Antibiotics are often administered to terminally ill patients until death, and antibiotic use contributes to the emergence of multidrug-resistant organisms (MDROs). We investigated antibiotic use and the isolation of MDROs among patients who died in general medical wards.
All adult patients who died in the general internal medicine wards at four acute care hospitals between January and June 2013 were enrolled. For comparison with these deceased patients, the same number of surviving, discharged patients was selected from the same divisions of internal medicine subspecialties during the same period.
During the study period, 303 deceased patients were enrolled; among them, 265 (87.5%) had do-not-resuscitate (DNR) orders in their medical records. Antibiotic use was more common in patients who died than in those who survived (87.5% vs. 65.7%, P<0.001). Among deceased patients with DNR orders, antibiotic use was continued in 59.6% of patients after obtaining their DNR orders. Deceased patients received more antibiotic therapy courses (two [interquartile range (IQR) 1–3] vs. one [IQR 0–2], P<0.001). Antibiotics were used for longer durations in deceased patients than in surviving patients (13 [IQR 5–23] vs. seven days [IQR 0–18], P<0.001). MDROs were also more common in deceased patients than in surviving patients (25.7% vs. 10.6%, P<0.001).
Patients who died in the general medical wards of acute care hospitals were exposed to more antibiotics than patients who survived. In particular, antibiotic prescription was common even after obtaining DNR orders in patients who died. The isolation of MDROs during the hospital stay was more common in these patients who died. Strategies for judicious antibiotic use and appropriate infection control should be applied to these patient populations.
Citation: Kwak YG, Moon C, Kim ES, Kim B-N (2016) Frequent Prescription of Antibiotics and High Burden of Antibiotic Resistance among Deceased Patients in General Medical Wards of Acute Care Hospitals in Korea. PLoS ONE 11(1): e0146852. doi:10.1371/journal.pone.0146852
Editor: Patrick M. Schlievert, University of Iowa Carver College of Medicine, UNITED STATES
Received: August 10, 2015; Accepted: December 21, 2015; Published: January 13, 2016
Copyright: © 2016 Kwak et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: Due to ethical considerations, an anonymized dataset is available upon request from the study's first author Yee Gyung Kwak (email@example.com).
Funding: These authors have no support or funding to report.
Competing interests: The authors have declared that no competing interests exist.
Antibiotic use contributes to increased antibiotic resistance due to the selection and expression of antibiotic resistance genes in bacterial populations . Antibiotic abuse can result in antibiotic resistance in individual patients. Antibiotic resistance has been associated with increased attributable length of hospital stay, mortality, and health care costs .
Physicians and family members often consider antibiotics to be a minimum treatment requirement for terminally ill patients, and they are often administered until death in those patients [3–5]. As in other Asian countries where palliative and hospice care services are not widely utilized for patients with terminal illnesses [6,7], most terminally ill patients in Korea are admitted to acute care hospitals near the end of their lives in order to receive life-sustaining treatments .
A recent investigation of intensive care units (ICUs) reported that dying patients without withdraw orders received more antibiotics and developed more multidrug-resistant organisms (MDROs) . Patients who acquire MDROs before death may serve as MDRO reservoirs, transmitting the organisms to surviving patients in the hospital setting [8,9].
While there have been many studies on antibiotic use and resistance in critically ill patients in ICUs and in patients who die in palliative and hospice care settings [3–5,10–12], scant attention has been paid to antibiotic use in patients who die in the general hospital wards. We hypothesized that a certain proportion of patients who died in the general medical wards of acute care hospitals might be exposed to antibiotics before death, and that the isolation of MDROs might be common in these patients during their hospital stay. The aim of this study was to examine the antibiotic use and isolation of MDROs among patients who died in the general medical wards of acute care hospitals and to compare these characteristics with those of surviving, discharged patients.
Materials and Methods
This retrospective study was conducted at four university-affiliated acute care hospitals in Korea. The study protocol was approved by the Institutional Review Board (IRB) of the Inje University Sanggye-Paik Hospital (SPIRB 13–037) as the central IRB for this multicenter study, which waived the need for written or oral informed consent from the participants.
All patients ≥18 years of age who died in the internal medicine wards between January and June 2013 were enrolled. For comparison with these deceased patients, we also enrolled the same number of surviving patients discharged from the same divisions of the internal medicine subspecialties during the same study period with differences in length of hospital stay ≤5 days. Patients who were hospitalized ≤2 days or >60 days, were transferred to other hospitals, or were discharged against medical advice were excluded.
Medical records were reviewed retrospectively, and data were collected using standardized case report forms. The data obtained included demographic characteristics, length of hospital stay, underlying diseases or conditions, and either clinical or microbiological bacterial infection during the admission and at the time of death or discharge. Data on antibiotic exposure for more than 24 hours and isolation of MDROs during the admission were also collected. An antibiotic therapy course was defined as the use of different antibiotics for at least one day during admission. Days of antibiotic use were defined as the total number of days that a single antibiotic (parenteral or oral) was administered, regardless of dosage . MDROs in this study included methicillin-resistant Staphylococcus aureus, vancomycin-resistant enterococci, third-generation cephalosporin-resistant Enterobacteriaceae, and carbapenem-resistant non-fermenters, including Acinetobacter baumannii and Pseudomonas aeruginosa . In addition, a do-not-resuscitate (DNR) order was defined as present if the patient had a preprinted DNR form or recorded verbal communication in their medical record .
Bivariable analysis was performed separately for each variable. P values were calculated using chi-square or Fisher’s exact tests for categorical variables. Continuous variables were compared using Student’s t (for normally distributed variables) or Mann—Whitney U tests (for non-normally distributed variables). P values <0.05 were considered statistically significant. All variables that were statistically significant in bivariable analyses were included in logistic regression analysis. Results from the multivariable analysis are presented as adjusted odds ratios (AORs) with 95% confidence intervals (CIs).
During the study period, a total of 15,777 patients were discharged from the general internal medicine wards of four hospitals, while 377 died in the hospital. Of these deceased patients, 74 were excluded, and the remaining 303 comprised the study population (Table 1). The same number of surviving patients was selected from among the 15,400 surviving patients who were discharged. Overall, the most common underlying disease was solid tumor (75.6% vs. 58.7%, P<0.001), followed by cardiovascular disease (42.9% vs. 34.7%, P = 0.037) and diabetes mellitus (27.4% vs. 24.8%).
Of the 303 deceased patients, 265 (87.5%) had DNR orders in their medical records (Table 2). On average, the first record of DNR order occurred on the 12th day (median, interquartile range [IQR] 4–21) of a 14-day hospital stay (median, IQR 8–26). The median time from DNR order to death was two (IQR 1–6) days, and 53 patients (19.6%) had DNR orders on the day of their death. Among deceased patients with DNR orders, 59.6% continued to receive antibiotics after providing a DNR order. Antibiotics were administered for a median of two (IQR 0–5) days after the day when DNR consent was provided in those patients. Comparison of deceased patients with DNR orders to those without orders revealed no significant differences in the proportion of patients that received antibiotic therapy (87.9% vs. 84.2%, P = 0.599) or the number of days of antibiotic use (median 13 [IQR 6–22] vs. 14 [IQR 4–23] days). Malignancy was the most common cause of death (62.4%), and 21 patients (6.9%) received cardiopulmonary resuscitation during admission.
As shown in Table 3, bacterial infections were more common in patients who died than in those who survived (57.8% vs. 40.6%, P<0.001). Antibiotic therapy lasting more than 24 hours was also more common in deceased than in surviving patients (87.5% vs. 65.7%, P<0.001). The median number of antibiotic therapy courses were two (IQR 1–3) and one (IQR 0–2) for deceased and surviving patients, respectively. Third-generation cephalosporins were the most commonly prescribed antibiotics in both patient groups (43.9% and 39.3%). However, antibiotics were used for significantly longer periods in deceased patients than in those who survived (median 13 [IQR 5–23] vs. 7 [IQR 0–18] days, P<0.001). MDROs were more commonly isolated in deceased patients than in surviving patients (25.7% vs. 10.6%, P<0.001).
Comparison of deceased patients who had MDROs during their admissions to those who did not (Table 4) revealed that bacterial infections were more common in deceased patients with MDROs than in those without MDROs (89.7% vs. 46.7%, P<0.001). The proportion of patients receiving antibiotic therapy was higher in deceased patients with MDROs than in those without MDROs (98.7% vs. 83.6%, P<0.001). The number of antibiotic therapy courses administered was also higher (median 3 vs. 2 courses, P<0.001), and antibiotics were used for longer periods of time in deceased patients with MDROs than in those without MDROs (median 22 vs. 10 days, P<0.001). DNR orders were less common in deceased patients with MDROs (79.5% vs. 90.2%, P = 0.014). In multivariable analyses, factors including bacterial infection during admission (AOR 6.19, 95% CI 2.67–14.39), biliary drainage tube (AOR 3.50, 95% CI 1.02–12.01), nasogastric or gastrostomy tube (AOR 3.08, 95% CI 1.56–6.09), and longer duration of antibiotic use (AOR 1.04, 95% CI 1.02–1.06) were associated with deceased patients who had MDROs during their admissions.
This study revealed extensive antibiotic exposure before death among deceased patients, even those who had provided DNR orders, in the general medical wards of acute care hospitals. Isolation of MDROs was more common in patients who died than in those who survived. Longer antibiotic use duration was associated with dying patients who had MDROs during admission. These findings support our hypothesis but a causality between the antibiotic exposure and isolation of MDROs in the deceased patients could not be defined because of limitations of this study (refer to below).
After opioid analgesics, antibiotics are the most commonly administered drugs on the day of death in terminally ill patients, and many of these patients receive antibiotics until death in Korea . In this study, about 90% of patients who died in the general internal medicine wards received at least one course of antibiotic therapy during their hospital stay. Given the proportion of patients with bacterial infection who received antibiotic therapy during admission, antibiotics might be abused at some point during admission in about 30% of deceased patients. It has not been determined whether the use of antibiotics for end-of-life treatment is appropriate and beneficial in terms of improving symptoms in terminally ill patients [7,12]. Furthermore, aggressive treatment of infection has not been found to alter the underlying disease process or improve survival rates among patients with terminal illness . Considering the risk of antibiotic resistance and the paucity of novel antibiotics in the development pipeline, the use of antibiotics in terminally ill patients should be rationalized .
Indiscriminate use of antibiotics in terminally ill patients may cause unintended consequences, including individual patient burdens as well as burdens extending to other patients in the same healthcare setting through the possible selection for antibiotic resistance . Although the deceased patients with MDROs had more bacterial infections during admission in this study, multivariable analyses indicated that longer duration of antibiotic use was associated with the isolation of MDROs in these patients. This result is consistent with previous findings that antibiotic courses of longer duration are associated with higher rates of resistance . MDROs that colonize or infect deceased patients can contaminate the environment or the hands of healthcare workers and spread to surviving patients or vice versa [9,11]. Therefore, careful selection of indications for and duration of antibiotic use at the end of life is important to prevent the development and dissemination of antibiotic resistance.
In Asia, physicians are typically more aggressive in their end-of-life treatment compared with their Western European counterparts due to cultural differences . They are less likely to withhold or withdraw life-sustaining treatment, and they will initiate antibiotic treatment in terminally ill patients . In this study, DNR consent was obtained from 87.5% of 303 patients who died in general medical wards in Korea. Among those deceased patients with DNR orders, 59.6% received antibiotic therapy even after providing a DNR order, although some of them had bacterial infections at the time of death. In addition, antibiotics were administered to patients who died regardless of the presence of DNR orders. This is consistent with the view that physicians may feel more comfortable in continuing to try to correct a theoretically reversible condition using antibiotics even in the face of an irreversible dying process .
This study has some limitations. First, we investigated the isolation of MDROs based on the results of clinical and hospital surveillance cultures limited to the hospital stay during the study period. Since the microbiology data before this admission and data on the time of isolation for each MDRO were not collected, we could not discriminate whether the MDROs isolated during this admission were already colonized before this admission or newly acquired. As a result, we could not determine the causal relation between the antibiotic use and acquisition of MDROs in the study patients. Second, since we included all patients who died in the internal medical wards, and not just those who were terminally ill, there may be differences in the population of deceased patients compared to previous studies. However, because almost 90% of patients had DNR orders, this difference was likely slight. Third, no was made distinction between colonization and infection. We did not determine the appropriateness of the antibiotic therapy administered to the study patients, so we were unable to estimate the proportion of controlled antibiotic use in patients who died in the general medical wards. Nevertheless, based on the finding in this study that many of the deceased patients had MDROs during their hospital stay before their deaths, it is likely that these patients with MDROs could be sources of transmission to other patients.
In conclusion, our study found that a substantial proportion of patients who died in the general medical wards of acute care hospitals were exposed to antibiotics before death. Antibiotic prescription was common even after obtaining their DNR orders in those deceased patients. In addition, the isolation of MDROs during their hospital stay was more common in patients who died than in those who survived. Because these patients may represent reservoirs of resistant bacteria that could be transmitted to surviving patients in hospital settings [8,9], strategies for judicious antibiotic use and appropriate infection control should be applied to this patient population. Further prospective studies are needed to evaluate whether overuse or abuse of antibiotics in the dying patients in general medical wards contributes to the acquisition and transmission of MDROs during their hospital stay.
Conceived and designed the experiments: YGK BNK. Performed the experiments: YGK CM ESK BNK. Analyzed the data: YGK. Contributed reagents/materials/analysis tools: YGK CM ESK BNK. Wrote the paper: YGK CM ESK BNK.
- 1. Opal SM, Calandra T. Antibiotic usage and resistance: gaining or losing ground on infections in critically ill patients? JAMA. 2009;302: 2367–2368. doi: 10.1001/jama.2009.1774. pmid:19952325
- 2. Spellberg B, Guidos R, Gilbert D, Bradley J, Boucher HW, Scheld WM, et al. The epidemic of antibiotic-resistant infections: a call to action for the medical community from the Infectious Diseases Society of America. Clin Infect Dis. 2008;46: 155–164. doi: 10.1086/524891. pmid:18171244
- 3. Daneman N, Gruneir A, Newman A, Fischer HD, Bronskill SE, Rochon PA, et al. Antibiotic use in long-term care facilities. J Antimicrob Chemother. 2011;66: 2856–2863. doi: 10.1093/jac/dkr395. pmid:21954456
- 4. Thompson AJ, Silveira MJ, Vitale CA, Malani PN. Antimicrobial use at the end of life among hospitalized patients with advanced cancer. Am J Hosp Palliat Care. 2012;29: 599–603. doi: 10.1177/1049909111432625. pmid:22218916
- 5. Oh DY, Kim JH, Kim DW, Im SA, Kim TY, Heo DS, et al. Antibiotic use during the last days of life in cancer patients. Eur J Cancer Care. 2006;15: 74–79.
- 6. Ivo K, Younsuck K, Ho YY, Sang-Yeon S, Seog HD, Hyunah B, et al. A survey of the perspectives of patients who are seriously ill regarding end-of-life decisions in some medical institutions of Korea, China and Japan. J Med Ethics. 2012;38: 310–316. doi: 10.1136/medethics-2011-100153. pmid:22252418
- 7. Heo DS. Life-sustaining medical treatment for terminal patients in Korea. J Korean Med Sci. 2013;28: 1–3. doi: 10.3346/jkms.2013.28.1.1. pmid:23341705
- 8. Levin PD, Simor AE, Moses AE, Sprung CL. End-of-life treatment and bacterial antibiotic resistance: a potential association. Chest. 2010;138: 588–594. doi: 10.1378/chest.09-2757. pmid:20472860
- 9. Ford PJ, Fraser TG, Davis MP, Kodish E. Anti-infective therapy at the end of life: ethical decision-making in hospice-eligible patients. Bioethics. 2005;19: 379–392. pmid:16222854
- 10. D'Agata E, Mitchell SL. Patterns of antimicrobial use among nursing home residents with advanced dementia. Arch Intern Med. 2008;168: 357–362. doi: 10.1001/archinternmed.2007.104. pmid:18299489
- 11. Albrecht JS, McGregor JC, Fromme EK, Bearden DT, Furuno JP. A nationwide analysis of antibiotic use in hospice care in the final week of life. J Pain Symptom Manage. 2013;46: 483–490. doi: 10.1016/j.jpainsymman.2012.09.010. pmid:23317761
- 12. Nagy-Agren S, Haley H. Management of infections in palliative care patients with advanced cancer. J Pain Symptom Manage. 2002;24: 64–70. pmid:12183096
- 13. Polk RE, Fox C, Mahoney A, Letcavage J, MacDougall C. Measurement of adult antibacterial drug use in 130 US hospitals: comparison of defined daily dose and days of therapy. Clin Infect Dis. 2007;44: 664–670. pmid:17278056
- 14. Boucher HW, Talbot GH, Bradley JS, Edwards JE, Gilbert D, Rice LB, et al. Bad bugs, no drugs: no ESKAPE! An update from the Infectious Diseases Society of America. Clin Infect Dis. 2009;48: 1–12. doi: 10.1086/595011. pmid:19035777
- 15. Kim DY, Lee KE, Nam EM, Lee HR, Lee KW, Kim JH, et al. Do-not-resuscitate orders for terminal patients with cancer in teaching hospitals of Korea. J Palliat Med. 2007;10: 1153–1158. pmid:17985972
- 16. Fins JJ, Miller FG, Acres CA, Bacchetta MD, Huzzard LL, Rapkin BD. End-of-life decision-making in the hospital: current practice and future prospects. J Pain Symptom Manage. 1999;17: 6–15. pmid:9919861
- 17. Safdar N, Maki DG. The commonality of risk factors for nosocomial colonization and infection with antimicrobial-resistant Staphylococcus aureus, enterococcus, gram-negative bacilli, Clostridium difficile, and Candida. Ann Intern Med. 2002;136: 834–844. pmid:12044132
- 18. Phua J, Joynt GM, Nishimura M, Deng Y, Myatra SN, Chan YH, et al. Withholding and withdrawal of life-sustaining treatments in intensive care units in Asia. JAMA Intern Med. 2015;175: 363–371. doi: 10.1001/jamainternmed.2014.7386. pmid:25581712
- 19. Marcus EL, Clarfield AM, Moses AE. Ethical issues relating to the use of antimicrobial therapy in older adults. Clin Infect Dis. 2001;33: 1697–1705. pmid:11595981