A Retrospective Review of Antimicrobial Usage and Indications at End-of-Life for Oncology Patients in a Tertiary Cancer Center in Saudi Arabia

Summary

Objectives:

To investigate antimicrobial usage patterns, indications, and outcomes in end-of-life (EOL) oncology patients at Princess Noorah Oncology Center (PNOC), King Abdulaziz Medical City, Jeddah, Saudi Arabia. The use of antimicrobials in oncology patients in EOL care has raised concerns about their appropriateness and the increasing issue of antimicrobial resistance.

Methods:

The study is a retrospective chart review of cancer patients treated at the PNOC from January 1, 2017, to January 31, 2022. The inclusion criteria included cancer patients ≥18 years who died during their admission. The exclusion criteria included subjects who did not meet the inclusion criteria.

Results:

Among the 503 patients analyzed, 89.7% received antimicrobial treatment. In 66.5% of cases, the antimicrobial prescriptions were justified by positive culture results, whereas 33.5% were administered without culture confirmation. The most used antimicrobials were Piperacillin-Tazobactam (20.2%) and Vancomycin (11.4%). Antimicrobial use was significantly associated with patient awareness of their diagnosis (p = 0.036) and metastatic status (p = 0.027). The median duration of antimicrobial administration was 7.0 days.

Conclusions:

The study highlights the prevalent use of antimicrobials for oncology patients in EOL care, with 89.7% taking antimicrobials. The findings emphasize the need for strong antimicrobial stewardship programs to optimize antimicrobial prescription, aiming to decrease antimicrobial resistance.

Introduction

Antimicrobial administration in end-of-life (EOL) care for oncology patients has emerged as a topic under debate within the medical community. Infection treatment through antimicrobial usage is an essential aspect of palliative and supportive care to alleviate various symptoms that may arise due to these infections. However, concern regarding the appropriateness and effectiveness of antimicrobial use in EOL settings is increasing, particularly due to the notable increase in antimicrobial resistance and its potential effects on the community [1].

Oncology patients frequently develop infections due to the tumor and antitumor treatment they receive, consequently compromising the patient's immune systems [2]. Antimicrobial prescriptions to manage these infections and maintain patient comfort considerably impact the their quality of life (QOL). However, aggressive usage of antimicrobials in EOL care patients could be harmful and may not always serve the best interests of the patients [3].

As the worldwide prevalence of drug-resistant infections continues to increase, antimicrobial resistance is a global concern. In addition, the side effects of antimicrobial treatment may adversely affect patient comfort and QOL. Considering these concerns, gaining a deep understanding of the current practice regarding antimicrobial usage patterns and outcomes in oncology patients in EOL care is crucial to optimizing patient care and developing clear future clinical decision-making guidelines [4]. Multiple studies have explored antimicrobial use for oncology patients in EOL care across various countries, and investigated the effects of different factors, including antimicrobial usage prevalence, common infection types, and appropriateness of prescribing antimicrobials [3,5]. However, the antimicrobial prescription practice in Saudi Arabia needs to be further investigated, and its effect on our patient population must be explored. A clear understanding of these patterns is crucial to enforce the clinical practice and improve the QOL for this specific patient group. Furthermore, sufficient data on local antimicrobial usage for the aforementioned oncology patients will help in developing a clear population- and practice-oriented antimicrobial stewardship program; such a program will considerably contribute to addressing the worldwide challenge of antimicrobial resistance [6]. Therefore, our study conducts a retrospective analysis of the use of antimicrobials and their indications for oncology patients in EOL care.

Methods

A retrospective study (chart review) was conducted from January 1, 2017, to January 31, 2022, at the Princess Noorah Oncology Center (PNOC) at King Abdulaziz Medical City, Jeddah (KAMC-JD). The PNOC is an oncology center with a total capacity of 125 oncology inpatient beds, which includes a bone marrow transplant unit.

The inclusion criteria were as follows: cancer patients aged 18 and above who died during their admission from January 1, 2017, to January 31, 2022. The exclusion criteria included subjects who did not meet the inclusion criteria.

The information was gathered using a collection form targeting oncology patients at KAMC-JD. The collection form included demographic information, reason for admission, length of stay, day of death, use of antimicrobials, type and route of antimicrobials used, indication of use, presence of foreign body at the site, type of infection/organism, presumed or confirmed infection, and clinical pharmacology input. Data were obtained from the patients' electronic health records. The collected data were filtered according to inclusion and exclusion criteria, organized in a database, and categorized into a data form.

A high white blood count (WBC) is defined as the presence of >11,000 white blood cells in a microliter of blood [7].

A high erythrocyte sedimentation rate (ESR) was defined as follows by age and gender:

• Men < 50 years old: more than 15 mm/h

• Men > 50 years old: more than 20 mm/h

• Women < 50 years old: more than 20 mm/h

• Women > 50 years old: more than 30 mm/h [8].

High C-reactive protein (CRP) was defined as more than 10 mg/L [9].

High procalcitonin was defined as more than 0.1 μg/L [9].

Bacteremia is defined as the existence of live bacteria in the bloodstream [10].

Infection is defined as the infiltration and growth of microorganisms that are not typically found inside the body, such as bacteria, viruses, fungi, or parasites. An infection can be subclinical and symptomless, or it can be clinically evident and symptomatic. Infections can either stay local or become systemic if they spread through the blood [11].

Support care in PNOC was defined as patient care focused on physical and symptom control. Patients and their families receive support on multiple levels, including social, spiritual, and psychological levels, as the patient might have a “do not resuscitate order” and may receive only palliative procedures, which may include palliative chemotherapy or radiation [Determination of goals of care. Administrative policy and procedures at KAMC-JD].

Comfort care in PNOC was defined as patient care focused on symptom and pain control and QOL as the patient goes through EOL care [Determination of goals of care. Administrative policy and procedures at KAMC-JD].

The Institutional Review Board at King Abdullah International Medical Research Center approved the study (NRJ22J/167/06). Because the study was retrospective, informed consent was not requested.

Statistical analysis

Analysis of the data was carried out through R Studio (R version 4.2.2). We used percentages and frequencies as well as the interquartile range (IQR) and the median to present categorical and continuous data, respectively. The prevalence of antimicrobial prescription was explored using a one-sample proportions test with continuity correction, and the result was articulated as the proportion and 95% confidence intervals (95% CIs). Factors associated with prescribing antimicrobial agents were assessed depending on the variable type. We employed the Wilcoxon rank sum test for numerical variables, and either Pearson's chi-squared test or Fisher's exact test for the categorical variables whenever applicable. A statistical significance p-value should be <0.05.

Results

In the beginning, data were collected from 510 patients. We excluded 7 patients who were younger than 18. Therefore, the current study included data from 503 patients. More than half were males (51.9%), and most were Saudis (97.0%) and married (85.5%). The median (IQR) age of the patients was 64.0 years (52.0, 75.0). All patients were admitted to PNOC, and 55.3% had comfort care as the target care regimen. Fewer than half of the patients were aware of their full prognoses (41.2%). Additional demographic information is provided in Table 1.

Regarding the clinical data, most patients had solid malignancies (87.9%) with evidence of metastasis in 83.7% (Table 1). The most common diagnoses were colon cancer (9.2%), breast cancer (8.8%), and hepatocellular cancer (7.8%). The most frequently received treatments were chemotherapy (59.8%) and surgery (42.0%).

Records of 367 infections were available; among these, urinary tract infections and pneumonia represented 19.9% and 18.8%, respectively. Records of the laboratory markers of infection and inflammation were available for 251 patients, including WBCs (39.0%) and CRP (19.9%; Table 2).

A total of 451 patients received antimicrobial agents, which represents 89.7% of the sample (95% CI, 86.6–92.1, Fig. 1). The prescription of antimicrobial drugs was significantly associated with awareness of the prognosis (93.2% versus 87.2% non-aware patients, p = 0.036) and having metastatic cancer (90.3% versus 81.4% patients with non-metastatic cancer, p = 0.027). Prescribing antimicrobials was justified by a positive culture in 300 out of 451 patients, representing 66.5% of those who received antimicrobial treatment. A total of 12 patients with a positive culture did not receive any treatment. Meanwhile, 151 patients with no positive culture received antimicrobial treatment, representing 33.5% of the total number who were prescribed antimicrobials. Notably, antimicrobial prescription was justified among 96.2% of the patients with positive cultures, which was a significantly higher proportion than that of those with a negative or non-available culture result (79.1%, p<0.001, Table 3).

Among the 451 patients who received antimicrobial drugs, 411 (91.1%), 127 (28.2%), and 44 (9.8%) received antibiotics, antifungal, and antiviral drugs, respectively. Focusing on combined therapies, 22.6%, 3.5%, and 2.9% received antibiotics and antifungal medications, antibiotics and antiviral drugs, and antiviral and antifungal drugs, respectively.

A total of 1,318 antimicrobial drugs were prescribed to the patients. The most common drugs were Piperacillin and Tazobactam therapy (20.2%), Vancomycin (11.4%), and Meropenem (8.9%). The median (IQR) duration of antimicrobial administration was 7.0 days (4.0–12.0). Antimicrobial prescribing was guided by culture results for 52.8% of the antimicrobials used, while 43.9% were prescribed without microbiological confirmation, and 2.9% of the medications were given for prophylactic indications.

Of the 411 patients who received antibiotics, the percentages who received one or 2 types of antibiotics were 24.8% and 24.6%, respectively (Fig. 2A). The percentages of patients who received one or 2 antibiotic courses were 27.3% and 29.3%, respectively (Fig. 2B). Most antibiotics were given intravenously (93.7%), and oral antibiotics were administered to 27.5% of the patients. Antibiotics were discontinued four days before death among 50.6% of the patients. The prescription of antibiotics was based on a positive culture in 280 patients, representing 68.1% of those who received antibiotics (Table 4). The most commonly isolated organisms detected in the positive cultures were Escherichia coli (33.6%), Candida species (25.7%), and Pseudomonas aeruginosa (11.4%).

Discussion

The use of antimicrobials in palliative care for oncology patients has sparked considerable debate, primarily due to concerns regarding its appropriateness and effectiveness, the increasing issue of antimicrobial resistance epidemic, and its impact on the patient's QOL [4,12].

To gain insight into regional practices and contribute to the global conversation on this crucial subject, we investigated antimicrobial usage patterns, indications, and outcomes in oncology patients undergoing EOL care at PNOC.

Our findings suggest a high frequency of antimicrobial usage among EOL oncology patients; 89.7% of the sample received antibiotics. This finding aligns with several research findings published in the United States and Europe, which reported that 70% to 90% of terminal oncology patients received antibiotics [13,14,15]. However, determining the appropriateness of these prescriptions is critical. In 66.5% of the patients, a positive culture justified the course of therapy, while 33.5% received antimicrobials without culture confirmation. This outcome corresponds Kim JH et al.'s (2025) findings, who reported that in patients with advanced cancer, 56% of antimicrobial prescriptions were not supported by microbiological evidence [16]. Additionally, 27% of the patients in our study had no culture done, which aligns with previous research that shows 7–56% of advanced cancer patients were treated without microbiological confirmation due to rapid decline, unspecific symptoms, and low yield culture due to difficulty in obtaining appropriate samples [3,17,18].

The clinical and demographic traits of our study population revealed a predominance of male patients (51.9%), a high prevalence of metastatic cancers (83.7%), and common diagnoses such as colon, breast, and hepatocellular cancers. These characteristics highlight the complexity of treating infections in patients with advanced malignancies and represent the larger population of oncology patients [19]. The most utilized medications were Piperacillin-Tazobactam, Vancomycin, and Meropenem, which is consistent with findings from earlier studies that highlighted the widespread usage of these agents in oncology settings [20,21]. Based on a study by Bantikassegn A. et al. (2024), broad-spectrum antibiotics such as Piperacillin combined with Tazobactam were noted as one of the most used prescribed antimicrobials for EOL patients [21].

A key finding in our study is the association between antimicrobial prescriptions and patient awareness of their full prognosis and metastatic status. A previous systematic review and meta-analysis reported that patients with comprehensive knowledge about their cancer diagnosis and prognosis tend to demonstrate better social functioning and overall QOL. This can influence the approach clinicians take in prescribing antimicrobials, with the goal of more aggressive management in well-informed patients [22]. Nevertheless, careful consideration of the possible harms and benefits of this approach is necessary, especially in the context of palliative care [23].

This study further emphasizes the widespread usage of antimicrobials, both prophylactically and empirically, with therapeutic indications being the most prevalent application. Although immunocompromised patients urgently require empirical treatment, the substantial amount of prophylactic usage raises concerns regarding its necessity and efficacy in palliatively improving patient outcomes [24]. Fairweather et al established that limited evidence exists to support the efficacy of frequent antimicrobial usage in palliative care settings in improving patient outcomes. The review noted that although prophylactic antimicrobial use is common, compelling data to support its frequent implementation in palliative care is seldom presented [24]. These findings highlight the necessity of effective antimicrobial stewardship programs to maximize antimicrobial usage, reduce redundant prescriptions, and minimize the risk of antimicrobial resistance [25].

In contrast to other research, the current study provides a focused investigation of a specific population in Saudi Arabia, thus providing insightful local data that further enhances the global understanding of antimicrobial usage in EOL care. Our findings highlight the potential to improve the appropriateness of antimicrobial prescribing practices, thereby aligning with worldwide trends. Enhancing the quality of care for EOL oncology patients, as well as the antimicrobial resistance epidemic, can be achieved by the implementation of targeted interventions, such as guidelines for antimicrobial usage in EOL care and education programs for healthcare providers.

Regarding limitations, the current study employed a retrospective design, thus rendering it more susceptible to multiple biases, such as recall or classification biases. In addition, the findings are based on data gathered from a single tertiary care facility. Future research should focus on overcoming these limitations and investigating optimal strategies for antimicrobial use in palliative care, as well as evaluating the impact of these strategies on patient outcomes and QOL.

In conclusion, the study highlights the prevalent use of antimicrobials for oncology patients in EOL care, with 89.7% taking antimicrobials. Concerns have been raised regarding the appropriateness of these treatments, as a significant percentage of prescriptions (33.5%) were filled without microbial evidence. The findings note that patients who are fully aware of their prognosis and those with metastatic cancer are more likely to be prescribed antimicrobials, which identifies a trend toward aggressive infection management in this population. The study findings emphasize the need for strong antimicrobial stewardship programs to optimize antimicrobial prescription, aiming to decrease antimicrobial resistance.

Acknowledgment

The authors would like to thank Editage (www.editage.com) for English language editing.

Disclosure

Authors have no conflict of interests, and the work was not supported or funded by any drug company.