Tre a tment of Acu t e

E x a c e r b a t i o n s i n C h ro n i c
O b s t r u c t i v e Pu l m o n a r y D i s e a s e
Rajesh Kunadharaju, MDa, Sanjay Sethi, MDa,b,*

KEYWORDS

Chronic obstructive pulmonary disease
Exacerbation
Antibiotics
Corticosteroids

KEY POINTS

Exacerbations in patients with chronic obstructive pulmonary disease are events that worsen qual-
ity of life, impact on lung function decline, and increase the risk of subsequent events and death.

The diagnosis rests on a clinical diagnosis based on the presence of worsening dyspnea and cough
and phlegm.

Its clinical presentation ranges from a mild illness only requiring additional rescue inhaler use to
acute respiratory failure requiring ventilatory support.

Irrespective of the site of treatment, the treatment modalities are the same, with the universal inten-
sification of bronchodilator therapy and selective application of systemic corticosteroids and anti-
biotics.

INTRODUCTION DIAGNOSIS
An acute exacerbation of chronic obstructive pul- A careful history and physical examination remain
monary disease (AECOPD) is a flare of chronic the basis of the diagnosis of an AECOPD. Despite
respiratory symptoms, including worsening years of study, there is no specific diagnostic test
shortness of breath, cough, and sputum produc- or biomarker that is the sine qua non for exacerba-
tion requiring an increase in need of rescue med- tions. It is a diagnosis of exclusion. Most labora-
ications or/and supplementation of additional tory and radiological evaluation in a patient with
treatment.1 It has a heterogeneous presentation suspected AECOPD is to exclude other cardiopul-
and is one of the most common reasons for hos- monary causes and determine the contribution of
pitalization, with an in-hospital mortality rate of comorbidities. Therefore, use of evaluation
3.9%.2 Its severity spectrum ranges from a mild methods are not prescriptive and formulaic, but
illness only requiring additional rescue inhaler should be guided by the clinical assessment.
use to acute respiratory failure requiring ventila- The diagnosis of an AECOPD requires the pres-
tory support. The treatment of exacerbations de- ence of underlying COPD (Fig. 1). Similar respira-
pends on the severity, which itself is a complex tory symptoms in an individual without COPD
sum of the severity of the acute event, the under- should be designated as acute bronchitis, a clin-
lying chronic disease, and associated patient ical entity with substantially different etiology,
comorbidities. prognosis, and treatment. This is easy in patients
chestmed.theclinics.com

a
Division of Pulmonary/Critical Care and Sleep Medicine, Department of Medicine, Jacobs School of Medicine
and Biomedical Sciences, University at Buffalo, SUNY, Buffalo, NY, USA; b Clinical and Translational Research
Center, Room 6045A, 875 Ellicott Street, Buffalo, NY 14203, USA
* Corresponding author. Clinical and Translational Research Center, Room 6045A, 875 Ellicott Street, Buffalo,
NY 14203.
E-mail address: [email protected]

Clin Chest Med 41 (2020) 439–451

https://doi.org/10.1016/j.ccm.2020.06.008
0272-5231/20/Ó 2020 Elsevier Inc. All rights reserved.
440 Kunadharaju & Sethi

Fig. 1. Approach to diagnosing AECOPD.

with established COPD diagnosis and previous poor outcomes. A small autopsy study showed
spirometry demonstrating airflow obstruction. that in patients who died within 24 hours of admis-
However, not infrequently the first presentation of sion for a COPD exacerbation, the primary causes
COPD is an acute exacerbation episode. In fact, of death were heart failure, pneumonia, pulmonary
an average delay of 2 years between the first acute thromboembolism, and COPD in 37%, 28%, 21%,
episode and the establishment of a COPD diag- and 14% of patients, respectively.5
nosis has been described.3 An assessment for Usually the testing described in Table 1 is suffi-
COPD risk factors is indicated in every adult pa- cient to rule out alternative diagnoses. On occa-
tient with a lower respiratory tract infection. If pre- sion, depending on clinical circumstances and
sent, consideration for treatment as an the result of these evaluations, further tests might
exacerbation and an evaluation for undiagnosed be indicated. For example, an atypical presenta-
COPD in the convalescent phase are indicated. tion with sudden onset of dyspnea with or
Once the presence of COPD is established or without an elevated D-dimer, would lead to a
suspected, the health care practitioner needs to computed tomography (CT) pulmonary angiogram
consider other entities in the differential diagnosis. to evaluate for pulmonary embolism. The presence
These include common ones such as pneumonia of atelectasis or mass lesion on the chest radio-
and congestive heart failure and less common graph would require a CT scan for further
ones such as pneumothorax, pulmonary embo- characterization.
lism, pleural effusion, lung carcinoma, and rib frac-
tures.1,4,5 Recently added to the list is vaping-
associated lung injury.6 The extent of evaluation DETERMINATION OF ETIOLOGY
is dictated by the severity of the exacerbation
Around 30% to 50% of AECOPD cases are asso-
and the site of treatment (Table 1). In mild-
ciated with a bacterial infection.7 However,
moderate exacerbations that are being evaluated
sputum Gram stain and culture are not recommen-
in the office, ascertaining absence of hypoxemia
ded in the management of most exacerbations.
with pulse oximetry is the only essential test, if
Studies using molecular diagnostics have shown
the clinical assessment does not support an alter-
that sputum cultures have limited sensitivity.8
native diagnosis. In patients who are severe
The presence of a respiratory pathogen on culture
enough to require emergency department evalua-
does not distinguish between chronic colonization
tion, a more extensive evaluation is indicated to
and acute infection, therefore having limited spec-
ensure appropriate management and prevent
ificity.9 The usual turn-around time of sputum
 Treatment of Acute Exacerbations in COPD 441

Table 1
Diagnostic evaluation of patients with suspected chronic obstructive pulmonary disease exacerbation

Clinical Setting Test Potential Diagnosis

Physician’s office Pulse oximetry Hypoxemia
Emergency Complete blood count Anemia, leukocytosis
department Comprehensive metabolic panel Electrolyte abnormalities, liver failure, kidney
failure
Brain natriuretic peptide Heart failure
Cardiac enzymes Heart failure, myocardial infarction
D Dimer Pulmonary embolism
Arterial blood gas measurement Hypoxemia, hypercapnia, and respiratory
acidosis
Electrocardiography Myocardial infarction, arrhythmias
Chest radiography Pneumonia, congestive heart failure,
pneumothorax, pleural effusion, rib
fractures
Point of care ultrasound – Heart failure, pneumothorax, pleural
if available effusion, pericardial effusion, venous
thromboembolism
Data from Global Initiative for Chronic Obstructive Pulmonary Disease. Global Strategy for the Diagnosis, Management,
and Prevention of COPD: 2020 Report. Available at: https://goldcopd.org/wp-content/uploads/2019/11/GOLD-2020-
REPORT-ver1.0wms.pdf.

culture results of 48 to 72 hours also limits their in AECOPD.12,13 A likely explanation is that bacte-
utility, as clinical decisions regarding treatment of rial exacerbations are mucosal infections, and
an AECOPD require point-of-care results. therefore not enough of a stimulus to induce
There are situations where a sputum culture robust hepatic production of these acute phase re-
could be useful. These include patients who fail actants. Studies with these biomarkers in
to respond to initial empiric antibiotic therapy, pa- AECOPD have not correlated them with careful
tients with risk factors for Pseudomonas aerugi- sputum microbiology and have not used the speed
nosa or other antibiotic-resistant bacteria, and of resolution as an endpoint. They have not shown
patients requiring hospitalization. Even in these their utility beyond the reliable, simple and no-cost
scenarios, the limited accuracy of this test should clinical symptom biomarkers of bacterial infection
be kept in mind when interpreting the results, with in exacerbations.
the clinical picture taking precedence. For Clinical symptom biomarkers of bacterial infec-
instance, in a hospitalized patient responding tion are useful and accurate in AECOPD. These
well to current antibiotics despite the sputum cul- include the presence of purulent sputum, espe-
ture yielding a resistant pathogen, a change of an- cially when the purulence emerged or increased
tibiotics is unnecessary.10 with the onset of the exacerbation. Also useful
Around 30% to 50% of AECOPD cases are are the Anthonisen criteria, where the probability
associated with a viral infection.7 Viral cultures of bacterial etiology in type 1 and 2 exacerbations
are now rarely performed. During the influenza is substantial and reliable enough to recommend
season, a rapid diagnostic test for influenza is use- antibiotic use (Box 1).14 Sputum purulence can
ful if acute onset of fever, myalgias, and coryza are be assessed readily by questioning the patient or
seen with exacerbation symptoms. More exten- visual examination of a sputum sample if available.
sive respiratory viral diagnostic panels have limited Another option is the use of a standardized color
utility in the absence of specific antiviral treatment. chart that has a sensitivity of 94.4% and specificity
Detection of a virus does not mean absence of a of 77.0% for a positive sputum culture and pro-
concurrent bacterial infection; as such, coinfection vides greater reproducibility.15
is seen in about 25% of exacerbations.11
Biomarkers for bacterial infection like procalci-
tonin and C-reactive protein (CRP) have been eval- DETERMINATION OF THE SEVERITY OF AN
uated in AECOPD to guide antibiotic therapy. EXACERBATION
Although these biomarkers perform well with The symptomatic criteria proposed by Anthonisen
parenchymal and systemic infections such as have been sometimes thought of as a measure of
pneumonia and sepsis, they have been unreliable exacerbation severity (see Box 1). However, they
442 Kunadharaju & Sethi

are more useful as an indication of etiology rather care resources or the required level of care1
than as a measure of severity. For example, a pa- (Box 2). However, this classification does not
tient could have a single symptom of increased help in deciding how to manage an exacerbation.
dyspnea of such severity that it requires hospitali- Other limitations of such an approach to defining
zation but still have a type 3 exacerbation, which severity include the inability to assess acute
clearly is not a mild exacerbation. change and the variability in the application of
Ideally, measures of severity of exacerbations health care interventions among practitioners
should reflect the extent of acute pathophysiology and systems. However, such a severity classifica-
(eg, the severity of lung dysfunction or the degree tion is useful in clinical trials and is also used in
of mucosal inflammation). An exacerbation is an determining future preventative care for exacerba-
amplification of chronic disease. Therefore, a sin- tions in an individual patient. More work needs to
gle measurement at the time of an acute exacer- be done to develop a uniform, reproducible, and
bation will not inform of the severity of the acute clinically relevant severity classification system
change, unless one takes the patient’s stable for COPD exacerbations.
baseline state into account. For instance, a
measured peak expiratory flow rate (PEFR) of SITE OF TREATMENT
60 L/min at the time of an exacerbation could
represent a small acute change in an individual Although more than 80% of patients with exacer-
whose baseline PEFR is 80 L/min or a large acute bations are managed as outpatients,16,17 patients
change in an individual with a baseline PEFR of with respiratory distress or at risk of such distress
240 L/min. Although both will require the same should be hospitalized to provide access to higher
level of care, the second individual has a more se- levels of care1 (Box 3). Mortality risk in AECOPD
vere exacerbation based on pathophysiological increases with the development of respiratory
impact. acidosis, change in mental status, presence of sig-
The most used severity classification of nificant comorbidities, and the need for ventilatory
AECOPD is based on the requirement for health support. The emergency department plays an
essential role in the triage and initial management
of severe COPD exacerbations. A well-developed
algorithm to standardize diagnosis, evaluation,
Box 1
Anthonisen classification of AECOPD based on
treatment, and triage decisions in the emergency
clinical findings department can improve the care of COPD
exacerbations.
Major Criteria:

Increased breathlessness, OUTPATIENT MANAGEMENT OF CHRONIC

Increased sputum volume OBSTRUCTIVE PULMONARY DISEASE
EXACERBATION

New or increased sputum purulence
Minor criteria: Outpatient management of AECOPD includes the
intensification of bronchodilator therapy and

Fever without other cause

Increased wheezing or cough Box 2

20% increase in respiratory or heart rate GOLD severity classification based on resource
compared with baseline requirements

Upper respiratory tract infection in the previ-
Mild exacerbation – treated with rescue in-
ous 5 days halers that is, short-acting bronchodilators
Types of exacerbation:
Moderate exacerbation – treated with rescue
inhalers, antibiotics and/or oral

Type 1 exacerbations – All of major criteria.
corticosteroids

Type 2 exacerbations - Any 2 major criteria

Severe exacerbation – Requires Emergency

Type 3 exacerbations - consist of any 1 of ma- Room visit or hospitalization including an
jor criteria along with at least 1 of the minor Intensive care unit admission
criteria
Data from Global Initiative for Chronic Obstructive
Data from Anthonisen NR, Manfreda J, Warren CP, Pulmonary Disease. Global Strategy for the Diagnosis,
et al. Antibiotic therapy in exacerbations of chronic Management, and Prevention of COPD: 2020 Report.
obstructive pulmonary disease. Ann Intern Med Available at: https://goldcopd.org/wp-content/
1987;106(2):196-204. uploads/2019/11/GOLD-2020-REPORT-ver1.0wms.pdf.
 Treatment of Acute Exacerbations in COPD 443

Box 3 is to reduce dyspnea and improve exercise toler-

Indications for inpatient care and critical care, ance, and it may help with mucus clearance by
modified from the GOLD guidelines improving the effectiveness of cough.
Inhaled short-acting b-adrenergic agonists
Indications for inpatient general medicine (SABAs) are the mainstay of therapy because of
admission: their rapid onset of action and efficacy. Albuterol

Dyspnea at rest and levalbuterol are the most used formulations

New hypoxemia at rest or marked increase in delivered by a multidose inhaler (MDI) with or
oxygen requirements in a patient on home without a spacer device or via a nebulizer. In the
oxygen preceding the exacerbation outpatient setting, the usual instructions are to
use the MDI every 3 to 4 hours as needed

Tachypnea (respiratory rate >30 breaths per
minute) (maximum 6 times/d) or the nebulizer every 5 to
6 hours as needed (maximum 4 times/d). If such

Tachycardia (heart rate >100 beats per frequency of use is not adequate to alleviate dys-
minute)
pnea, that usually indicates the need for further

Unable to perform activities of daily living at in-person evaluation. Although a nebulizer is
home without excessive dyspnea easier to use in the acute setting, it should be

Presence of serious comorbidities including noted that there is no significant efficacy differ-
pneumonia, cardiac arrhythmia, heart failure, ence between properly administered MDI and
diabetes mellitus, renal failure, or liver failure nebulizer treatments.18 Common adverse effects

Failure of outpatient management or emer- are headache, nausea, palpitations, tremor, and
gency department management vomiting.

Insufficient home support Inhaled short-acting muscarinic antagonists
(SAMAs) can provide additional bronchodilation

Frailty
in combination with inhaled SABAs. Ipratropium
Indications for intensive care admission: is the most used formulation by MDI or nebulizer

Cardiopulmonary arrest every 4 to 6 hours. Common adverse effects are
dry mouth, tremor, and urinary retention.19

Severe hemodynamic instability

Deteriorating mental status and inability to
protect the airway Systemic Corticosteroids

Agonal breathing Oral corticosteroids are commonly prescribed in

Risk for respiratory failure presenting with outpatients with AECOPD despite scant data.
cyanosis, pH less than 7.35, PaCO2 greater Most trials of systemic corticosteroids in AECOPD
than 45 have been in hospitalized or emergency depart-

Hypoxia (SpO2 <88%) despite supplemental ment patients, who are likely more severe than
oxygen and requiring noninvasive ventilation office-based patients. A short course of systemic
corticosteroids is thought to be benign; however,

Use of accessory muscles of respiration, para-
studies in asthma demonstrate that intermittent
doxic chest wall and abdominal movements.
steroid courses are associated with worse long-
Data from Global Initiative for Chronic Obstructive term health outcomes.20 Emerging data also
Pulmonary Disease. Global Strategy for the Diagnosis, show that higher levels of blood and sputum eo-
Management, and Prevention of COPD: 2020 Report.
Available at: https://goldcopd.org/wp-content/
sinophils are a biomarker for steroid response,
uploads/2019/11/GOLD-2020-REPORT-ver1.0wms.pdf. and conversely, when eosinophils are not
elevated, clinical failure may be increased in
AECOPD with systemic corticosteroids.21,22
Microbiome studies have demonstrated that a
decisions about the use of oral glucocorticoids short course of oral corticosteroids used to treat
and oral antibiotics. AECOPD can cause an increased abundance of
airway microbial species that lasts for several
Bronchodilator Therapy
weeks.23 All these data points do not support uni-
Airway caliber is reduced in exacerbations by a versal prescription of systemic corticosteroids in
combination of increased luminal mucus, mucosal AECOPD, and suggest that such a practice could
edema, and inflammation and bronchospasm. do more harm than good in some patients.
Therefore, bronchodilators, primarily by the How does one choose when to use systemic
inhaled route, are universally indicated in AECOPD corticosteroids in AECOPD? No clear guidelines
management. The goal of bronchodilator therapy exist, but the criteria the authors use are outlined
444 Kunadharaju & Sethi

in Box 4. When indicated, the usual dose is oral prominent wheezing may indicate less probability
prednisone 40 mg every day for five days.1 of antibiotic benefit. Once it is determined who to
Whether extended courses up to 14 days are use- treat, the next important question is the choice of
ful has never been systematically examined in out- specific antibiotics. Unlike corticosteroids, there
patients, although they are not better than short is significant variation in antimicrobial activity,
courses in hospitalized patients. Having said pharmacokinetic/pharmacodynamic properties,
that, if there appears to be an incomplete or and adverse event profiles among antibiotics,
short-lived response after 5 days, a longer course making use of the same antibiotic in all exacerba-
could be tried if other reasons for nonresponse are tions inappropriate. Rational antibiotic choice for
ruled out clinically. Adverse effects include hyper- AECOPD requires a risk stratification approach
glycemia, heartburn, infection, gastrointestinal (Fig. 2). In AECOPD, one can define 2 categories
bleeding, psychomotor disturbance, and steroid of risk26: the risk for a poor clinical outcome (ie,
myopathy, although most of these (except hyper- treatment failure, early relapse, hospitalization,
glycemia) are usually seen after prolonged use.1,24 death) or the risk of antimicrobial resistance. Indi-
viduals who carry these risks need early, broad-
spectrum aggressive antibiotic therapy to optimize
Antibiotics
outcomes.
The primary role of antibiotics in AECOPD is to The risk factors for poor outcomes in AECOPD
hasten the host response-dependent resolution have been reproducibly identified in several obser-
process and prevent complications. The weight vational studies and secondary analyses of ran-
of evidence, from meta-analyses, placebo- domized controlled trials. It is a simple matter to
controlled trials, and database studies, supports determine these risk factors and then stratify an
improved outcomes with antibiotics in all but AECOPD patient as complicated or uncompli-
mild single-symptom exacerbations. The evidence cated. Absence of all the risk factors defines an
from placebo-controlled randomized trials was uncomplicated patient, whereas a complicated
recently summarized in a 2018 meta-analysis patient will have 1 or more of these risk factors
that included 19 randomized trials in 2663 pa- (see Fig. 2).27–30 As these risk factors have been
tients.25 Unlike corticosteroid trials, several of independently associated with poor outcome in
these antibiotic trials have been done in the outpa- exacerbations, it is likely they are additive. Compli-
tient setting. Among outpatients, a significant risk cated patients are more likely to be hospitalized or
reduction (risk ratio 0.69, 95% confidence interval die because of an exacerbation, making the failure
(CI) 0.53 to 0.90) for treatment failure within of initial antibiotic treatment expensive and
4 weeks was shown in 9 trials with 1332 partici- severely detrimental in these patients.
pants. Common adverse effects are diarrhea and The risk for infection by an antimicrobial-
the concern of the promotion of antibiotic resistant pathogen includes the risk for infection
resistance.26 with intrinsically drug-resistant pathogens (eg, P
As about half the exacerbations are bacterial, aeruginosa) and the risk of infection with
the first decision is who to treat. That is based on antibiotic-resistant strains of pathogens that are
looking for evidence for bacterial infection, as dis- generally antibiotic susceptible (eg, Streptococcus
cussed above. A lack of bronchitic symptoms and pneumoniae). Because P aeruginosa is associated
with severe COPD, patients at risk for this path-
ogen are a subgroup of complicated patients
Box 4 with additional risk factors (Fig. 3). Presence of
Proposed indications for systemic steroid use in risk for P aeruginosa requires the selection of anti-
acute exacerbation of chronic obstructive biotics that are active against this pathogen. The
pulmonary disease risk for antibiotic-resistant strains of the common
pathogens is best predicted by recent antibiotic

Prominent wheezing
use for any reason, with recent usually defined

Severe underlying COPD where treatment as the past 3 months.31 Selection of resistant path-
failure could result in hospitalization ogens because of recent use is often specific to

Blood eosinophils that are elevated or high that antibiotic class. Therefore, this risk is easily
normal (at baseline or at exacerbation) handled by using an antimicrobial from a different

Absent or minimal bronchitic symptoms class.
(cough and sputum) of exacerbation Antibiotic choices for uncomplicated patients

Previous exacerbations have required cortico- include a macrolide, a cephalosporin, or trimetho-
steroid treatment for resolution prim/sulfamethoxazole. Amoxicillin is inadequate
because of the substantial incidence of b-
 Treatment of Acute Exacerbations in COPD 445

Fig. 2. The authors’ approach to antibiotic treatment of AECOPD in outpatients. * Antiviral therapy for influenza
is also indicated for exacerbations triggered by influenza infection { Suspicion for other cardiopulmonary disor-
ders (heart failure, pneumothorax) and more severe infections (eg, pneumonia) should be absent for the diag-
nosis of an acute COPD exacerbation. D Age alone is not a strict risk factor but should be considered as
additive to other risk factors. > Selection among antibiotic choices is based on local microbial sensitivity patterns,
patient comorbidities, prior infecting organisms, potential adverse events and drug interactions, and also pro-
vider and patient preferences. In particularly, modifications to this regimen may be needed for patients with a
history of drug-resistant Pseudomonas based on severity of illness, degree of suspicion for Pseudomonas, and
prior susceptibility profiles of pseudomonal isolates. x If recent antibiotic exposure (eg, within the past 3 months),
select an antibiotic from a different class than the most recent agent used. U Trimethoprim-sulfamethoxazole is a
reasonable alternative when macrolides and cephalosporins cannot be used due to allergy, potential adverse ef-
fects, or availability. z Because fluoroquinolone resistance is prevalent among Pseudomonas aeruginosa strains,
we obtain a sputum Gram stain and culture with susceptibility testing for these patients to help guide subsequent
management decisions. For most other outpatients, obtaining a sputum culture is not needed unless the patient
fails to respond to empiric treatment. y Levofloxacin has lesser activity against Pseudomonas than ciprofloxacin
but has greater activity against S. pneumoniae and M. catarrhalis is thus a reasonable alternative to ciprofloxacin
for patients who are at increased risk of Pseudomonas infection but lack microbiologic evidence of Pseudomonas
infection or colonization. Adapted with permission from: Sethi S, Murphy TF. Management of infection in exac-
erbations of chronic obstructive pulmonary disease. In: UpToDate, Post TW (Ed), UpToDate, Waltham, MA. (Ac-
cessed on [Date].) Copyright Ó 2020 UpToDate, Inc. For more information visit HYPERLINK “http://www.
uptodate.com” www.uptodate.com.
446 Kunadharaju & Sethi

Fig. 3. The authors’ approach to antibiotic treatment of AECOPD in the hospitalized patient. * Antiviral therapy
for influenza is also indicated for exacerbations triggered by influenza infection. { Selection among antibiotic
choices is based on local microbial sensitivity patterns, patient comorbidities, prior infecting organisms, potential
adverse events and drug interactions, and also provider and patient preferences. Modifications to these regimens
may be needed for patients with suspicion for specific pathogens and/or history of drug-resistant organisms (eg,
drug-resistant Pseudomonas). D If recent antibiotic exposure (eg, within the past 3 months), select an antibiotic
from a different class than the most recent agent used. Adapted with permission from: Sethi S, Murphy TF. Man-
agement of infection in exacerbations of chronic obstructive pulmonary disease. In: UpToDate, Post TW (Ed), Up-
ToDate, Waltham, MA. (Accessed on [Date].) Copyright Ó 2020 UpToDate, Inc. For more information visit
HYPERLINK ”http://www.uptodate.com” www.uptodate.com.

lactamase production among common pathogens recommended in the subgroup at risk for P aerugi-
involved in AECOPD.32 The authors also avoid tet- nosa, with the caveat that emerging resistance to
racyclines, because recent studies with doxycy- this antibiotic has compromised its efficacy.35
cline have shown limited bacteriologic and Therefore, it is prudent to obtain a sputum culture
clinical efficacy.33,34 Complicated patients can be if possible before starting antibiotic treatment
treated with amoxicillin/clavulanate or a respira- whenever this pathogen is suspected, to allow
tory fluoroquinolone. Ciprofloxacin is for subsequent adjustment of antibiotics.
 Treatment of Acute Exacerbations in COPD 447

Adjunctive Care CARE ON THE GENERAL MEDICAL FLOOR

Exacerbations are an opportune time to discuss Management of AECOPD on the general medical
and encourage smoking cessation in current floor involves reversing airflow limitation with
smokers. An assessment of proper inhaler tech- inhaled short-acting bronchodilators and systemic
nique and education is another important interven- glucocorticoids, treating infection, addressing co-
tion. The data showing benefit with mucolytics, morbid conditions, ensuring adequate oxygena-
expectorants, and cough suppressants in tion, and close monitoring.
AECOPD are scant and of poor quality. The use
of these agents beyond the provision of adequate Bronchodilator Therapy
hydration is best individualized.
The principles of bronchodilator treatment for
AECOPD in the hospital setting are similar to out-
Failure to Respond to Initial Treatment patients. Most commonly, SABA and SAMA
agents are administered together. The nebulized
Clinical improvement in an AECOPD after route is often preferred initially because of ease
commencing treatment should be expected within and reliability. As the patient improves, the fre-
48 to 72 hours, although complete resolution takes quency of short-acting bronchodilators can be
several more days. Absence of improvement or reduced and switched over to MDI. Long-acting
deterioration should lead to a prompt reassess- bronchodilators are not appropriate during an
ment in person. This should include a careful clin- exacerbation and usually can be resumed or initi-
ical evaluation to exclude conditions described in ated on discharge.
the differential diagnosis, and to assess the wors-
ening of comorbidities that have symptoms that Corticosteroids
overlap with AECOPD. A classic example is the
decompensation of chronic congestive heart fail- The benefit of systemic corticosteroids for
ure following an exacerbation induced by AECOPD is supported by randomized controlled
increased stress, hypoxemia, and steroid-related trials in the hospitalized setting. Systematic re-
fluid retention. Additional diagnostic tests could views and meta-analyses of these trials have
be useful, as indicated in Table 1. If bacterial infec- shown that systemic corticosteroids improve
tion indicators have developed or persisted, a lung function, reduce treatment failure and
sputum Gram stain and culture should be consid- relapse, and shorten the length of hospital stay.
ered to identify resistant pathogens. If alternative However, they do not reduce mortality, and the
diagnoses and comorbid conditions are excluded most common adverse effect is hyperglycemia.36
as a cause of treatment nonresponse, switching Unless specific contraindications exist, sys-
to an alternative antibiotic and/or the addition of temic corticosteroids should be prescribed in all
systemic corticosteroids should be considered. hospitalized exacerbations. Although this has
been common practice for the last several de-
cades, there has been a substantial change in ste-
HOSPITAL MANAGEMENT OF CHRONIC roid dose and duration in recent years. After the
OBSTRUCTIVE PULMONARY DISEASE SCCOPE (Systemic Corticosteroids in Chronic
EXACERBATION Obstructive Pulmonary Disease Exacerbations)
trial published in 1999 established the benefit of
Patients who progress despite outpatient man- steroids in hospitalized AECOPD, the study dose,
agement or have indications for hospitalization methylprednisolone 125 mg four times daily for
are managed on the medical floor or intensive 3 days, followed by oral steroid taper for a total
care unit (ICU) based on the severity of the presen- of 14 days became the standard dose and dura-
tation. The basic tenets of therapy of the exacer- tion.37 Although subsequent trials suggested that
bation with bronchodilators, antibiotics, and lower doses and shorter durations of systemic cor-
corticosteroids do not change with hospitalization. ticosteroids are sufficient, it was only after the
However, hospitalization does allow for the provi- recent REDUCE trial published in 2013 that the
sion of these treatments in a more intensive current dose of prednisone 40 mg for 5 days
manner and by alternative routes. In addition, became the standard of care.38 Observational
frequent assessment for signs of respiratory studies have shown the previous dosing regimens
distress and changes in mental and hemodynamic to be associated with more adverse effects such
status is possible in the inpatient setting, as well as as hyperglycemia and infections. There appears
the provision of supplemental oxygen and ventila- to be no significant difference in clinical outcomes
tory support. between the parenteral and oral route.36 Most
448 Kunadharaju & Sethi

clinicians prefer to give the first 1 to 2 doses paren- preferred, and the total duration of 5 to 7 days is
terally to avoid any undetected impaired absorp- adequate.
tion, and then switch to the oral route.
Although often practiced, the use of a longer Antiviral Agents
course of steroids (eg, 14 days with tapering) in
patients who appear to be slowly responding or Antiviral therapy is recommended in patients who
failed a prior short course is not supported by cur- are either outpatients or hospitalized for an
rent evidence and needs to be studied. AECOPD with clinical and laboratory evidence of
influenza infection. Oseltamivir is the most
commonly used antiviral medication, with baloxa-
Antibiotics vir a newly available alternative choice.45
Antibiotics are of proven benefit in hospitalized ex-
acerbations of COPD. In the 2018 meta-analysis of Oxygen Supplementation
placebo-controlled randomized trials, 5 trials in In AECOPD patients presenting with hypoxia, sup-
803 hospitalized patients showed reduced treat- plemental oxygen is a critical component of treat-
ment failure with a relative risk of 0.76 (95% confi- ment with a target arterial oxygen tension of 60 to
dence interval [CI] 0.58–1.00) favoring antibiotic 70 mm Hg or an oxygen saturation of 88% to
use.25 Two large database studies from the United 92%.46,47 Supplemental oxygen use in severe
States also strongly support the benefit of antibi- COPD carries the risk of worsening hypercapnia
otics in hospitalized AECOPD, demonstrating due to combinations of blunting of ventilatory
less treatment failure in antibiotic-treated patients drive, the Haldane effect in the red blood cell,
(odds ratio [OR], 0.87; 95% CI, 0.82–0.92),39 as and worsening ventilation-perfusion mismatch.
well as reductions in in-hospital mortality (relative Therefore, the paradigm of oxygen supplementa-
risk [RR], 0.60; 95% CI, 0.50–0.73) and 30-day tion in AECOPD should be start low, go slow,
readmission for COPD (RR, 0.87; 95% CI, 0.79– with careful titration. The use of titrated oxygen
0.96).40 treatment in hospitalized AECOPD is associated
Whether all hospitalized patients should be with lower mortality rates and a lower likelihood
treated with antibiotics is unclear, with 1 study of hypercapnia or respiratory acidosis than the
suggesting that it is safe to withhold antibiotics in use of nontitrated oxygen.48 The predominant
patients who are milder and do not have any clin- mechanism of hypoxemia in AECOPD is a
ical signs of bacterial infection. However, the ac- ventilation-perfusion mismatch; therefore, it is
curacy of the criteria of bacterial infection has easily correctable with low levels of oxygen sup-
not been adequately tested in hospitalized pa- plementation. Hypoxemia in AECOPD not
tients. In the authors’ experience, often hospital- responding to a relatively low FiO2 supplementa-
ized AECOPD patients, when asked about the tion should prompt evaluation of alternative or
color of sputum, state they can feel chest conges- concomitant diagnoses such as pulmonary
tion but are unable to expectorate sputum emboli, acute respiratory distress syndrome, pul-
because of severely limited expiratory flow. Unless monary edema, or pneumonia.
specific contraindications are present, the authors
treat all hospitalized patients with an antibiotic.
Almost all patients hospitalized primarily for the CARE IN THE INTENSIVE CARE UNIT
exacerbation will fall into the complicated group. Indications for ICU admission in AECOPD include
Furthermore, the Pseudomonas subgroup is likely acute hypoxic or hypercapnic respiratory failure,
larger among patients requiring hospitalization nonresponse to low-flow supplemental oxygen,
compared with complicated outpatients, and its altered mental status with the inability to protect
isolation has been associated with increased mor- the airway, severe hemodynamic compromise, or
tality.41–44 A careful assessment for risk factors for impending respiratory failure. Along with the treat-
Pseudomonas infection is important in hospital- ment modalities discussed previously, these pa-
ized patients, including a review of previous tients require close monitoring and often
sputum microbiology. Antibiotic choices for hospi- noninvasive or invasive ventilatory support.
talized patients without risk factors for P aerugi-
nosa include a respiratory fluoroquinolone or a
Bronchodilator Therapy
third-generation cephalosporin (see Fig. 3).
Among hospitalized patients at risk for P aerugi- Nonintubated patients are usually provided inten-
nosa, antibiotic choices include cefepime, ceftazi- sive nebulized bronchodilator treatment with
dime, piperacillin/tazobactam, or ciprofloxacin. both SABAs and SAMAs. Ventilated patients are
Intravenous therapy for the first 1 to 2 days is provided such treatment by puffs from an MDI
 Treatment of Acute Exacerbations in COPD 449

delivered into the inspiratory limb of the ventilator infections, compared with patients who received
circuit on a regular basis. invasive ventilation.52 Although these were not
randomized controlled trials, a substantial body
Corticosteroids of evidence supports the use of noninvasive venti-
Large randomized controlled trials of systemic lation as the preferred modality in AECOPD pa-
corticosteroids in AECOPD excluded patients tients. Noninvasive ventilation is contraindicated
requiring ventilatory support. A meta-analysis of in patients who are unable to cooperate, unable
smaller trials showed a nonsignificant difference to protect the airway or clear secretions, have
in treatment success and mortality rates between altered mental status, have a facial deformity, are
steroids and placebo in both noninvasively or inva- at high aspiration risk, or have had recent esopha-
sively ventilated patients.49 Despite the lack of geal stenosis.24 Patients who do not respond to
good quality data in the intensive care setting, noninvasive ventilation with worsening mental sta-
given the significant benefits observed in hospital- tus or blood gases should be quickly escalated to
ized patients, the authors use systemic corticoste- mechanical ventilation.
roids in AECOPD patients in the ICU. They use the Invasive ventilation is indicated in severe
parenteral route and a dose of 40 to 60 mg/d for 5 AECOPD patients with status postrespiratory or
to 7 days. It is important to avoid high doses and cardiac arrest, acute respiratory failure with an
prolonged durations, as that could make the inability to tolerate noninvasive ventilation,
benefit-risk ratio for this intervention unfavorable. increased respiratory secretions, altered mental
ICU patients are highly prone to secondary infec- status, inability to protect the airway, high risk of
tious complications of steroid use, both bacterial aspiration, cardiac arrhythmias, and severe hemo-
and fungal. In fact, disseminated aspergillosis dynamic stability.1
has been described in AECOPD patients in the
ICU treated with high-dose steroids.50 Adjunctive Care

Antibiotics AECOPD patients who are hospitalized should

receive deep venous thrombosis prophylaxis,
A single placebo-controlled randomized trial has early physical therapy, and nutrition. Smoking
been conducted with antibiotics in AECOPD pa- cessation counseling and inhaler training prior to
tients in the ICU. In this study in 93 patients, pa- discharge and pulmonary rehabilitation
tients who received a fluoroquinolone, ofloxacin, commencing a few weeks after discharge are
had a significant reduction in treatment failure, important.
mortality, and length of hospital stay. The authors
treat all AECOPD patients in the ICU with antibi-
SUMMARY
otics, with antibiotic choices based on the consid-
erations used for hospitalized patients (see The management of AECOPD often still is a knee-
Fig. 3).25,51 jerk bundle approach of the standard interventions
discussed previously. Emerging data are showing
Ventilatory Support that except for bronchodilators, more selective
Ventilatory support is needed in patients with se- and thoughtful use of treatment modalities in
vere AECOPD presenting with clinical signs of res- AECOPD is prudent. Universal prescription of sys-
piratory muscle fatigue, increased work of temic steroids and antibiotics can negate the po-
breathing, severe hypoxia, respiratory acidosis tential benefits of these interventions. Newer
with pH no more than 7.35 and PaCO2 of at least modalities of diagnosis and treatment, such as
45 mm Hg or severe hemodynamic compromise.1 better biomarkers, antibiotics, and anti-
Ventilatory support can be provided via noninva- inflammatory agents are needed to improve the
sive ventilation with bilevel positive airway pres- current success rates of AECOPD treatment. The
sure (BiPAP) with a face mask or by intubation profound influence of noninvasive ventilation in
and mechanical ventilation. High- flow nasal can- reducing mortality in critically ill AECOPD patients
nula has become an important modality to treat is a good example. In the interim, thoughtful appli-
hypoxic respiratory failure; however, as most cation of antibiotics and corticosteroids is
AECOPD patients are or can become hypercap- indicated.
nic, it is not useful in this setting.
AECOPD patients who receive noninvasive REFERENCES
ventilation have a lower mortality rate and a shorter
length of hospital and ICU stay. Additionally, they 1. Global initiative for chronic obstructive lung disease
have had fewer complications, especially 2020. Available at: https://goldcopd.org/wp-content/
450 Kunadharaju & Sethi

uploads/2019/11/GOLD-2020-REPORT-ver1.0wms. 15. Stockley RA, O’Brien C, Pye A, et al. Relationship of

pdf. Accessed May 30, 2020. sputum color to nature and outpatient management
2. Jinjuvadia C, Jinjuvadia R, Mandapakala C, et al. of acute exacerbations of COPD. Chest 2000;
Trends in outcomes, financial burden, and mortality 117(6):1638–45.
for acute exacerbation of chronic obstructive pulmo- 16. Buist AS, McBurnie MA, Vollmer WM, et al. Interna-
nary disease (COPD) in the United States from 2002 tional variation in the prevalence of COPD (the
to 2010. COPD. J Chronic Obstr Pulm Dis 2017; BOLD Study): a population-based prevalence study.
14(1):72–9. Lancet 2007;370(9589):741–50.
3. Yawn BP, Wollan P, Rank M. Exacerbations in the 17. Jackson H, Hubbard R. Detecting chronic
pre- and post-COPD diagnosis periods. Pragmat obstructive pulmonary disease using peak flow
Obs Res 2013;4:1–6. rate: cross sectional survey. BMJ 2003;327(7416):
4. Myint PK, Lowe D, Stone RA, et al. UK National 653–4.
COPD Resources and Outcomes Project 2008: pa- 18. van Geffen WH, Douma WR, Slebos DJ, et al. Bron-
tients with chronic obstructive pulmonary disease chodilators delivered by nebuliser versus pMDI with
exacerbations who present with radiological pneu- spacer or DPI for exacerbations of COPD. Cochrane
monia have worse outcome compared to those Database Syst Rev 2016;(8):CD011826.
with non-pneumonic chronic obstructive pulmonary 19. McCrory DC, Brown CD. Anticholinergic bronchodi-
disease exacerbations. Respiration 2011;82(4): lators versus beta2-sympathomimetic agents for
320–7. acute exacerbations of chronic obstructive pulmo-
5. Zvezdin B, Milutinov S, Kojicic M, et al. nary disease. Cochrane Database Syst Rev
A postmortem analysis of major causes of early 2003;(1):CD003900.
death in patients hospitalized with COPD exacerba- 20. Bleecker ER, Menzies-Gow AN, Price DB, et al. Sys-
tion. Chest 2009;136(2):376–80. tematic literature review of systemic corticosteroid
6. Werner AK, Koumans EH, Chatham-Stephens K, use for asthma management. Am J Respir Crit
et al. Hospitalizations and deaths associated with Care Med 2020;201(3):276–93.
EVALI. N Engl J Med 2020;382(17):1589–98. 21. Bafadhel M, McKenna S, Terry S, et al. Blood eosin-
7. Sethi S. Infectious etiology of acute exacerbations of ophils to direct corticosteroid treatment of exacerba-
chronic bronchitis. Chest 2000;117(5 Suppl 2): tions of chronic obstructive pulmonary disease: a
380s–5s. randomized placebo-controlled trial. Am J Respir
8. Desai H, Eschberger K, Wrona C, et al. Bacterial Crit Care Med 2012;186(1):48–55.
colonization increases daily symptoms in patients 22. Bafadhel M, Davies L, Calverley PM, et al. Blood
with chronic obstructive pulmonary disease. Ann eosinophil guided prednisolone therapy for exacer-
Am Thorac Soc 2014;11(3):303–9. bations of COPD: a further analysis. Eur Respir J
9. Jacobs DM, Pandit U, Sethi S. Acute exacerbations 2014;44(3):789–91.
in chronic obstructive pulmonary disease: should we 23. Huang YJ, Sethi S, Murphy T, et al. Airway micro-
use antibiotics and if so, which ones? Curr Opin biome dynamics in exacerbations of chronic
Infect Dis 2019;32(2):143–51. obstructive pulmonary disease. J Clin Microbiol
10. Sethi S, File TM, Dagan R. Acute exacerbations of 2014;52(8):2813–23.
chronic bronchitis: diagnosis and therapy; first, 24. Wedzicha JA, Miravitlles M, Hurst JR, et al. Manage-
determine the cause of the worsening symptoms. ment of COPD exacerbations: a European Respira-
J Respir Dis 2003;24(6):257–63. tory Society/American Thoracic Society guideline.
11. Wilkinson TM, Hurst JR, Perera WR, et al. Effect of Eur Respir J 2017;49(3):1600791.
interactions between lower airway bacterial and rhi- 25. Vollenweider DJ, Frei A, Steurer-Stey CA, et al. Anti-
noviral infection in exacerbations of COPD. Chest biotics for exacerbations of chronic obstructive pul-
2006;129(2):317–24. monary disease. Cochrane Database Syst Rev
12. Niewoehner DE. Procalcitonin level-guided treat- 2018;(10):CD010257.
ment reduced antibiotic use in exacerbations of 26. Sethi S, Murphy TF. Acute exacerbations of chronic
COPD. ACP J Club 2007;146(3):57. bronchitis: new developments concerning microbi-
13. Stolz D, Christ-Crain M, Bingisser R, et al. Anti- ology and pathophysiology–impact on approaches
biotic treatment of exacerbations of COPD: a ran- to risk stratification and therapy. Infect Dis Clin North
domized, controlled trial comparing procalcitonin- Am 2004;18(4):861–82, ix.
guidance with standard therapy. Chest 2007; 27. Balter MS, La Forge J, Low DE, et al. Canadian
131(1):9–19. guidelines for the management of acute exacerba-
14. Anthonisen NR, Manfreda J, Warren CP, et al. Anti- tions of chronic bronchitis: executive summary.
biotic therapy in exacerbations of chronic obstruc- Can Respir J 2003;10(5):248–58.
tive pulmonary disease. Ann Intern Med 1987; 28. Miravitlles M, Murio C, Guerrero T. Factors associ-
106(2):196–204. ated with relapse after ambulatory treatment of
 Treatment of Acute Exacerbations in COPD 451

acute exacerbations of chronic bronchitis. DAFNE patients hospitalized with acute exacerbation of
Study Group. Eur Respir J 2001;17(5):928–33. COPD treated with systemic steroids. Chest 2013;
29. Wilson R, Anzueto A, Miravitlles M, et al. Moxifloxa- 143(1):82–90.
cin versus amoxicillin/clavulanic acid in outpatient 41. Gallego M, Pomares X, Espasa M, et al. Pseudo-
acute exacerbations of COPD: MAESTRAL results. monas aeruginosa isolates in severe chronic
Eur Respir J 2012;40(1):17–27. obstructive pulmonary disease: characterization
30. Wilson R, Jones P, Schaberg T, et al. Antibiotic treat- and risk factors. BMC Pulm Med 2014;14:103.
ment and factors influencing short and long term 42. Garcia-Vidal C, Almagro P, Romani V, et al. Pseudo-
outcomes of acute exacerbations of chronic bron- monas aeruginosa in patients hospitalised for COPD
chitis. Thorax 2006;61(4):337–42. exacerbation: a prospective study. Eur Respir J
31. Desai H, Richter S, Doern G, et al. Antibiotic resis- 2009;34(5):1072–8.
tance in sputum isolates of Streptococcus pneumo-
43. Parameswaran GI, Sethi S. Pseudomonas infection
niae in chronic obstructive pulmonary disease is
in chronic obstructive pulmonary disease. Future Mi-
related to antibiotic exposure. COPD 2010;7(5):
crobiol 2012;7(10):1129–32.
337–44.
32. Sethi S, Breton J, Wynne B. Efficacy and safety of 44. Boixeda R, Almagro P, Diez-Manglano J, et al. Bac-
pharmacokinetically enhanced amoxicillin- terial flora in the sputum and comorbidity in patients
clavulanate at 2,000/125 milligrams twice daily with acute exacerbations of COPD. Int J Chron
for 5 days versus amoxicillin-clavulanate at 875/ Obstruct Pulmon Dis 2015;10:2581–91.
125 milligrams twice daily for 7 days in the treat- 45. Dobson J, Whitley RJ, Pocock S, et al. Oseltamivir
ment of acute exacerbations of chronic bron- treatment for influenza in adults: a meta-analysis of
chitis. Antimicrob Agents Chemother 2005;49(1): randomised controlled trials. Lancet 2015;
153–60. 385(9979):1729–37.
33. Daniels JM, Snijders D, de Graaff CS, et al. Antibi- 46. Austin MA, Wills KE, Blizzard L, et al. Effect of high
otics in addition to systemic corticosteroids for acute flow oxygen on mortality in chronic obstructive pul-
exacerbations of chronic obstructive pulmonary dis- monary disease patients in prehospital setting:
ease. Am J Respir Crit Care Med 2010;181(2): randomised controlled trial. BMJ 2010;341:c5462.
150–7. 47. Plant PK, Owen JL, Elliott MW. One year period prev-
34. van Velzen P, Ter Riet G, Bresser P, et al. Doxycy- alence study of respiratory acidosis in acute exacer-
cline for outpatient-treated acute exacerbations of bations of COPD: implications for the provision of
COPD: a randomised double-blind placebo- non-invasive ventilation and oxygen administration.
controlled trial. Lancet Respir Med 2017;5(6):492–9. Thorax 2000;55(7):550–4.
35. Rehman A, Patrick WM, Lamont IL. Mechanisms of 48. Lellouche F, Bouchard PA, Roberge M, et al. Auto-
ciprofloxacin resistance in Pseudomonas aerugi- mated oxygen titration and weaning with FreeO2 in
nosa: new approaches to an old problem. J Med Mi- patients with acute exacerbation of COPD: a pilot
crobiol 2019;68(1):1–10. randomized trial. Int J Chron Obstruct Pulmon Dis
36. Walters JA, Tan DJ, White CJ, et al. Systemic corti- 2016;11:1983–90.
costeroids for acute exacerbations of chronic
49. Abroug F, Ouanes I, Abroug S, et al. Systemic corti-
obstructive pulmonary disease. Cochrane Database
costeroids in acute exacerbation of COPD: a meta-
Syst Rev 2014;(9):CD001288.
analysis of controlled studies with emphasis on
37. Niewoehner DE, Erbland ML, Deupree RH, et al. Ef-
ICU patients. Ann Intensive Care 2014;4:32.
fect of systemic glucocorticoids on exacerbations of
chronic obstructive pulmonary disease. N Engl J 50. Bulpa PA, Dive AM, Garrino MG, et al. Chronic
Med 1999;340(25):1941–7. obstructive pulmonary disease patients with inva-
38. Leuppi JD, Schuetz P, Bingisser R, et al. Short-term sive pulmonary aspergillosis: benefits of intensive
vs conventional glucocorticoid therapy in acute ex- care? Intensive Care Med 2001;27(1):59–67.
acerbations of chronic obstructive pulmonary dis- 51. Nouira S, Marghli S, Belghith M, et al. Once daily
ease: the REDUCE randomized clinical trial. JAMA oral ofloxacin in chronic obstructive pulmonary dis-
2013;309(21):2223–31. ease exacerbation requiring mechanical ventilation:
39. Rothberg MB, Pekow PS, Lahti M, et al. Antibiotic a randomised placebo-controlled trial. Lancet
therapy and treatment failure in patients hospitalized 2001;358(9298):2020–5.
for acute exacerbations of chronic obstructive pul- 52. Rochwerg B, Brochard L, Elliott MW, et al. Official
monary disease. JAMA 2010;303(20):2035–42. ERS/ATS clinical practice guidelines: noninvasive
40. Stefan MS, Rothberg MB, Shieh MS, et al. Associa- ventilation for acute respiratory failure. Eur Respir J
tion between antibiotic treatment and outcomes in 2017;50(2):1602426.