Procalcitonin-Guided Antibiotic Prescribing for Acute Exacerbations of Chronic Obstructive Pulmonary Disease in the Emergency Department - MDedge
←
→
Page content transcription
If your browser does not render page correctly, please read the page content below
Procalcitonin-Guided Antibiotic Prescribing
for Acute Exacerbations of Chronic
Obstructive Pulmonary Disease in the
Emergency Department
Leah J. Nguyen, PharmD, BCPS; Andrew Varker, PharmD, BCPS; Pamela Slaughter, PharmD, BCPS;
Daniel Boyle, PharmD, BCOP; and Negin Nekahi, MD, PhD
Purpose: Acute exacerbations of chronic obstructive pulmonary tients (73 patients/group) would provide 80% power to detect a
disease (AECOPD) can be caused by viral, bacterial, or environ- between-group difference of 10% in the percentage of patients
mental factors. Recent studies have suggested that procalcitonin who were prescribed antibiotics. Categorical variables were ex-
serum levels may help reduce unnecessary antibiotic use with- pressed using estimates of their frequency and percentages.
out statistically significant differences in rates of treatment failure Percentages were compared using Fisher exact tests. For all
for AECOPD. The purpose of this quality improvement project tests, the significance level was set at 0.05.
was to create a procalcitonin-based algorithm to aid emergency Results: Seventy-three patients were included in the prein-
department (ED) clinicians in the management of patients with tervention group, and 77 patients were included in the
AECOPD who do not require hospitalization and to evaluate its postintervention group. Patients in the preintervention and pos-
efficacy and practicality. The primary outcome of this project was tintervention groups had similar representation in GOLD catego-
the rate of antibiotic prescriptions before and after the initiation ries: 52% and 51% for D, 17.8% and 23.4% for C, 21.9% and
of the algorithm. 16.8% for B, and 8.2% and 7.8% for A, respectively. The rate of
Methods: This study used an observational, retrospective, pre- antibiotic prescriptions decreased by 20% after implementation
and posteducation/intervention design. Clinicians were educated from 83.6% before to 63.6% after implementation (P = .01). The
individually on the use of procalcitonin, and a copy of the algo- differences in reexacerbation rates between the preinterven-
rithm was made available to each clinician and posted in the ED. tion and postintervention groups were similar: 19.2% vs 23.4%
Patients who were discharged from the ED with a diagnosis of at 30 days, 12.3% vs 11.7% at 60 days, and 4.1% vs 9.1% at
an AECOPD were identified using International Classification of 90 days, respectively. Prior to education and introduction of the
Diseases, Tenth Revision codes. Patient charts were reviewed procalcitonin algorithm, procalcitonin was ordered for 1.4% of
from November 2018 to March 2019 for the preimplementa- AECOPD cases. Postimplementation, procalcitonin was or-
tion period and November 2019 to March 2020 for the postim- dered for 28.6% of AECOPD cases and used in clinical decision
plementation period. The rate of antibiotic prescriptions and the making 81.8% of the time.
number of procalcitonin tests ordered before and after the intro- Conclusions: In this study of the implementation of a treatment
duction of the algorithm were analyzed. In addition, information algorithm for patients with mild and moderate AECOPD who
on COPD Global Initiative for Chronic Obstructive Lung Disease present to the ED, procalcitonin was shown to reduce the rate of
(GOLD) grouping and 30-, 60-, and 90-day reexacerbation rates antibiotic prescriptions without an observable difference in reex-
were collected. It was estimated that a sample size of 146 pa- acerbation rates 30, 60, and 90 days after presentation.
T
Author affiliations he Global Initiative for Chronic Ob- most common cause.1-4 However, determin-
can be found at the structive Lung Disease (GOLD) guide- ing the etiology of an exacerbation can be
end of the article.
Correspondence: lines define acute exacerbations of difficult based on symptoms alone and can
Leah Nguyen chronic obstructive pulmonary disease lead to an excessive and unnecessary use of
(leah.nguyen1@va.gov) (AECOPD) as a sudden worsening of respira- antibiotics. Only the change in sputum color
Fed Pract. 2021;38(6).
tory symptoms that require additional inter- is considered highly sensitive and specific for
Published online June 12. ventions. Exacerbations are classified as mild bacterial causes.1 As a result, there has been
doi:10.12788/fp.0141 (treated with short-acting bronchodilators an increased interest in the use of acute bio-
only), moderate (treated with antibiotics and/ markers to determine whether antibiotics are
or oral corticosteroids), or severe (treatment necessary.
requiring hospitalization). Exacerbations Procalcitonin (PCT) is an acute phase re-
must include increased dyspnea, and other actant that increases in response to inflam-
symptoms may involve increased sputum mation, especially inflammation caused by a
volume and purulence, cough, and a change bacterial infection. Recent studies have sug-
in sputum color. These symptoms can be due gested that PCT may be used in patients ex-
to viral, bacterial, or environmental causes, periencing an AECOPD to reduce antibiotic
with viral respiratory infections being the use without impacting rates of treatment
264 • FEDERAL PRACTITIONER • JUNE 2021 mdedge.com/fedpracfailure.5-9 A majority of these studies have TABLE 1 GOLD Classification of Patients
been in the inpatient setting or a combina-
GOLD Preintervention, % Postintervention, %
tion of inpatient and outpatient settings. Classificationsa (n = 73) (n = 77)
The purpose of this study was to create
and to evaluate the efficacy and practical- A 8.2 7.8
ity of a PCT-based algorithm to aid emer-
B 21.9 16.8
gency department (ED) clinicians in the
evaluation of patients with AECOPD who C 17.8 23.4
do not require hospitalization. The primary
outcome of this project was the rate of an- D 52 51
tibiotic prescriptions before and after the
Abbreviation: GOLD, Global Initiative for Chronic Obstructive Lung Disease.
initiation of the algorithm. a
A, low risk for an exacerbation, fewer symptoms; B, low risk for an exacerbation,
more symptoms; C, high risk of an exacerbation, fewer symptoms; D, high risk for an
Methods exacerbation, more symptoms.
This was an observational, retrospective,
pre/post assessment at the Phoenix Veter- mote utilization, a PCT quick order option
ans Affairs Health Care System (PVAHCS) was added to the ED laboratory order menu.
in Arizona. Patients who were discharged ED clinicians were individually ed-
from the ED with a diagnosis of an AE- ucated by the antimicrobial stewardship
COPD were identified using International and emergency medicine pharmacists, an
Classification of Diseases, Tenth Revision infectious disease physician champion,
(ICD-10) codes. Patient charts were re- and the pharmacy resident. Clinicians
viewed from November 2018 to March were educated about PCT and its use in
2019 for the preimplementation group and the setting of AECOPD to aid in the de-
from November 2019 for March 2020 in termination of bacterial infections. Each
the postimplementation group. The peri- clinician received an electronic copy the al-
ods were chosen to reflect similar seasons gorithm and summary of the study proto-
for both the pre- and postimplementa- col before implementation and 3 months
tion interventions. Patients were excluded after implementation for follow-up educa-
from analysis if they required hospital ad- tion. In addition, a printed copy of the al-
mission, were immunocompromised, on gorithm was posted in multiple clinician
chronic antimicrobial therapy, had no doc- workstations within the ED. For the first
umented medical history of COPD, or if month of implementation, the project lead
they were presenting primarily for med- was available full-time in the ED to encour-
ication refills. Information collected in- age algorithm use and to field questions or
cluded the rate of antibiotic prescriptions, concerns from clinicians.
procalcitonin test orders, COPD GOLD
classification, and 30-, 60-, and 90-day re- Outcome Measures
exacerbation rates. The primary outcome was the rate of anti-
A PCT-based algorithm (Figure 1) was biotic prescriptions pre- and postinterven-
developed and approved by the PVAHCS tion. The safety endpoints were 30-, 60-, and
Antimicrobial Stewardship Program, the 90-day reexacerbation rates. Reexacerbation
Pharmacy and Therapeutics committee, and rates were defined by ICD-10 codes and doc-
ED leadership. PCT threshold values were umentation from a primary care visit or sub-
based on values approved by the US Food sequent ED visit. The secondary outcomes
and Drug Administration and previous stud- were the rate of PCT tests ordered and used
ies—antibiotics were discouraged for PCT for treatment decisions. Other areas of in-
levels ≤ 0.25 ng/mL but could be considered terest were antibiotic prescribing trends, du-
for PCT levels > 0.25 ng/mL.5,8,9 Clinicians ration of therapy, and patient COPD GOLD
were not required to use the algorithm, and classification.
the use of clinical judgement was encour-
aged. The recommended antibiotic therapies Statistical analysis
were based on previously approved PVAHCS It was estimated that a sample size of
antimicrobial stewardship guidance. To pro- 146 patients (73 patients/group) would
mdedge.com/fedprac JUNE 2021 • FEDERAL PRACTITIONER • 265Chronic Obstructive Pulmonary Disease
TABLE 2 Pre- and Postintervention Outcomes
Preintervention Postintervention P
Outcomes (n = 73) (n = 77) value
Antibiotic prescribed, No. (%) 61 (83.6) 49 (63.6) .01
Azithromycin 27 (37.0) 21 (27.3) -
Doxycycline 24 (32.9) 017 (22.1) -
Levofloxacin 8 (11.0) 2 (2.6) -
Other 2 (3.3) 9 (18.4) -
Duration of therapy, d 6.8 6.2 -
Reexacerbation rates, No. (%)
30 d 14 (19.2) 18 (23.4) -
60 d 9 (12.3) 9 (11.7) -
90 d 3 (4.1) 7 (9.1) -
Procalcitonin tests, No. (%)
Ordered 1 (1.4) 22 (28.6) -
Used 0 (0) 18 (81.8) -
Documented change in sputum color, No. (%) 13 (17.8) 13 (16.9) -
provide 80% power to detect a between- for 1.4% of AECOPD cases. Postinterven-
group difference of 10% in the percentage tion, PCT was ordered for 28.6% of mild-
of patients who were prescribed antibiotics. to-moderate AECOPD cases and used in
Categorical variables were expressed using clinical decision making per clinical doc-
estimates of frequency and percentages. umentation 81.8% of the time. PCT was
Percentages were compared using Fisher used in 5 GOLD group B patients, 5 GOLD
exact tests. For all tests, the significance group C patients, and 7 GOLD group D pa-
level was set at 0.05. tients. In all cases, PCT was < 0.25 ng/mL.
In 4 cases PCT was ordered but not used:
RESULTS 1 GOLD group D patient refused tradi-
Seventy-three patients were included in the tional treatment with oral corticosteroids,
preintervention group and 77 in the postint- which resulted in the clinician prescribing
ervention group. The GOLD classification antibiotics, and the other 3 cases did not
rates were similar between the groups (Table use PCT based on clinical decision making.
1). In addition, > 90% of patients were White The rate of PCT tests ordered for mild-to-
males and all patients were aged ≥ 50 years, moderate AECOPD over time is depicted
which is characteristic of the US Department in Figure 2.
of Veterans Affairs (VA) population. The average duration of antibiotic ther-
The percentage of antibiotic prescriptions apy was about 6 days pre- and postinter-
decreased by 20% after implementation, falling vention. This is longer than the PVAHCS
from 83.6% before to 63.6% after the imple- recommended duration of 5 days but is
mentation (P =.01). The documented change consistent with the GOLD guidelines rec-
in sputum color remained low compared with ommended duration of 5 to 7 days. 1
antibiotic prescriptions: 17.8% preimplemen- Azithromycin is recommended as a first-
tation and 16.9% postimplementation. The line treatment option at the PVAHCS based
reduction in antibiotic prescriptions was asso- on the local antibiogram, and it remained
ciated with limited differences observed in 30-, the most commonly prescribed antibiotic
60-, and 90-day reexacerbation rates pre- and pre- and postintervention. Outcomes of in-
postintervention: 19.2% vs 23.4%, 12.3% vs terest are detailed in Table 2.
11.7%, and 4.1% vs 9.1%, respectively.
Prior to the education, introduction of DISCUSSION
the algorithm, and implementation of the The implementation of PCT-guided antibi-
PCT quick-order menu, PCT was ordered otic prescribing for patients with mild and
266 • FEDERAL PRACTITIONER • JUNE 2021 mdedge.com/fedpracChronic Obstructive Pulmonary Disease
FIGURE 1 Algorithm Used to Aid Interpretation of Procalcitonin Results
Diagnosis of COPD
exacerbation
Procalcitonin
< 0.25 ng/mL > 0.25 ng/mL
Bacterial etiology: Bacterial etiology:
unlikely unknown
Use clinical judgement to asses patient for
Use clinical judgement to asses patients need for
alternative infectious etiology (eg, pneumonia)
antibiotic therapy as procalcitonin may be falsely
elevated due to severe trauma, surgery, vasculitis,
Antibiotic therapy for COPD exacerbation
burns ESRD, etc.
discouraged
Continue with standard of care for COPD
Continue with standard of care for COPD
management; steriods, breathing treatments/
management; steroids, breathing treatment/
inhalers, etc.
inhalers, etc.
If antibiotic therapy is planned, consider facility guidance for treatment of COPD exacerbation as follows:
• Preferred empiric antibiotic therapy: Oral azithromycin 500 mg every 24 h for 3 d or 500 mg on day 1 and
250 mg every 24 h for 4 d.
•A lternative empiric antibiotic therapy: Oral doxycycline 100 mg every 12 h for 5 d or oral amoxicillin/
clavulanate 875/125 mg every 12 h for 5 d or oral cefuroxime 500 mg every 12 h for 5 d or oral levofloxacin 500
mg every 24 h for 5 d.
Abbreviations: COPD, chronic obstructive pulmonary disease; ESRD, end stage renal disease.
moderate AECOPD who presented to the Many health care providers (HCPs) in
ED resulted in a 20% reduction in antibiotic the ED were unfamiliar with PCT prior to
prescriptions, falling from 83.6% before the implementation. A primary concern with
intervention to 63.6% afterward (P = .01). this study was its impact on diagnostic
The measured decrease in antibiotic prescrip- stewardship. Preimplementation, ED clini-
tions is consistent with other studies evaluat- cians ordered PCT 8 times for any cause.
ing the use of acute phase reactants to guide Postintervention, ED clinicians ordered PCT
antibiotic prescribing for AECOPD.10,11 In 180 times for any cause: 36% of these orders
addition, there was no observed difference were for patients with AECOPD who were
in reexacerbation rates. This adds to the in- discharged from the ED or who required
creasing body of evidence that antibiot- hospital admission. The other orders were
ics are overprescribed in mild and moderate for other respiratory conditions, including
AECOPD.12 This is exemplified in our data asthma exacerbations, pneumonia, bronchi-
by the low percentage of patients who had a tis, sinusitis, pharyngitis, nonspecific respi-
documented change in sputum color; symp- ratory infections, and respiratory failure.
toms that are well known to be highly spe- The early phase of the COVID-19 pan-
cific and sensitive for a bacterial infection in demic coincided with the postinterven-
AECOPD. tion phase of this project. PVAHCS started
mdedge.com/fedprac JUNE 2021 • FEDERAL PRACTITIONER • 267Chronic Obstructive Pulmonary Disease
FIGURE 2 Procalcitonin Tests Ordered for Mild and pelled to maintain patient satisfaction, thus
Moderate AECOPD 2019-2020 leading to unnecessary antibiotic prescrip-
tions.14 The viral illness pack helped fulfill
10 the patient’s expectation to receive treat-
ment after seeking care. In addition, the
Procalcitonin test orders, No.
8
project lead was available full time during
6
the first month of PCT algorithm imple-
mentation to address questions and con-
4 cerns, which may have improved HCPs
overall confidence in using PCT.
2
Limitations
0 Limitations of this project include its pop-
November December January February March
2019 2020 ulation and its retrospective nature. The
PVAHCS patient population is predomi-
Abbreviation: AECOPD, acute exacerbations of chronic obstructive pulmonary disease. nantly older, more White, and more male
compared with the general civilian popula-
preparing for the pandemic in March 2020, tion, and results may not be generalizable
and the first confirmed diagnosis at the fa- to other populations. Data were limited to
cility occurred mid-March. COVID-19 may documentation in the electronic health re-
have contributed to the sharp increase in cord. The population was based on data ex-
PCT tests. There is currently no well-de- traction by the ICD-10 code, which may
fined role for PCT in the diagnosis or man- not be an accurate capture of the total pop-
agement of COVID-19, but ED clinicians ulation as HCPs may not select the most
may have increased their use of PCT tests accurate ICD-10 code on documenta-
to help characterize the etiology of the large tion. Another potential limitation was the
influx of patients presenting with respira- COVID-19 pandemic which may have re-
tory symptoms.13 sulted in HCPs ordering PCT more fre-
quently as more patients presented to the
Strengths ED with undifferentiated respiratory symp-
Strengths of this project include its multi- toms. Finally, there were minimal differ-
modal implementation and overall prag- ences observed in reexacerbation rates;
matic design, which reflects real-world however, although the sample size was
utilization of procalcitonin by ED HCPs. powered to detect a difference in antibiotic
The HCPs were not mandated to follow prescriptions, the sample size was not pow-
the procalcitonin algorithm, and the use ered to detect a statistically significant dif-
of clinical judgment was strongly encour- ference in the primary safety outcome.
aged. This project occurred concomitantly
with the VA Infectious Disease Academic CONCLUSIONS
Detailing education program. The pro- PCT-guided antibiotic prescribing signif-
gram focused on clinician education for the icantly reduced the number of antibiotic
proper diagnosis and treatment of respira- prescriptions without an observable in-
tory tract infections. In addition, viral ill- crease in reexacerbation rates for patients
ness packs were introduced as part of this with mild and moderate AECOPD in the
initiative to reduce unnecessary antibiotic ED. This study provides a pragmatic eval-
prescribing. The viral illness pack included uation of PCT-guided antibiotic prescrib-
standard items for symptom relief, such as ing for patients with AECOPD solely in the
saline nasal spray, cough drops, and hand outpatient setting. Acute phase reactants
sanitizer, as well as an explanation card of like PCT can play a role in the management
why the patient was not receiving antibi- of AECOPD to reduce unnecessary antibi-
otics. Several studies have suggested that otic prescriptions.
patients expect a prescription for an anti-
biotic when they present with respiratory Author affilitations
Leah Nguyen is an Emergency Medicine Clinical Phar-
tract symptoms, and HCPs often are com- macy Specialist at the Portland Veterans Affairs Health Care
268 • FEDERAL PRACTITIONER • JUNE 2021 mdedge.com/fedpracChronic Obstructive Pulmonary Disease
System in Oregon; Andrew Varker is an Infectious Dis- Infect Dis. 2019;68(5):725-730. doi:10.1093/cid/ciy552
ease Clinical Pharmacy Specialist; Pamela Slaughter and 6. M athioudakis AG, Chatzimavridou-Grigoriadou V, Cor-
Daniel Boyle are Emergency Medicine Clinical Pharmacy lateanu A, Vestbo J. Procalcitonin to guide antibiotic ad-
Specialists; Negin Nekahi is an Infectious Disease Special- ministration in COPD exacerbations: a meta-analysis. Eur
ist; Leah Nguyen was a Pharmacy Resident at the time this Respir Rev. 2017;26(143):160073. Published 2017 Jan 31.
article was written; all at Phoenix Veterans Affairs Health doi:10.1183/16000617.0073-2016
Care System in Arizona. 7. v an der Does Y, Rood PP, Haagsma JA, Patka P,
van Gorp EC, Limper M. Procalcitonin-guided ther-
Author disclosures apy for the initiation of antibiotics in the ED: a system-
The authors report no actual or potential conflicts of interest atic review. Am J Emerg Med. 2016;34(7):1286-1293.
with regard to this article. doi:10.1016/j.ajem.2016.03.065
8. H uang DT, Yealy DM, Filbin MR, et al. Procalcito-
Disclaimer nin-guided use of antibiotics for lower respiratory
The opinions expressed herein are those of the authors and do tract infection. N Engl J Med. 2018;379(3):236-249.
not necessarily reflect those of Federal Practitioner, Frontline doi:10.1056/NEJMoa1802670
Medical Communications Inc., the US Government, or any of its 9. Picart J, Moiton MP, Gaüzère BA, Gazaille V, Combes
agencies. This article may discuss unlabeled or investigational X, DiBernardo S. Introduction of a PCT-based algo-
use of certain drugs. Please review the complete prescribing rithm to guide antibiotic prescription in COPD ex-
information for specific drugs or drug combinations—including acerbation. Med Mal Infect. 2016;46(8):429-435.
indications, contraindications, warnings, and adverse effects— doi:10.1016/j.medmal.2016.07.008
before administering pharmacologic therapy to patients. 10. S chuetz P, Chiappa V, Briel M, Greenwald JL. Pro-
calcitonin algorithms for antibiotic therapy deci-
References sions: a systematic review of randomized controlled
1. G
lobal Initiative for Chronic Obstructive Lung Disease. trials and recommendations for clinical algo-
Global strategy for the diagnosis, management and pre- rithms. Arch Intern Med. 2011;171(15):1322-1331.
vention of chronic obstructive pulmonary disease: 2020 doi:10.1001/archinternmed.2011.318
report. Accessd June 2, 2021. http://www.goldcopd.org 11. Butler CC, Gillespie D, White P, et al. C-reactive pro-
/wp-content/uploads/2019/12/GOLD-2020-FINAL-ver1.2 tein testing to guide antibiotic prescribing for COPD
-03Dec19_WMV.pdf exacerbations. N Engl J Med. 2019;381(2):111-120.
2. George SN, Garcha DS, Mackay AJ, et al. Human rhi- |doi:10.1056/NEJMoa1803185
novirus infection during naturally occurring COPD 12. Vollenweider DJ, Frei A, Steurer-Stey CA, Garcia-Aymerich
exacerbations. Eur Respir J. 2014;44(1):87-96. J, Puhan MA. Antibiotics for exacerbations of chronic ob-
doi:10.1183/09031936.00223113 structive pulmonary disease. Cochrane Database Syst
3. S eemungal T, Harper-Owen R, Bhowmik A, et Rev. 2018;10(10):CD010257. Published 2018 Oct 29.
al. Respiratory viruses, symptoms, and inflam- doi:10.1002/14651858.CD010257.pub2
matory markers in acute exacerbations and sta- 13. C enters for Disease Control and Prevention. In-
ble chronic obstructive pulmonary disease. Am terim clinical guidance for management of patients
J Respir Crit Care Med. 2001;164(9):1618-1623. with confirmed coronavirus disease (COVID-19). Up-
doi:10.1164/ajrccm.164.9.2105011 dated February 16, 2021. Accessed May 14, 2021.
4. Rohde G, Wiethege A, Borg I, et al. Respiratory viruses in https://www.cdc.gov/coronavirus/2019ncov/hcp/clinical
exacerbations of chronic obstructive pulmonary disease -guidance-management-patients.html
requiring hospitalisation: a case-control study. Thorax. 14. Gaarslev C, Yee M, Chan G, Fletcher-Lartey S, Khan R. A
2003;58(1):37-42. doi:10.1136/thorax.58.1.37 mixed methods study to understand patient expectations
5. Bremmer DN, Moffa MA, Ma K, et al. Acute exacerbations for antibiotics for an upper respiratory tract infection. An-
of chronic obstructive pulmonary disease with a low pro- timicrob Resist Infect Control. 2016;5:39. Published 2016
calcitonin concentration: impact of antibiotic therapy. Clin Oct 20. doi:10.1186/s13756-016-0134-3
mdedge.com/fedprac JUNE 2021 • FEDERAL PRACTITIONER • 269You can also read
