carbapenem-resistant Enterobacterales often face treatment failure, prolonged hospital stay, high expenses and a high possibility of death.⁵⁹

The antibiotic resistance phenotype results of isolates from ESBL agar showed high rate of resistance for ofloxacin (ranging from 97.11% to 100%), amoxicillin + clavulanic acid (ranging from 79.41% to 100%), and cefotaxime (ranging from 50% to 100%) among the tested antibiotics. Multidrug resistance was common among Enterobacter cloacae, Escherichia coli and Klebsiella pneumoniae strains. Similar observations were made,⁴⁴ indicating significant resistance not only to β-lactam antibiotics due to ESBL production but also to other antibiotic families, including quinolones. In this study, the carbapenems (imipenem and ertapenem) showed the highest effectiveness against ESBL-producing Gram-negative bacteria (GNBs). These carbapenems are considered expensive last-resort drugs in the region but have demonstrated favorable activity against GNBs that produce this enzyme.^60–62 This finding is consistent with a report which indicates the widespread distribution of ESBL producers among Enterobacterales.⁵⁰ The study of antibiotic susceptibility patterns of carbapenem-resistant isolates revealed that the majority of isolates were resistant to all tested antibiotics, including carbapenems. Distribution of carba types according to species isolated and ESBL production showed that 52.31% of isolates were ESBL producers, but none non-Enterobacterales was found to be ESBL producing. All isolates selected were found to be carbapenemase producers. NDM (43.08%) was the most detected type of carbapenemase followed by other types (30.77%) using the aforementioned phenotypic method. The NDM enzyme was frequently encountered among Enterobacterales as one of the most common carbapenemases, and it was also observed in A. baumannii and P. aeruginosa isolates.^63,64 Most of non-Enterobacterales were found to produce other types of carbapenemase. This is consistent with a study reporting that carbapenemase producers are becoming highly distributed among Enterobacterales.⁵⁰ Non-Enterobacterales including Pseudomonas spp employ mainly porin expression reduction and increased chromosomal cephalosporinase activity against carbapenems.⁶⁵

The study extensively examined the presence of ESBL genes among the isolates. The most frequently detected ESBL genes were bla_CTXM-1 and bla_CTXM-15 (91.49%), followed by bla_TEM (36.17%). These genes are associated with resistance to a broad range of β-lactam antibiotics. Similar findings have been reported in other studies conducted in Cameroon (96% of isolates), Indonesia (94.5%), and Tunisia (91%), indicating the global dominance of CTX-M-type ESBLs.^57,66,67 The study also investigated the presence of other resistance genes associated with quinolone and carbapenem resistance. The isolates exhibited the presence of quinolone resistance genes, such as qnrb (70.21%), qnrs (63.82%), and aac(6′)-ib-cr (53.19%), indicating the potential for reduced efficacy of fluoroquinolone antibiotics in treating infections caused by ESBL-producing bacteria. This result supports the concept co-expression due to the isolates harboring several antibiotic resistance genes. Furthermore, the presence of qnr genes and aac(6′)-ib-cr alongside ESBL-encoding genes further strengthens the possibility of co-selection.⁶⁸ These findings confirm the multidrug-resistant nature of ESBL-PE, which severely limits available therapeutic options and contributes to the widespread use of carbapenems. The detection of carbapenemase-encoding genes in a study is vital for understanding and addressing the growing problem of antibiotic resistance. It provides critical information for infection control, treatment decisions, and the development of strategies to preserve the effectiveness of antibiotics in the face of rising resistance. In this study, bla_NDM (54.68%) was the most detected carbapenemase genes mainly in Klebsiella pneumoniae (91.66%), Enterobacter cloacae (77.77%) and Pseudomonas aeruginosa (75%). Our results are consistent with others studies in Sudan, Senegal, and South Africa.^49,50,59 This confirms that bla_NDM could demonstrate a capacity for rapid dissemination.⁶⁹ The plasmids that carry carbapenemase genes, such as bla_NDM, exhibit significant diversity and can harbor numerous additional resistance genes, including ESBL-alleles. The presence of multiple resistance genes in certain isolates, as observed in this study, indicates the presence of multidrug-resistant pathogens. These pathogens are responsible for treatment failures, outbreaks of infections, and increased treatment costs.

The primary aim was to determine the prevalence of ESBL and carbapenemase-producing bacteria at a single time point in hospitalized post-surgical patients, providing baseline data for future longitudinal studies on acquisition dynamics. The one-day design and >48-hour hospitalization criterion limit differentiation between community- and hospital-acquired colonization. Without admission screening or follow-up, acquisition timing and transmission pathways cannot be confirmed. Also, Limited clinical and hospital data restrict analysis of risk factors and transmission routes. Therefore, conclusions about nosocomial spread should be interpreted cautiously. The absence of whole genome sequencing prevents detailed epidemiological and transmission analysis. Future studies with longitudinal sampling, comprehensive data, and genomic tools are needed to clarify acquisition and guide interventions.

Conclusion

This study reports a high one-day prevalence of fecal carriage of extended-spectrum β-lactamase (ESBL) and carbapenemase-producing Gram-negative bacteria among post-surgical patients hospitalized for over 48 hours in Benin. The predominance of Escherichia coli, Klebsiella pneumoniae, Pseudomonas aeruginosa, and Acinetobacter baumannii, combined with high resistance to β-lactams and fluoroquinolones, underscores the clinical and public health significance of these pathogens. Genotypic analysis revealed frequent detection of bla_CTX-M and bla_NDM, along with co-resistance genes, indicating the circulation of multidrug-resistant organisms in the hospital setting. While this cross-sectional study cannot confirm nosocomial acquisition, it provides important baseline data to inform future surveillance and infection control efforts in Beninese hospitals.

Abbreviations

CTX-M: Cefotaxime-Munich; ESBL: Extended-Spectrum β-lactamase; ESBL-PE: ESBL Producing Enterobacterales; GNB: Gram Negative Bacilli; MBL: Metallo-betalactamase; NDM: New Delhi Metallo-betalactamase.

Data Sharing Statement

All data generated and/or analyzed during the current study are included in this published article. The datasets used and/or analyzed during this study are also available from the corresponding author on reasonable request.

Ethical Approval

The study proposal was reviewed and approved by the Ethics and Research Committee of the Institute of Applied Biomedical Sciences (CER-ISBA) under number 154. Written informed consent was obtained from each patient or their parent/guardian before participation, accompanied by a concise explanation of the study’s objective. The research work (sampling from hospitalized patients, sample processing and data analysis) in our study was conducted in accordance with the Declaration of Helsinki.

Acknowledgments

The authors are very grateful to Alida Oussou, Donald Bokossa, Lydie Comlan, Adonias Houefonde and Yedia Djohoun for their great help during the implementation of this study. They are also very grateful to all the staff of the hospitals involved in the study and their willing to support any kind of interventions for a better care of patients. They thank the patients who accepted to participate in this study.

Author Contributions

All authors made a significant contribution to the work reported, whether that is in the conception, study design, execution, acquisition of data, analysis and interpretation, or in all these areas; took part in drafting, revising or critically reviewing the article; gave final approval of the version to be published; have agreed on the journal to which the article has been submitted; and agree to be accountable for all aspects of the work.

Funding

This research received no external funding.

Disclosure

The authors declare no conflict of interest.

References

1. Kedišaletše M, Phumuzile D, Angela D, Andrew W, Mae NF. Epidemiology, risk factors, and clinical outcomes of carbapenem-resistant Enterobacterales in Africa: a systematic review. JGAR. 2023;35:297–306.

2. Chelaru EC, Muntean AA, Hogea MO, Muntean MM, Popa MI, Popa GL. The importance of carbapenemase-producing Enterobacterales in African countries: evolution and current burden. Antibiotics. 2024;13(4):295. doi:10.3390/antibiotics13040295

3. Akingbade OA, Ogiogwa IJ, Okonko IO, et al. Plasmid profile of isolated Klebsiella species in a Tertiary Hospital in Abeokuta, Ogun State, Nigeria. World Appl Sci J. 2012;3:371–378.

4. Shristi R, Gokhale S, Adhikari B. Prevalence of extended spectrum beta-lactamases among Escherichia coli and Klebsiella spp isolates in Manipal Teaching Hospital, Pokhara, Nepal. JMID. 2015;5(2):69–75.

5. WHO. WHO bacterial priority pathogens list, 2024: bacterial pathogens of public health importance to guide research, development and strategies to prevent and control antimicrobial resistance; 2024. Available from: https://www.who.int/publications-detail-redirect/9789240093461. Accessed July 16, 2024.

6. Valverde A, Grill F, Coque TM, et al. High rate of intestinal colonization with extended-spectrum-β-lactamase-producing organisms in household contacts of infected community patients. J Clin Microbiol. 2008;46(8):2796–2799. doi:10.1128/JCM.01008-08

7. Woerther PL, Burdet C, Chachaty E, Andremont A. Trends in human fecal carriage of extended-Spectrum β-Lactamases in the community: toward the globalization of CTX-M. Clin Microbiol Rev. 2013;26(4):744–758. doi:10.1128/CMR.00023-13

8. La Fauci V, Costa GB, Genovese C, Palamara MAR, Alessi V, Squeri R. Drug-resistant bacteria on hands of healthcare workers and in the patient area: an environmental survey in Southern Italy’s hospital. Rev Esp Quimioter. 2019;32(4):303–310. doi:10.1056/NEJMoa1306801

9. Tajeddin E, Rashidan M, Razaghi M, et al. The role of the intensive care unit environment and health-care workers in the transmission of bacteria associated with hospital acquired infections. J Infect Public Health. 2016;9(1):13–23. doi:10.1016/j.jiph.2015.05.010

10. Weisenberg SA, Mediavilla JR, Chen L, et al. Extended spectrum beta-lactamase-producing Enterobacteriaceae. Int Travelers Non-Travelers New York City. 2012;7(9):e45141.

11. Ouédraogo AS, Sanou S, Kissou A, et al. Fecal carriage of Enterobacteriaceae producing extended-spectrum beta-lactamases in hospitalized patients and healthy community volunteers in Burkina Faso. MDR. 2017;23(1):63–70. doi:10.1089/mdr.2015.0356

12. Umezawa K, Asai S, Ohshima T, et al. Outbreak of drug-resistant Acinetobacter baumannii ST219 caused by oral care using tap water from contaminated hand hygiene sinks as a reservoir. Am J Infect Control. 2015;43(11):1249–1251. doi:10.1016/j.ajic.2015.06.016

13. Khan A, Miller WR, Arias CA. Mechanisms of antimicrobial resistance among hospital-associated pathogens. Expert Rev Anti Infect Ther. 2018;16(4):269–287. doi:10.1080/14787210.2018.1456919

14. Abdeta A, Bitew A, Fentaw S, et al. Phenotypic characterization of carbapenem non-susceptible gram-negative bacilli isolated from clinical specimens. PLoS One. 2021;16(12):e0256556. doi:10.1371/journal.pone.0256556

15. Raj S, Sharma T, Pradhan D, et al. Comparative analysis of clinical and genomic characteristics of hypervirulent Klebsiella pneumoniae from hospital and community settings: experience from a Tertiary Healthcare Center in India. Microbiol Spectr. 2022;10(5):e00376–22. doi:10.1128/spectrum.00376-22

16. Wei L, Wu Q, Zhang J, et al. Prevalence, virulence, antimicrobial resistance, and molecular characterization of Pseudomonas aeruginosa isolates from drinking water in China. Front Microbiol. 2020;11:544653. doi:10.3389/fmicb.2020.544653

17. Wu W, Feng Y, Tang G, Qiao F, McNally A, Zong Z. NDM metallo-β-lactamases and their bacterial producers in health care settings. Clin Microbiol Rev. 2019;32(2):e00115–18. doi:10.1128/CMR.00115-18

18. Jia H, Sun Q, Ruan Z, Xie X. Characterization of a small plasmid carrying the carbapenem resistance gene bla_OXA-72 from community-acquired Acinetobacter baumannii sequence type 880 in China. Infect Drug Resist. 2019;1545–1553. doi:10.2147/IDR.S202803

19. Zhu Y, Xiao T, Wang Y, et al. Socioeconomic burden of bloodstream infections caused by carbapenem-resistant Enterobacteriaceae. Infect Drug Resist. 2021;14:5385–5393. doi:10.2147/IDR.S341664

20. Bezabih YM, Sabiiti W, Alamneh E, et al. The global prevalence and trend of human intestinal carriage of ESBL-producing Escherichia coli in the community. J Antimicrob Chemother. 2021;76(1):22–29. doi:10.1093/jac/dkaa399

21. Sariola S, Butcher A, Cañada JA, Aïkpé M, Compaore A. Closing the GAP in antimicrobial resistance policy in Benin and Burkina Faso. mSystems. 2022;7(4):e00150–22. doi:10.1128/msystems.00150-22

22. Legba BB, Dougnon V, Koudokpon H, et al. Assessment of blood cultures and antibiotic susceptibility testing for bacterial sepsis diagnosis and utilization of results by clinicians in Benin: a qualitative study. Front Public Health. 2023;10:5487. doi:10.3389/fpubh.2022.1088590

23. Anago E, Ayi-Fanou L, Akpovi CD, et al. Antibiotic resistance and genotype of beta-lactamase producing Escherichia coli in nosocomial infections in Cotonou, Benin. Ann Clin Microbiol Antimicrob. 2015;14(1):1–6. doi:10.1186/s12941-014-0061-1

24. Dougnon V, Koudokpon H, Hounmanou YMG, et al. High prevalence of multidrug-resistant bacteria in the Centre Hospitalier et Universitaire de la Mère et de l’Enfant Lagune (CHU-MEL) reveals implications of poor hygiene practices in healthcare. SN Compr Clin Med. 2019;1(12):1029–1037. doi:10.1007/s42399-019-00149-3

25. Gbotche E, Dougnon V, Chabi Y, et al. Molecular characterization of Enterobacteriaceae producing β-lactamase and methicillin-resistant staphylococci isolated from the hospital environment and catheters in two public hospitals in Benin, Republic of Benin. Bio-Research. 2020;18(2):1164–1176. doi:10.4314/br.v18i2.5

26. Deguenon E, Dougnon V, Houssou VMC, et al. Hospital effluents as sources of antibiotics residues, resistant bacteria and heavy metals in Benin. SN Appl Sci. 2022;4(8):206. doi:10.1007/s42452-022-05095-9

27. Markkanen MA, Haukka K, Pärnänen KM, et al. Metagenomic analysis of the abundance and composition of antibiotic resistance genes in Hospital Wastewater in Benin, Burkina Faso, and Finland. Msphere. 2023;8:e00538–22. doi:10.1128/msphere.00538-22

28. Ombelet S, Kpossou G, Kotchare C, et al. Blood culture surveillance in a secondary care hospital in Benin: epidemiology of bloodstream infection pathogens and antimicrobial resistance. BMC Infect Dis. 2022;22(1):1–15. doi:10.1186/s12879-022-07077-z

29. Yehouenou C, Bogaerts B, Vanneste K, et al. First detection of a plasmid-encoded New-Delhi metallo-beta-lactamase-1 (NDM-1) producing Acinetobacter baumannii using whole genome sequencing, isolated in a clinical setting in Benin. Ann Clin Microbiol Antimicrob. 2021;20(1):1–7. doi:10.1186/s12941-020-00411-w

30. Yehouenou CL, Kpangon AA, Affolabi D, et al. Antimicrobial resistance in hospitalized surgical patients: a silently emerging public health concern in Benin. Ann Clin Microbiol Antimicrob. 2020;19(1):54. doi:10.1186/s12941-020-00398-4

31. Koudokpon H, Dougnon V, Hadjadj L, et al. First sequence analysis of genes mediating extended-spectrum beta-lactamase (ESBL) bla-TEM, SHV-and CTX-M production in isolates of Enterobacteriaceae in Southern Benin. Int J Infect. 2018;5(4):e83194.

32. Afle FCD, Agbankpe AJ, Johnson RC. Healthcare-associated infections: bacteriological characterization of the hospital surfaces in the University Hospital of Abomey-Calavi/so-ava in South Benin (West Africa). BMC Infect Dis. 2019;19:1–7. doi:10.1186/s12879-018-3648-x

33. Dougnon VT, Sintondji K, Koudokpon CH, et al. Investigating Catheter-related infections in Southern Benin Hospitals: identification, susceptibility, and resistance genes of involved bacterial strains. Microorganisms. 2023;11(3):617. doi:10.3390/microorganisms11030617

34. EUCAST. Comité de l’Antibiograme de la Société Française de Microbiologie. Société Française de Microbiologie; 2022. Available from: https://www.sfm-microbiologie.org/boutique/comite-de-lantibiograme-de-la-sfm-casfm/. Accessed November 17, 2022.

35. Tan TY, Ng LSY, He J, Koh TH, Hsu LY. Evaluation of screening methods to detect plasmid-mediated AmpC in Escherichia coli, Klebsiella pneumoniae, and Proteus mirabilis. AAC. 2009;53(1):146–149. doi:10.1128/AAC.00862-08

36. Weldhagen GF, Poirel L, Nordmann P. Ambler Class A Extended-Spectrum β-Lactamases in Pseudomonas aeruginosa: novel Developments and Clinical Impact. AAC. 2003;47(8):2385–2392. doi:10.1128/AAC.47.8.2385-2392.2003

37. Kumari M, Bhattarai NR, Rai K, Pandit TK, Khanal B. Multidrug-resistant Acinetobacter: detection of ESBL, MBL, bla genotype, and biofilm formation at a Tertiary Care Hospital in Eastern Nepal. Int J Microbiol. 2022;2022(1):8168000. doi:10.1155/2022/8168000

38. Dallenne C, Da Costa A, Decré D, Favier C, Arlet G. Development of a set of multiplex PCR assays for the detection of genes encoding important β-lactamases in Enterobacteriaceae. J Antimicrob Chemother. 2010;65(3):490–495. doi:10.1093/jac/dkp498

39. Cerezales M, Biniossek L, Gerson S, et al. Novel multiplex PCRs for detection of the most prevalent carbapenemase genes in Gram-negative bacteria within Germany. J Med Microbiol. 2021;70(3):001310. doi:10.1099/jmm.0.001310

40. Ciesielczuk H, Hornsey M, Choi V, Woodford N, Wareham DW. Development and evaluation of a multiplex PCR for eight plasmid-mediated quinolone-resistance determinants. J Med Microbiol. 2013;62(12):1823–1827. doi:10.1099/jmm.0.064428-0

41. Bercion R, Mossoro-Kpinde D, Manirakiza A, Le Faou A. Increasing prevalence of antimicrobial resistance among Enterobacteriaceae uropathogens in Bangui, Central African Republic. JIDC. 2009;3(03):187–190. doi:10.3855/jidc.34

42. Ogbolu DO, Daini OA, Ogunledun A, Alli AO, Webber MA. High levels of multidrug resistance in clinical isolates of Gram-negative pathogens from Nigeria. Int J Antimicrob Agents. 2011;37(1):62–66. doi:10.1016/j.ijantimicag.2010.08.019

43. Yala JF, Mabika RM, Bisseye C, et al. Phenotypic and genotypic characterization of extended-spectrum-beta-lactamases producing-Enterobacteriaceae (ESBLE) in patients attending Omar Bongo Ondimba military hospital at Libreville (Gabon). J Mol Microbiol Biotechnol. 2016;4(6):944–949.

44. Dikoumba AC, Onanga R, Boundenga L, Bignoumba M, Ngoungou EB, Godreuil S. Prevalence and characterization of extended-spectrum beta-lactamase-producing Enterobacteriaceae in major Hospitals in Gabon. MDR. 2021;27(11):1525–1534. doi:10.1089/mdr.2020.0497

45. Herindrainy P, Randrianirina F, Ratovoson R, et al. Rectal carriage of extended-spectrum beta-lactamase-producing gram-negative Bacilli in community settings in Madagascar. PLoS One. 2011;6(7):e22738. doi:10.1371/journal.pone.0022738

46. Ouedraogo AS, Sanou M, Kissou A, et al. High prevalence of extended-spectrum ß-lactamase producing Enterobacteriaceae among clinical isolates in Burkina Faso. BMC Infect Dis. 2016;16(1):326. doi:10.1186/s12879-016-1655-3

47. Tansarli GS, Poulikakos P, Kapaskelis A, Falagas ME. Proportion of extended-spectrum β-lactamase (ESBL)-producing isolates among Enterobacteriaceae in Africa: evaluation of the evidence—systematic review. J Antimicrob Chemother. 2014;69(5):1177–1184. doi:10.1093/jac/dkt500

48. Amer WH, Khalil HS, Wahab MAAEL. Risk factors, phenotypic and genotypic characterization of carpabenem resistant Enterobacteriaceae in Tanta University Hospitals, Egypt. Int J Infect Control. 2016;12(2). doi:10.3396/IJIC.v12i2.012.16

49. Perovic O, Britz E, Chetty V, Singh-Moodley A. Molecular detection of carbapenemase-producing genes in referral Enterobacteriaceae in South Africa: a short report: clinical update. SAMJ. 2016;106(10):975–977. doi:10.7196/SAMJ.2016.v106i10.11300

50. Elbadawi HS, Elhag KM, Mahgoub E, et al. Detection and characterization of carbapenem resistant Gram-negative bacilli isolates recovered from hospitalized patients at Soba University Hospital, Sudan. BMC Microbiol. 2021;21(1):1–9. doi:10.1186/s12866-021-02133-1

51. Mushi MF, Mshana SE, Imirzalioglu C, Bwanga F. Carbapenemase genes among multidrug resistant gram-negative clinical isolates from a tertiary hospital in Mwanza, Tanzania. BioMed Res Int. 2014;2014:1–6. doi:10.1155/2014/303104

52. Okoche D, Asiimwe BB, Katabazi FA, Kato L, Najjuka CF. Prevalence and characterization of carbapenem-resistant Enterobacteriaceae isolated from Mulago National Referral Hospital, Uganda. PLoS One. 2015;10(8):e0135745. doi:10.1371/journal.pone.0135745

53. Dougnon V, Chabi Y, Koudokpon H, et al. Prescription of antibiotics as a source of emerging antibiotic resistance: knowledge, attitudes, and practices of medical staff in the Dassa-Glazoué and Savalou-Bantè’s health zones (Benin, West Africa). Int J One Health. 2020;6:2455–8931. doi:10.14202/IJOH.2020.34-40

54. Hosseinzadeh-sohi N, Najar-Peerayeh S, Bakhshi B. Intestinal Carriage of Extended-spectrum β-lactamase and Carbapenemase-producing Enterobacteriaceae in Hemodialysis Patients. Jundishapur J Microbiol. 2022;15(3). doi:10.5812/jjm-118946

55. Zhao Z, Xu X, Liu M, Wu J, Lin J, Li B. Fecal carriage of carbapenem-resistant Enterobacteriaceae in a Chinese university hospital. AJIC. 2014;42(5):e61–4.

56. Toudji AG, Djeri B, Karou SD, Tigossou S, Ameyapoh Y, De Souza C. Prévalence des souches d’entérobactéries productrices de bêta-lactamases à spectre élargi isolées au Togo et de leur sensibilité aux antibiotiques. IJBCS. 2017;11(3):1165–1177.

57. Lonchel CM, Meex C, Gangoué-Piéboji J, et al. Proportion of extended-spectrum ß-lactamase-producing Enterobacteriaceae in community setting in Ngaoundere, Cameroon. BMC Infect Dis. 2012;12:1–7. doi:10.1186/1471-2334-12-53

58. Zhang H, Zhou Y, Guo S, Chang W. High prevalence and risk factors of fecal carriage of CTX-M type extended-spectrum beta-lactamase-producing Enterobacteriaceae from healthy rural residents of Taian, China. Front Microbiol. 2015;6:239. doi:10.3389/fmicb.2015.00239

59. Sarr H, Niang AA, Diop A, et al. The Emergence of Carbapenem-and Colistin-Resistant Enterobacteriaceae in Senegal. Pathogens. 2023;12(8):974. doi:10.3390/pathogens12080974

60. Ejikeugwu PC, Ugwu CM, Araka CO, et al. Imipenem and meropenem resistance amongst ESBL producing Escherichia coli and Klebsiella pneumoniae clinical isolates. Int Res J Microbiol. 2012;3(10):339–344.

61. Dirar M, Bilal N, Ibrahim ME, Hamid M, Hamid ME. Resistance patterns and phenotypic detection of β-lactamase Enzymes among Enterobacteriaceae isolates from referral hospitals in Khartoum State, Sudan. Cureus. 2020;12(3). doi:10.7759/cureus.7260

62. Ibadin EE, Omoregie R, Anogie NA, Igbarumah IO, Ogefere HO. Prevalence of extended spectrum β-lactamase, AmpC β-lactamase and metallo-β-lactamase among Gram negative bacilli recovered from clinical specimens in Benin City, Nigeria. Int J Enteric Pathog. 2017;5(3):85–91. doi:10.15171/ijep.2017.20

63. Cantón R, Akóva M, Carmeli Y, et al. Rapid evolution and spread of carbapenemases among Enterobacteriaceae in Europe. CMI. 2012;18(5):413–431. doi:10.1111/j.1469-0691.2012.03821.x

64. Munoz-Price LS, Poirel L, Bonomo RA, et al. Clinical epidemiology of the global expansion of Klebsiella pneumoniae carbapenemases. Lancet Infect Dis. 2013;13(9):785–796. doi:10.1016/S1473-3099(13)70190-7

65. Meletis G, Exindari M, Vavatsi N, Sofianou D, Diza E. Mechanisms responsible for the emergence of carbapenem resistance in Pseudomonas aeruginosa. Hippokratia. 2012;16(4):303.

66. Dahmen S, Bettaieb D, Mansour W, Boujaafar N, Bouallègue O, Arlet G. Characterization and molecular epidemiology of extended-spectrum beta-lactamases in clinical isolates of Enterobacteriaceae in a Tunisian University Hospital. Microb Drug Resist. 2010;16(2):163–170. doi:10.1089/mdr.2009.0108

67. Severin JA, Mertaniasih NM, Kuntaman K, et al. Molecular characterization of extended-spectrum beta-lactamases in clinical Escherichia coli and Klebsiella pneumoniae isolates from Surabaya, Indonesia. J Antimicrob Chemother. 2010;65(3):465–469. doi:10.1093/jac/dkp471

68. Tadesse BT, Ashley EA, Ongarello S, et al. Antimicrobial resistance in Africa: a systematic review. BMC Infect Dis. 2017;17:616. doi:10.1186/s12879-017-2713-1

69. Karabay O, Altindis M, Koroglu M, Karatuna O, Aydemir ÖA, Erdem AF. The carbapenem-resistant Enterobacteriaceae threat is growing: NDM-1 epidemic at a training hospital in Turkey. Ann Clin Microbiol Antimicrob. 2016;15(1):1–6. doi:10.1186/s12941-016-0118-4

Introduction

Obstructive sleep apnea (OSA) is characterized by recurrent sleep disturbances leading to intermittent hypoxemia, hypercapnia, and disrupted sleep architecture.^1–3 An epidemiological study indicates that the estimated prevalence of OSA with an apnea-hypopnea index (AHI) ≥5 events/hour ranges from 17% to 26% in men and 9% to 28% in women, with rates of 4% to 9% in middle-aged men and 1% to 2% in middle-aged women.⁴ According to the diagnostic criteria of AHI ≥5 events/hour, it is estimated that there are 176 million OSA patients in China, with 66 million being moderate to severe cases.^5,6 Studies have demonstrated that OSA can lead to nocturnal hypoxia, resulting in sympathetic activation, inflammation, oxidative stress, metabolic disorders, and endothelial dysfunction, potentially leading to multi-organ and multi-system damage such as cardiovascular diseases, diabetes, arrhythmias, cerebrovascular accidents, and cognitive impairments.^3,7–9 Untreated severe OSA patients have a mortality rate 3.8 times higher than that of the general population. The probability of stroke occurrence in the OSA population is 4.33 times higher than in the control group, with a mortality rate 1.98 times higher than the control group.^10,11

The primary assessment indicators for OSA currently include total AHI, lowest blood oxygen saturation, longest apnea duration, and longest hypoventilation period. Clinical experience demonstrates that relying solely on these indicators does not provide a comprehensive reflection of the severity of the patient’s condition. The longest apnea duration, longest hypoventilation period, and lowest blood oxygen saturation measurements offer insights at specific time points, while the AHI value merely indicates the frequency of events throughout the night without capturing the duration and extent of hypoxia.^12,13 This lack of effective evaluation indicators leads to inadequate understanding and assessment of OSA conditions, resulting in insufficient patient awareness of the condition’s severity and subsequently lower rates of consultation, diagnosis, and treatment for OSA.¹⁴ Economical and efficient screening for patients with OSA, coupled with a thorough evaluation of their condition, can enhance patient awareness and facilitate early intervention, thereby generating significant societal benefits and enhancing the overall quality of life for the population. The China-PAR model has been developed to validate 10-year risk predictions for atherosclerotic cardiovascular disease (ASCVD) among the Chinese population. This model incorporates various factors, including gender, age, waist circumference, total cholesterol, high-density lipoprotein cholesterol, treated or untreated systolic blood pressure, current smoking status (yes/no), diabetes mellitus (yes/no), and family history of ASCVD.¹⁵ It emphasizes anthropometric measurements and risk factors that are commonly observed in patients with OSA. Therefore, the China-PAR model is particularly suitable for assessing the 10-year ASCVD risk in patients with OSA in our study.

The investigation into cardiovascular risks among patients with OSA during episodes of hypoxia is gaining increasing significance. Nocturnal hypoxia can induce oxidative stress, inflammation, sympathetic activation, endothelial dysfunction, metabolic dysregulation, and pro-thrombotic effects and drive ASCVD and myocardial ischemia.^16–18 Previous studies have shown that recurrent hypoxia in OSA is related to pulmonary arterial hypertension, and vascular and microvascular diseases in patients.^1,2,10,11,19 The importance of new body indices and hypoxic burden metrics in the assessment of cardiovascular disease risk in OSA patients cannot be overlooked. For example, Yeşildağ M et al reported that the body roundness index, a novel body metric for defining and predicting cardiovascular risk in patients with OSA, was significantly associated with the oxygen desaturation index, a critical metric that measures hypoxic burden and links OSA to cardiovascular conditions.²⁰ However, the association between hypoxia events and patient prognosis assessment remains unclear. The hypoxic burden is calculated by dividing the sum of the product of the desaturation area associated with each pause or hypopnea event by the total sleep time. It can be automatically calculated by advanced sleep monitoring data analysis software, reflecting the intensity and duration of hypoxic events.²¹ The percentage of time with blood oxygen saturation below 90% (T90%) reflects the total duration of hypoxia throughout the night.²² However, due to the lack of comprehensive prognostic analysis, there are currently no standardized grading criteria for the hypoxic burden and T90%. The role of hypoxic burden and T90% in assessing the condition and prognosis of OSA patients remains to be further explored.

Polysomnography (PSG) is considered the primary diagnostic tool for identifying OSA. However, the equipment associated with PSG is costly, requires a specific installation environment, involves complex operation, demands high proficiency from medical personnel, necessitates completion in medical facilities during the examination process, and often leads to a suboptimal patient experience. These factors make it challenging to use PSG efficiently in screening for OSA among high-risk populations. Studies have shown that portable sleep monitors demonstrate a high level of agreement with PSG outcomes and offer reliable diagnostic capabilities for OSA, prompting an increased recognition of their clinical utility.^23,24

Here, by using portable sleep apnea monitors, we investigated the relationship between oxygen desaturation burden and T90% and myocardial ischemia and cardiovascular risk, promoting the establishment of a more comprehensive OSA assessment system. Our research has the potential to serve as a foundation for OSA patients to receive enhanced diagnostic and therapeutic interventions, ultimately leading to improved patient outcomes.

Materials and Methods

Ethics Statement

This study was carried out in accordance with the Declaration of Helsinki. All procedures were performed in compliance with relevant laws and institutional guidelines and have been approved by the Ethics Committee of Shenzhen Qian Hai She Kou Free Trade Zone Hospital (No.: 2021K-057). Informed content was obtained from all patients.

Study Participants

This prospective observational study was conducted at a single tertiary center in Southern China from January 2022 to March 2024. Hospitalized patients who underwent portable sleep apnea screening tests after being highly suspected of having OSA were screened. Only patients who were subsequently diagnosed with OSA were included. Inclusion criteria: 1) patients with daytime sleepiness, lack of energy after waking up, fatigue, or insomnia. 2) patients who woke up due to suffocation, wheezing, or choking at night. 3) patients with habitual snoring or breathing interruptions. 4) patients with dizziness. 5) patients with hypertension or poor response to hypertension treatment. 6) patients with emotional disorders. Exclusion criteria: 1) patients with known coronary artery diseases; 2) patients with a history of stroke; 3) patients with arrhythmias or those who used antiarrhythmic medications; 4) patients with cardiomyopathy; 5) patients with valvular heart disease; 6) patients who had electrolyte imbalances; 7) patients with severe kidney insufficiency.

Diagnostic Criteria for OSA

The diagnosis and initial assessment of OSA were based on the International Classification of Sleep Disorders.²⁵ The diagnostic threshold for OSA was defined by an AHI of ≥ 5 events per hour, where AHI values of 5–15 were classified as mild, 15–30 as moderate, ≥ 30 as severe, and ≥ 60 as extremely severe. The severity of nocturnal hypoxia was classified into five categories: normal, mild, moderate, severe, and very severe, determined by the lowest oxygen saturation levels: ≥90%, 85%-89%, 80%-84%, and <80%.

Sleep Monitoring

Participants were instructed to wear the portable sleep monitor, Phillip Alice Night One (Respironics, CA, USA), for overnight sleep monitoring. This device has been validated against PSG and has demonstrated a high level of agreement with PSG outcomes.^23,24 Sleep duration was assessed by considering the position, nasal airflow patterns, and self-reported sleep onset time of patients. Data was collected using Sleepware G3 4.0.1.0 (Respironics, CA, USA), manually reviewed and interpreted. Then, the hypoxic burden and other assessment parameters were automatically calculated. The recorded indicators included: AHI, hypoxic burden, supine hypoxic burden, non-supine hypoxic burden, lowest blood oxygen saturation, percentage of total sleep time with apneas, percentage of total sleep time with hypopneas, and percentage of total sleep time with blood oxygen saturation below 90% (T90%), 85% (T85%), 80% (T80%).

Electrocardiogram (ECG) Examinations

All patients underwent routine ECG examinations. The ECGs were assessed and reported by physicians at or above the attending level. The ECG indicators for evaluating myocardial ischemia included: 1) ST segment depression or elevation (ST segment depression in two or more adjacent leads, with limb leads ≥ 0.05 mV or precordial leads ≥ 0.1 mV); 2) T wave flattening or inversion); 3) ST-T changes; 4) positive PTFV1 (P-wave terminal force in electrocardiogram lead V1).

Assessment of the 10-Year Atherosclerotic Cardiovascular Disease (ASCVD) Risk

The process for assessing the 10-year ASCVD risk using the China-PAR model involved two steps.¹⁵ In Step 1, high-risk individuals were directly identified based on specific criteria, including 1) diabetic patients aged ≥40 years; 2) LDL ≥4.9 mmol/L or total cholesterol ≥7.2 mmol/L; 3) chronic kidney disease at stage 3/4. Due to the potential impact of kidney insufficiency on electrolytes and electrocardiographic waveforms, patients with chronic kidney insufficiency were excluded from this study. In Step 2, the remaining patients were subjected to the 10-year ASCVD risk assessment using the China-PAR model. Patients were categorized as either having low or moderate-to-high risk of ASCVD.

Data Collection

The baseline clinical data of patients were collected, including medical history, personal history, medication history, height, weight, abdominal circumference, and current residence. The data on biochemical indicators (such as total cholesterol, high-density lipoprotein, low-density lipoprotein (LDL), and triglycerides) were also collected.

Sample Size Calculation and Power Analysis

The sample size was determined using the standard formula. Assuming a moderate effect size corresponding to an expected odds ratio (OR) of approximately 3–10 and an estimated risk of ASCVD around 50% (α=0.05), we calculated a necessary sample size of approximately 200 to 350 participants for adequate statistical power.

Power analysis was conducted to validate the statistical power of our sample size, using the following formula: Power = 1 – β, where β indicates Type II error rate (commonly set at 0.2 for 80% power).

Statistical Analysis

Statistical analyses were conducted using SPSS 19.0 software. The normally distributed continuous data are expressed as mean ± standard deviation, and comparisons between two groups were performed using independent samples t-test. The non-normally distributed continuous data are expressed as median (interquartile range), and group comparisons were conducted using the Kruskal–Wallis H-test. Categorical data are presented as counts (percentages) and were analyzed using the χ²-test. A P-value of < 0.05 was considered statistically significant.

Binary logistic regression was used to analyze the relationships between the factors in the model and the risk for moderate-to-high 10-year ASCVD and myocardial ischemia. The factors analyzed in the model included AHI, hypoxic burden quartiles, percentage of total sleep time with apneas, percentage of total sleep time with hypopneas, percentage of total sleep time with apneas and hypopneas, T90%, T85%, T80%, and grade of lowest blood oxygen saturation. Hypoxic burden, percentage of total sleep time with hypopneas, percentage of total sleep time with apneas, and percentage of total sleep time with apneas and hypopneas were categorized into four quartiles. Confounders including gender, age, waist circumstance, total cholesterol, high-density lipoprotein, treated or untreated systolic blood pressure, current smoking (yes/no), diabetes mellitus (yes/no), and family history of ASCVD were adjusted. The OR and 95% confidence interval (CI) values were recorded. The receiver operating characteristic (ROC) curve was plotted to assess the diagnostic values of the hypoxic burden and T90%. The area under the curve (AUC), P-value, the optimal cut-off value, sensitivity, specificity, and Youden’s index were calculated. The maximum Youden’s index corresponds to the optimal balance of sensitivity and specificity, thereby determining the optimal cutoff value.

Results

Baseline Clinical Data of Patients

This study included a total of 311 hospitalized patients diagnosed with OSA from January 2022 to March 2024. Basic clinical information of the patients is summarized in Table 1. The median age of the included patients was 53 years, with 75.6% being male and 24.4% female. Among the patients, 51.4% (160 cases) showed ECG changes of myocardial ischemia. Additionally, 55.3% (172 cases) had moderate-to-high 10-year ASCVD risk. As shown in Table 2, the first to third hypoxic burden quartiles for patients with low 10-year ASCVD risk were 44.2 min/h, 73.8 min/h, and 123.4 min/h, respectively. In contrast, the corresponding quartiles for patients with moderate to high risk were 83.0 min/h, 176.1 min/h, and 373.2 min/h. For patients who exhibited negative changes on the ECG indicating myocardial ischemia, the first to third quartiles were 42.8 min/h, 69.8 min/h, and 129.4 min/h, compared to 95.3 min/h, 179.3 min/h, and 384.3 min/h for those with positive changes.

Table 1 Clinical Information of Patients

Table 2 The Quartile Distribution of Hypoxic Burden Among Patients with Moderate-to-High ASCVD Risk and Myocardial Ischemia

Comparison of Sleep Monitoring Characteristics in Patients with Different ASCVD Risk Levels

Kruskal–Wallis H-test indicated significant differences in sex, age, and medical history (Table 3). Independent samples t-test revealed no significant difference between the low 10-year ASCVD risk group and the moderate-to-high 10-year ASCVD risk group with the lowest blood oxygen saturation (Table 3). However, significant differences were observed in AHI, hypoxic burden, supine hypoxic burden, non-supine hypoxic burden, percentage of total sleep time with apnea, percentage of total sleep time with hypopnea, percentage of total sleep time with apnea and hypopnea, T90%, T85%, and T80% (Table 3).

Table 3 Comparison of Sleep Monitoring Parameters in OSA Patients with Low versus Moderate-to-High 10-year ASCVD Risk

Analysis of Baseline Clinical Data and Sleep Monitoring Measures Concerning Myocardial Ischemia Changes

As shown in Table 4, the Kruskal–Wallis H-test indicated that there were no significant differences in sex and age between patients with and without myocardial ischemia changes on ECG. The independent samples t-test showed no significant differences in body mass index, total cholesterol, high-density lipoprotein, or LDL between the two groups. In terms of sleep monitoring indicators, there were no significant differences between the two groups in the percentage of total sleep time with hypopneas, T85%, T80%, or lowest blood oxygen saturation. However, significant differences were found in AHI, hypoxic burden, supine hypoxic burden, non-supine hypoxic burden, percentage of total sleep time with apneas, percentage of total sleep time with apneas and hypopneas, and T90% between the two groups.

Table 4 Comparison of Clinical Data and Sleep Monitoring Indicators Between Patients with and without Myocardial Ischemia Changes on ECG

Analysis of Factors Influencing 10-Year ASCVD Risk and Myocardial Ischemia in OSA Patients

The binary logistic regression analysis was then performed to evaluate factors influencing 10-year ASCVD risk and myocardial ischemia in OSA patients. The factors included in the model were hypoxic burden grading, AHI grading, percentage of total sleep time with hypopnea, percentage of total sleep time with apnea, percentage of total sleep time with apnea and hypopnea, T90%, T85%, and T80%. Our results showed that hypoxic burden and T90% were significantly associated with the high 10-year ASCVD risk in OSA patients. For the index of hypoxic burden grade, the P-value was 0.024, while hypoxic burden between 75%-100% quartile showed statistically significant risk compared with patients in 0%-25% quartile (Q1), with an OR of 13.5 and a 95% CI of 1.684 to 109.051 (Table 5). For T90%, the P-value was 0.022, with an OR of 1.085 and a 95% CI of 1.012 to 1.164 (Table 5). Similarly, hypoxic burden grade and T90% were significantly associated with myocardial ischemia in patients with OSA. For the index of hypoxic burden grade, the P-value was 0.044. Both Q2, Q3, and Q4 exhibited statistically significant risks compared to patients in the 0%-25% quartile (Q1), with P-values of 0.011, 0.008, and 0.048, respectively. The OR was 3.241 (95% CI: 1.306–8.043), 4.497 (95% CI: 1.494–13.538), and 4.850 (95% CI: 1.017–23.139) for Q2, Q3, and Q4, respectively. For T90%, the P-value was 0.019, with an OR of 1.059 and a 95% CI of 1.009 to 1.110 (Table 6). Additionally, the power analysis showed that with the expected ORs of the hypoxic burden ranging from 3 to 11, our study provided a power greater than 90%. For T90%, even though the ORs were between 1.059 and 1.085, as a continuous variable, the analysis indicated that changes in this index across its range still resulted in sufficient power, estimated between 85% and 92%, confirming the reliability of our findings. These results revealed a significant association between T90% and hypoxic burden with high 10-year ASCVD risk and myocardial ischemia.

Table 5 Analysis of Risk Factors for High 10-Year ASCVD Risk in OSA Patients

Table 6 Analysis of Risk Factors for Myocardial Ischemia in OSA Patients

ROC Curve analysis of the Diagnostic Value of Hypoxic Burden for Moderate-to-High 10-Year ASCVD Risk and Myocardial Ischemia in OSA Patients

To determine the value of the hypoxic burden and T90% in diagnosing 10-year ASCVD risk and myocardial ischemia, the ROC curve was plotted. The results showed that the hypoxic burden and T90% (Figure 1) had diagnostic values for moderate-to-high 10-year ASCVD risk (Table 7). For hypoxic burden, the area under the curve was 0.747, with P<0.0001. The Youden’s index was used to determine the cut-off value of the hypoxic burden, which was 125.8 (%min/h). The corresponding sensitivity and specificity were 62.2% and 77.0%, respectively. For T90%, the AUC was 0.754. The cut-off value of T90% was 3.05%, with the corresponding sensitivity and specificity of 65.1% and 71.9%, respectively.

Table 7 Diagnostic Value of Hypoxic Burden and T90% for Moderate-to-High 10-Year ASCVD Risk in OSA Patients

Figure 1 ROC curve analysis of hypoxic burden and T90% in diagnosing moderate-to-high 10-year ASCVD risk in OSA patients.

Similarly, the hypoxic burden and T90% (Figure 2) also showed diagnostic values for myocardial ischemia (Table 8) in OSA patients. The AUC for the hypoxic burden and T90% was 0.769 and 0.741, respectively. The cut-off value for the hypoxic burden was 112.6 (%min/h), with a sensitivity of 70.6% and a specificity of 70.2%. For T90%, the cut-off value was 4.20%, the sensitivity was 60.6%, and the specificity was 77.5%.

Table 8 Diagnostic Value of Hypoxic Burden and T90% for Myocardial Ischemia in OSA Patients

Figure 2 ROC curve analysis of hypoxic burden and T90% in diagnosing myocardial ischemia in OSA patients.

Discussion

The results of this study revealed a strong association between hypoxic burden and T90% and both moderate-to-high 10-year ASCVD risk and myocardial ischemia in OSA patients. These indicators, reflecting the intensity and duration of hypoxic events during sleep, provided valuable insights into the severity of OSA and its cardiovascular implications. In fact, the cutoff values derived from this study are highly consistent with clinically observed OSA severity and have significant implications for evaluating patients with a high AHI, short-duration events, and low overall nocturnal hypoxia. These metrics provide essential clinical guidance in determining whether patients require more aggressive treatments, such as continuous positive airway pressure therapy. Rather than relying solely on AHI and minimum oxygen saturation for severity assessment, clinicians are more inclined to consider measures of hypoxia duration and intensity. The hypoxic burden index and T90% more accurately meet the requirements of clinicians. Notably, the traditional evaluation indicators, such as AHI and the lowest blood oxygen saturation, did not have such associations, highlighting the limitations of current OSA severity assessment indicators. The higher hypoxic burden and T90% underscore the risk of intensity and duration of hypoxic events in 10-year ASCVD risk and myocardial ischemia for OSA patients. Additionally, the findings emphasize the importance of incorporating these advanced indicators into routine clinical practice for a more comprehensive evaluation of cardiovascular risk in OSA patients. These indicators can help clinicians identify high-risk patients early and develop tailored treatment strategies.

The association of cardiovascular risk with hypoxic events in OSA patients is receiving increasing attention. Previous studies^1,2,10,11,19 have indicated that recurrent hypoxia in OSA is related to pulmonary hypertension and vascular and microvascular diseases. The oxygen desaturation index, a critical metric that measures hypoxic burden, links OSA to cardiovascular conditions in the context of hypoxia. It is correlated with the body roundness index, a novel body metric used to define and predict cardiovascular risk in patients with OSA.²⁰ However, the relationship between hypoxic events and patient prognosis still requires further investigation. The hypoxic burden, which reflects the intensity and duration of hypoxic events, can be automatically calculated by advanced sleep monitoring data analysis software. The T90% reflects the total time spent with blood oxygen saturation below 90% throughout the night, providing insight into the overall severity of OSA during the night.²² This study indicates that the hypoxic burden and T90% were significantly related to both the 10-year ASCVD risk and myocardial ischemia. As the increase of the hypoxic burden, the risk of moderate-to-high 10-year ASCVD and myocardial ischemia also elevated. Although previous studies have investigated the relationship between T90% and hypoxic burden-related indicators and cardiovascular disease, there is currently a lack of cut-off values for these indicators for risk prediction, leading to an inability to classify the severity of patients and resulting in different research conclusions.^26,27 This study established cut-off values for T90% and hypoxic burden in predicting the risk of myocardial ischemia and cardiovascular disease. These results may provide a basis for defining the severity and offer important guidance for the clinical treatment of myocardial ischemia and cardiovascular complications in patients with OSA.

PSG is limited in clinical use due to its high cost and complex operation. Portable sleep monitors, however, show high consistency with PSG in diagnosing OSA.²³ This study utilized portable sleep monitoring and included 311 OSA patients of different ages and genders from southern China. Given that the portable sleep monitor we used in this study is validated for accuracy and reliability against traditional PSG, it enables clinicians to perform comprehensive assessments in a variety of settings, including home care. This flexibility allows for the timely identification of at-risk patients and could facilitate earlier interventions, which are critical for aggressive management strategies in OSA. Furthermore, the China-PAR model, which is tailored for the Chinese population, was used to assess the 10-year ASCVD risk, allowing for an evaluation of long-term cardiovascular risk. Additionally, ECG was used to assess current myocardial ischemia, providing a comprehensive assessment of both long-term cardiovascular risk and current myocardial ischemia risk in OSA patients.

This study has some limitations. For example, the exclusion of patients with electrolyte disturbances and kidney insufficiency might have underestimated the impact of oxygen desaturation on myocardial ischemia and cardiovascular risk assessment. Another limitation of this study is the modest AUC values and small ORs of the hypoxic burden and T90%, which may limit their clinical utility. Further research is warranted to confirm our findings and explore the applicability of these indices in diverse clinical settings.

Conclusion

This study demonstrates that hypoxic burden and T90% are valuable indicators for assessing cardiovascular risk and myocardial ischemia in OSA patients. These metrics have diagnostic potential for identifying myocardial ischemia and cardiovascular diseases, with implications for clinical practice. While these findings highlight the diagnostic potential of nocturnal hypoxia monitoring for cardiovascular complications, our study is single-center research; further validation in larger, prospective cohorts is necessary before widespread clinical adoption. Additionally, the impact of hypoxia on both short-term and long-term cardiovascular risk events in OSA patients warrants further research and attention.

Abbreviations

OSA, Obstructive sleep apnea; ASCVD, Atherosclerotic cardiovascular disease; PSG, Polysomnography; ECG, Electrocardiogram.

Data Sharing Statement

The datasets generated and/or analyzed during the current study are available from the corresponding author on reasonable request.

Author Contributions

Wenmei Zeng: Conceptualization, Methodology, Resources, Software, Funding acquisition, Writing – original draft, Writing – review & editing. Sulong Wu: Conceptualization, Methodology, Resources, Writing – original draft, Writing – review & editing. Zhuofan Liu: Conceptualization, Investigation, Resources, Writing – review & editing. Long Yuan: Conceptualization, Investigation, Resources, Writing – review & editing. Bilin Chen: Software, Methodology, Resources, Writing – review & editing. Yan Rong: Conceptualization, Funding acquisition, Formal analysis, Data curation, Writing – review & editing, Project administration. All authors gave final approval of the version to be published; have agreed on the journal to which the article has been submitted; and agree to be accountable for all aspects of the work.

Funding

This work was supported by the Technical R & D and Creative Design Project sub-funds, Nanshan, Shenzhen, China [grant numbers NS2021002; NS2022013].

Disclosure

The authors report no conflicts of interest in this work.

References

1. Labarca G, Vena D, Hu WH, et al. Sleep apnea physiological burdens and cardiovascular morbidity and mortality. Am J Resp Crit Care. 2023;208(7):802–813. doi:10.1164/rccm.202209-1808OC

2. Driendl S, Stadler S, Arzt M, Zeman F, Heid IM, Baumert M. Nocturnal hypoxemic burden and micro- and macrovascular disease in patients with type 2 diabetes. Cardiovasc Diabetol. 2024;23(1):195. doi:10.1186/s12933-024-02289-w

3. Yoon CW, Park HK, Bae EK, Rha JH. Sleep apnea and early neurological deterioration in acute ischemic stroke. J Stroke Cerebrovasc Dis. 2020;29(2):104510. doi:10.1016/j.jstrokecerebrovasdis.2019.104510

4. Ferini-Strambi L, Fantini ML, Castronovo C. Epidemiology of obstructive sleep apnea syndrome. Minerva med. 2004;95(3):187–202.

5. Olaithe M, Pushpanathan M, Hillman D, et al. Cognitive profiles in obstructive sleep apnea: a cluster analysis in sleep clinic and community samples. J Clin Sleep Med. 2020;16(9):1493–1505. doi:10.5664/jcsm.8564

6. Huang SLQ, Li Q-Y. Prevalence of obstructive sleep apnea-hypopnea syndrome in Chinese adults aged over 30 yr in Shanghai. Zhonghua Jie He He Hu Xi Za Zhi. 2003;26(5):268–272.

7. Takagi T, Nakamura K, Hashimoto H, et al. The impact of sleep apnea on right atrial structural remodeling with atrial fibrillation. J Cardiol. 2020;75(6):665–672. doi:10.1016/j.jjcc.2019.12.012

8. Hui M, Li Y, Ye J, Zhuang Z, Wang W. Obstructive sleep apnea-hypopnea syndrome (OSAHS) comorbid with diabetes rather than OSAHS alone serv es an independent risk factor for chronic kidney disease (CKD). Ann Palliat Med. 2020;9(3):858–869. doi:10.21037/apm.2020.03.21

9. Barreto PR, Diniz DLO, Lopes JP, et al. Obstructive sleep apnea and wake-up stroke – a 12 months prospective longitudinal study. J Stroke Cerebrovasc Dis. 2020;29(5):104564. doi:10.1016/j.jstrokecerebrovasdis.2019.104564

10. Testelmans D, Spruit MA, Vrijsen B, et al. Comorbidity clusters in patients with moderate-to-severe OSA. Sleep Breath. 2022;26(1):195–204. doi:10.1007/s11325-021-02390-4

11. Thanaviratananich S, Cheng H, Chirakalwasan N, Reutrakul S. Association between nocturnal hypoxemic burden and glucose metabolism. Sleep Breath. 2022;26(3):1465–1470. doi:10.1007/s11325-021-02464-3

12. Hui XX, Xiao Y. Annual review of sleep-disordered breathing in 2023. Zhonghua Jie He He Hu Xi Za Zhi. 2024;47(1):59–63. doi:10.3760/cma.j.cn112147-20231031-00280

13. Brooker EJ, Landry SA, Thomson LDJ, et al. Obstructive sleep apnea is a distinct physiological endotype in individuals with comorbid insomnia and sleep apnea. Ann Am Thorac Soc. 2023;20(10):1508–1515. doi:10.1513/AnnalsATS.202304-350OC

14. Simpson L, Hillman DR, Cooper MN, et al. High prevalence of undiagnosed obstructive sleep apnoea in the general population and methods for screening for representative controls. Sleep Breath. 2013;17(3):967–973. doi:10.1007/s11325-012-0785-0

15. Yang X, Li J, Hu D, et al. Predicting the 10-year risks of atherosclerotic cardiovascular disease in Chinese population: the China-PAR project (prediction for ASCVD risk in China). Circulation. 2016;134(19):1430–1440. doi:10.1161/CIRCULATIONAHA.116.022367

16. Roffe C, Sills S, Halim M, et al. Unexpected nocturnal hypoxia in patients with acute stroke. Stroke. 2003;34(11):2641–2645. doi:10.1161/01.STR.0000095188.65567.4F

17. Baguet JP, Minville C, Tamisier R, et al. Increased aortic root size is associated with nocturnal hypoxia and diastolic blood pressure in obstructive sleep apnea. Sleep. 2011;34(11):1605–1607. doi:10.5665/sleep.1406

18. Goody PR, Nachtsheim L, Hosen MR, et al. Analysis of nocturnal, hypoxia-induced miRNAs in sleep apnea patients. PLoS One. 2022;17(3):e0263747. doi:10.1371/journal.pone.0263747

19. Huang Z, Duan A, Zhao Z, et al. Sleep-disordered breathing patterns and prognosis in pulmonary arterial hypertension: a cluster analysis of nocturnal cardiorespiratory signals. Sleep Med. 2024;113:61–69. doi:10.1016/j.sleep.2023.11.016

20. Yeşildağ M, Şentürk Z, Bekci TT, Guney İ. The usefulness of new body indices in determining the risk of cardiovascular disease in cases with obstructive sleep apnea syndrome. Int J Gen Med. 2024;17:5523–5534. doi:10.2147/IJGM.S489904

21. Azarbarzin A, Sands SA, White DP, Redline S, Wellman A. The hypoxic burden: a novel sleep apnoea severity metric and a predictor of cardiovascular mortality-reply to ‘the hypoxic burden: also known as the desaturation severity parameter’. Eur Heart J. 2019;40(35):2994–2995. doi:10.1093/eurheartj/ehz273

22. Henríquez-Beltrán M, Dreyse J, Jorquera J, et al. Is the time below 90% of SpO2 during sleep (T90%) a metric of good health? A longitudinal analysis of two cohorts. Sleep Breath. 2024;28(1):281–289. doi:10.1007/s11325-023-02909-x

23. Kuna ST. Portable-monitor testing: an alternative strategy for managing patients with obstructive sleep apnea. Respir Care. 2010;55(9):1196–1215.

24. Canadian Agency for D. Technologies in H. Portable monitoring devices for diagnosis of obstructive sleep apnea at home: review of accuracy, cos t-effectiveness, guidelines, and coverage in Canada. CADTH Technol Overv. 2010;1(4):e0123.

25. Sateia MJ. International classification of sleep disorders-third edition: highlights and modifications. Chest. 2014;146(5):1387–1394. doi:10.1378/chest.14-0970

26. Kendzerska T, Gershon AS, Hawker G, Leung RS, Tomlinson G. Obstructive sleep apnea and risk of cardiovascular events and all-cause mortality: a decade-long historical cohort study. PLoS Med. 2014;11(2):e1001599. doi:10.1371/journal.pmed.1001599

27. Cao W, Luo J, Huang R, Xiao Y. The association between sleep breathing impairment index and cardiovascular risk in male patients with obstructive sleep apnea. Nat Sci Sleep. 2022;14:53–60. doi:10.2147/NSS.S343661