1132678

research-article2022
PMTXXX10.1177/87551225221132678Journal of Pharmacy TechnologyReinert and Kormanyos

Review Article

Journal of Pharmacy Technology

Pharmacologic Management of Central

2023, Vol. 39(1) 29­–34
© The Author(s) 2022
Article reuse guidelines:
Fever: A Review of Evidence for sagepub.com/journals-permissions
DOI: 10.1177/87551225221132678
https://doi.org/10.1177/87551225221132678

Bromocriptine, Propranolol, and Baclofen journals.sagepub.com/home/pmt

Justin P. Reinert, PharmD, MBA, BCCCP1

and Zsanett Kormanyos, PharmD2

Abstract
Objective: The purpose of this review was to evaluate the clinical data supporting bromocriptine, propranolol, and
baclofen in the pharmacologic management of central fever. Data Sources: A comprehensive literature review was
performed between January 2018 and August 2022 using the following keywords: “central fever” NOT “fever” OR
“infection” OR “infectious” AND “neurocritical” OR “neurology” AND “treatment” AND “medication” OR “medicine”
OR “drug” OR “pharmaceutical.” Study Selection and Data Extraction: A total of 6 case reports met specified
inclusion criteria, with 2 reporting on each of the evaluated medications. Data Synthesis: Significant heterogeneity
exists regarding dosing strategies and duration of treatment with these medications for the management of central fever.
Although each medication demonstrated the ability to restore normothermia, the variation in underlying cause of the fever
and lack of cross-over evaluation between different medications makes a definitive treatment strategy for any of these
agents elusive. Conclusions: The development of a central fever has been associated with poor outcomes in patients who
have suffered a critical neurologic injury. Although their exact mechanism for this indication has not been fully elucidated,
anecdotal evidence seemingly supports the use of bromocriptine, propranolol, and baclofen.

Keywords
neurocritical care, central fever, medication efficacy, medication safety, clinical pharmacy, evidence-based medicine

Background analysis demonstrated a 30% incidence of developing a

central fever as a result of an ICH. Furthermore, the pres-
While critically ill patients admitted to an intensive care ence of an IVH or an expansion in volume of the ICH was
unit may be at an increased risk of developing an acute found to be predictors developing a central fever. In evalu-
infectious process, there are other sequalae associated with ating patient outcomes at 90 days utilizing the modified
a traumatic brain injury that require the attention of clini- Rankin score, central fever was associated with poorer out-
cians. Specifically, those patients suffering from an intra- comes in 100% of patients, compared with 46% of patients
cranial hemorrhagic event may be at risk for developing a who remained afebrile during their hospitalization from an
noninfectious fever.1,2 These central fevers have been asso- intracranial hemorrhage. These results solidify the need to
ciated with poor outcomes in patients with an intracerebral expeditiously identify, diagnose, and manage central fevers
hemorrhage (ICH) or an intraventricular hemorrhage (IVH) in an effort to improve patient outcomes.1
and may be a result of either a mechanical injury associated Unfortunately, the diagnosis of central fever is largely one
with the hemorrhage to the thermoregulatory centers of the of exclusion.4 The work conducted by Hocker et al4 conducted
brain, elevated intracranial pressure as a result of an intra-
cranial hemorrhage, or a combination of both.1-3 A study
1
conducted by Commichau et al3 determined that up to 23% Clinical Pharmacy, College of Pharmacy and Pharmaceutical Sciences,
of neurocritical intensive care patients developed a central The University of Toledo, Toledo, OH, USA
2
Department of Pharmaceutical Care, Mercy Health St. Vincent Medical
fever of at least 38.3°C, indicating a high prevalence of this Center, Toledo, OH, USA
issue in this patient population.
A study conducted by Honig et al1 sought to define the Corresponding Author:
Justin P. Reinert, Assistant Professor, Clinical Pharmacy, College of
correlation between an ICH or IVH, the development of a Pharmacy and Pharmaceutical Sciences, The University of Toledo,
central fever, and patient outcomes. A total of 141 patients 3000 Arlington Avenue, HEB 114A, Toledo, OH 43614, USA.
were evaluated in their prospective observational cohort Email: [email protected]
30 Journal of Pharmacy Technology 39(1)

a retrospective cohort analysis of 526 neurocritically ill total of 6 case reports were evaluated in this review. Two of
patients in an effort to determine a set of reasonable diagnos- the included pieces describe treatment with bromocriptine,
tic criteria for central fever. Independent risk factors for the 2 with propranolol, and 2 with baclofen. Seven patient
development of a central fever were identified as the require- encounters are described in the following results, which can
ment of a blood transfusion, a diagnosis of a subarachnoid also be reviewed in Table 1.
hemorrhage (SAH), IVH, or an intracranial tumor, with a
documented fever within 72 hours of admission to the neuro-
Bromocriptine
critical intensive care unit. When combined with a negative
infectious work-up, Hocker et al4 determined the probability Kang et al5 reported a case of a 25-year-old female patient
of the fever being of central origin was 90%. who presented to the emergency department with a chief
Despite the proposed definitions and diagnosis criteria, complaint of drowsiness and urinary incontinence. She was
the optimal management strategy of patients who have found to have a calcified mass in the suprasellar region with
developed a central fever has not been fully elucidated in associated obstructive hydrocephalus. She later underwent
the literature, with data being largely derived from case tumor resection for the craniopharyngioma and was subse-
series and case reports. The pharmacologic management of quently found to be febrile 35 days postoperatively, with
central fever varies by institution and healthcare provider, sustained temperatures above 39.6°C. After being diag-
but revolves around 3 medications: bromocriptine, propran- nosed with both a central fever and mixed autonomic hyper-
olol, and baclofen. Each of their proposed mechanisms for activity disorder, the patient was treated with 0.05 mg/kg of
this indication is unique, with the targeting of hypothalamic bromocriptine every 8 hours. The frequency was subse-
atrophy, parasympathetic activity, and regulation of brown quently increased to every 6 hours followed by every 5
adipose tissue, respectively.5-10 The objective of this review hours, targeted to ensure efficacy. Her fevers drastically
was to evaluate the clinical data supporting bromocriptine, improved after 3 days of bromocriptine therapy, which was
propranolol, and baclofen as used in the management of ultimately continued for a total of 32 days and led to her
central fever. clinical improvement with a noted absence of fevers. The
patient did not experience any adverse effects from her bro-
mocriptine therapy.5
Methods A 44-year-old male patient who presented with stroke-
A comprehensive review of the literature was conducted on like symptoms and was found to have a large IVH and
PubMed, MEDLINE, Scopus, and Web of Science between hematomas in the pons and midbrain and underwent an
January 2018 and August 2022 with the assistance of a emergent ventricular drainage was described by Yu et al.6
medical reference librarian and using the following termi- Following the placement of a ventriculoperitoneal (VP)
nology: “central fever” NOT “fever” OR “infection” OR shunt and tracheostomy for extubation failures, he was
“infectious” AND “neurocritical” OR “neurology” AND transferred to a general medical floor. Although intermit-
“treatment” AND “medication” OR “medicine” OR “drug” tent fevers had been present since admission, the patient
OR “pharmaceutical.” Reports that detailed the pharmaco- had sustained hyperthermic episodes ranging from 37 to
logic management of a diagnosed central fever in a neuro- 41°C after leaving the neurocritical care unit. He was sub-
critically ill patient, described efficacy and safety endpoints, sequently treated with a 14-day empiric course of ampicil-
and provided dosing and duration of treatment were lin/sulbactam, though his infectious work-up continued to
included in this review. For the purposes of this review, effi- be grossly negative. The patient was started on 2.5 mg of
cacy endpoints were considered to be study-determined oral bromocriptine per day for 3 days, after which the dose
success or failure of the intervention, coupled with an was increased to 5 mg daily. The patient received bro-
objective metric documenting the success or failure of the mocriptine therapy for 2 weeks, after which the medica-
intervention (ie, temperature). Safety endpoints were con- tion was stopped and he was discharged to a rehabilitation
sidered to be a disclosure of any adverse drug events, or the facility. A follow-up at 1 month did not reveal any addi-
absence thereof, reported by the authors of the study. tional documented fevers. No adverse effects of bro-
Reports describing cases of uncertain fever etiology and mocriptine were reported.6
those not available in English or readily translatable to
English were excluded. A schematic of the search method-
Propranolol
ology can be viewed in Figure 1.
The use of propranolol for the management of central fevers
in traumatic brain injury patients was described in a case
Results series reported by Meythaler et al7 in 1994. The first patient
Following the application of inclusion and exclusion crite- in their report was a 20-year-old female who presented fol-
ria and the removal of duplicate or nonrelevant results, a lowing a motor vehicle accident but without hemorrhagic
Reinert and Kormanyos 31

Figure 1. Schematic of search methodology.

changes noted on computed topography scans. On the sec- patient’s fevers again sustained greater than 38.9°C and
ond day of her admission, she developed sustained fevers were managed with the resumption of propranolol therapy.
exceeding 38.9°C and was initiated on propranolol 10 mg The second patient, a 21-year-old male suffering from a
by mouth every 6 hours, which was then increased to 20 mg midbrain hemorrhage had similar experience with propran-
every 6 hours to better control her concomitant autonomic olol, in that his fevers which exceeded 40.1°C were man-
dysfunction. The patient’s fevers decreased by an average aged with propranolol 30 mg by mouth every 6 hours. The
of 2°C Celsius. Of interest in this particular case is the fact final case also demonstrated efficacy in a 15-year-old
that, when propranolol doses were held for a procedure, the female with an IVH associated with a motor vehicle
32 Journal of Pharmacy Technology 39(1)

Table 1. Results.

Report type, Time to

number of resolution
Author Medication patients Precipitating event Dosing of fever Comments
Kang et al5 Bromocriptine Case report Mass in suprasellar Initial: 0.05 mg/kg by mouth 72 hours Fever presented 35
N=1 region, obstructive 3 times daily days postoperatively.
hydrocephalus, Maintenance: 0.05 mg/kg by Bromocriptine ordered
craniopharyngioma mouth 5 times daily for 32 days. Patient
weight not specified
Yu et al6 Bromocriptine Case report Intraventricular Initial: 2.5 mg by mouth 72 hours Bromocriptine ordered
N=1 hemorrhage daily for 14 days
Maintenance: 5 mg by
mouth daily
Meythaler Propranolol Case report Traumatic brain Initial: 10–20 mg by mouth 72 hours Fever returned in
et al7 N=1 injuries every 6 hours one patient when
Maintenance: 30 mg by propranolol was held for
mouth every 6 hours a procedure. Propranolol
continued postdischarge
in all patients for 1-2
months, then weaned
Garg et al8 Propranolol Case report Frontal and temporal
Initial: 10 mg by mouth 48 hours Hypothermia reported.
N=1 contusions with twice daily Propranolol continued
associated midline
Maintenance: 10 mg by for 17 days
shift mouth 3 times per daily
Huang et al9 Baclofen Case report Basilar artery Initial:30 mg by mouth daily Unspecified Duration not reported
N=1 occlusion Maintenance: 30 mg by
mouth daily
Lee et al10 Baclofen Case report Pontine hemorrhage Initial: 30 mg by mouth daily 72 hours Baclofen continued
N=1 Maintenance:60 mg by through discharge,
mouth daily duration not reported

accident. She developed sustained fevers greater than normal. No adverse effects other than the incident regarding
40.6°C and was started on propranolol 10 mg by mouth hypothermia were reported.8
every 6 hours, which was ultimately increased to 30 mg
every 6 hours. The patient became afebrile after 24 hours of
Baclofen
propranolol therapy and remained so for the duration of her
hospitalization. She was ultimately discharged to home A 68-year-old-female who suffered a basilar artery occlu-
with a prescription for 10 mg of propranolol by mouth every sion and subsequently developed a central fever was suc-
6 hours which was discontinued 2 months following dis- cessfully returned to normothermia with baclofen therapy.9
charge, as she remained fever free. None of the patients in Huang et al9 initially attempted external thermoregulation
this case series had any adverse drug effects associated with with cooling blankets, but ultimately decided to initiate
propranolol. baclofen therapy once the patient’s temperature reached
More recently, Garg et al8 described a 26-year-old male 40°C. Following the administration of baclofen 30 mg by
who suffered a motor vehicle accident with imaging dem- mouth per day, the patient’s body temperatures returned to
onstrating contusions in the frontal and temporal lobes with normal. No adverse effects were reported.9
a 9 mm midline shift. The day following his decompressive Lee et al10 described a central fever in a 46-year-old
craniotomy, the patient developed fevers exceeding 38.9°C female following a pontine hemorrhage. She developed a
with a max of 41.1°C, and a continuously negative infec- fever that was sustained greater than 38°C for which she
tious work-up. The patient was stared on propranolol 10 mg was treated with a host of antibiotics, despite cultures,
by mouth twice a day, which was later increased to 10 mg 3 imaging, and labs indicating the absence of an infection.
times per day, and achieved continued normothermia. On The patient was started on baclofen 30 mg by mouth per
Day 17 of admission, the patient experienced an episode of day, which was ineffective at breaking the fever and resulted
hypothermia at 36.1°C, after which the propranolol was dis- in the dose being increased to 60 mg by mouth per day. The
continued, and the patient’s temperatures returned to patient became normothermic after 3 days and was
Reinert and Kormanyos 33

ultimately discharged to a skilled facility with the baclofen bromocriptine’s elimination half-life of 4 to 6 hours, it may
therapy ongoing. No adverse effects were reported.10 be reasonable to consider either twice daily or 3 times daily
dosing for this medication, though the data from Yu et al6
proved that once daily bromocriptine was efficacious for
Discussion this indication.5,12 Conversely, Kang et al5 dosed the medi-
Each of the medications described in this review have dif- cation as frequently as every 5 hours to ensure efficacy.
fering mechanisms of action that may contribute to their Similarly, the elimination half-life for propranolol is 3 to 6
efficacy in the management of central fever, though much hours, making it a candidate for 3 or 4 times daily dosing,
information remains to be elucidated regarding their exact however; the evaluation by Garg et al8 demonstrated effi-
mechanisms. Bromocriptine is a centrally acting dopamine cacy using a twice daily regimen.7,16 A disparity exists
2 (D2) agonist that acts upon receptors that are anatomically between the elimination half-life data for baclofen and the
within close proximity to the hypothalamus and may help to efficacy established by Huang et al9 and Lee et al10 While
decrease any physical hypertrophy of the hypothalamic the half-life elimination is approximately 3 to 5 hours, both
region, thereby reducing central fever.5,6 Propranolol, a reports described efficacy with once daily dosing regi-
nonselective beta-adrenergic antagonist, has been histori- mens.17 This may indicate that lower doses of baclofen are
cally utilized for a host of indications ranging from the required to mediate brown adipose tissue than are required
management of heart rate and blood pressure to parasympa- for other indications. Clinicians should consider starting
thetic hyperactivity and essential tremor.7,8 Its lipophilic with a once daily regimen and increasing the frequency
properties allow it to readily penetrate the blood-brain-bar- based on clinical response.9,10
rier, and its ability to work within the central nervous sys- The heterogeneity of the case reports included in this
tem (CNS) has been attributed to its efficacy in the review make a definitive dosing strategy and the selection
management of central fever, though a definitive mecha- of an ideal agent elusive, however; the duration of treat-
nism in this regard remains elusive. Baclofen’s efficacy in ment necessary prior to the restoration of normothermia
central fever is likely attributed to its gamma-aminobutyric is consistent regardless of agent. Each of the 6 patients
acid inhibition and the associated downstream effects on discussed saw a reduction of their central fever toward a
brown adipose tissue, which is associated with thermoregu- normothermic range in approximately 72 hours, meaning
latory functions not attributed with shivering.9,10 that this timeframe could be used by clinicians to justify
Although the absence of adverse drug events were not treatment success, treatment failure, or the need for a
reported in any of the literature reviewed, clinicians should dose adjustment.
remain cognizant of those effects associated with bro- There are several limitations of this review that war-
mocriptine, propranolol, and baclofen. Outside of com- rant discussion. First, it is possible that our search meth-
monly associated adverse drug effects, a few specific odology and inclusion criteria inadvertently omitted
considerations should be given to each of the medication relevant results, though every effort was made by the
discussed. Although a duration of use has not been estab- authors to mitigate as much of this risk as possible. The
lished, abrupt bromocriptine discontinuation may lead to heterogeneity of dosing and duration data makes drawing
dopamine agonist withdrawal syndrome (DAWS).11 specific conclusions about those clinical questions impos-
Tapering the medication as clinically appropriate may be sible, though several important inferences can be made
warranted to prevent this complication.11,12 Propranolol and have been outlined previously in this article. While
may induce first degree, second degree, or complete atrio- our assessment was unable to perform statistical analysis
ventricular blockages, and should be used with caution in of clinical outcomes due to the variability in cases and
patients with autonomic dysfunction or who are otherwise small number of included patients, the efficacy and safety
at risk of developing bradyarrhythmias.13,14 Baclofen may of each medication was demonstrated. Finally, the vari-
decrease the seizure threshold in patients and should be ances in patient demographic information, when pro-
used cautiously in patients with underlying seizure disor- vided, along with the mechanism and classification of
ders or those who are at risk of developing a seizure and injuries does not support universal guidance.
receiving prophylactic antiepileptic medications.15 It is also
imperative that clinicians recognize the potential of each of
Conclusion
these medications to complicate or impede a meaningful
neurologic examination, as they are each associated with The high incidence of central fever amongst neurocriti-
varying degrees of CNS depressive symptoms, including cally ill patients demands that clinicians in this practice
somnolence, confusion, asthenia, and drowsiness.11-15 area be aware of pharmacologic management strategies.
Dosing and duration for these agents should be judi- Bromocriptine, propranolol, and baclofen have been
ciously scrutinized and selected based on the lowest effec- evaluated for this indication and have been found to be
tive dose that simultaneously spares adverse effects. Given largely efficacious and safe. Drug selection, dose titration,
34 Journal of Pharmacy Technology 39(1)

and duration of therapy should be guided by patient-spe- mixed autonomic hyperactivity after neurosurgery: a case
cific parameters in an effort to minimize any associated report. J Korean Med Sci. 2012;27(8):965-968.
adverse drug effects. Continued evaluation of these agents 6. Yu KW, Huang YH, Lin CL, Hong CZ, Chou LW. Effectively
for the management of central fever is warranted. managing intractable central hyperthermia in a stroke patient
by bromocriptine: a case report. Neuropsychiatr Dis Treat.
2013;9:605-608.
Declaration of Conflicting Interests
7. Meythaler JM, Stinson AM 3rd. Fever of central origin in
The author(s) declared no potential conflicts of interest with traumatic brain injury controlled with propranolol. Arch Phys
respect to the research, authorship, and/or publication of this Med Rehabil. 1994;75(7):816-818.
article. 8. Garg M, Garg K, Singh PK, et al. Neurogenic fever in severe
traumatic brain injury treated with propranolol: a case report.
Funding Neurol India. 2019;67(4):1097-1099.
9. Huang YS, Hsiao MC, Lee M, Huang YC, Lee JD. Baclofen
The author(s) received no financial support for the research,
successfully abolished prolonged central hyperthermia in a
authorship, and/or publication of this article.
patient with basilar artery occlusion. Acta Neurol Taiwan.
2009;18(2):118-122.
ORCID iD 10. Lee HC, Kim JM, Lim JK, Jo YS, Kim SK. Central hyper-
Justin P. Reinert https://orcid.org/0000-0003-0321-5608 thermia treated with baclofen for patient with pontine hemor-
rhage. Ann Rehabil Med. 2014;38(2):269-272.
11. Summary safety review—dopamine agonists. Government of
References Canada. Published June 8, 2021. Accessed February 15, 2022.
1. Honig A, Michael S, Eliahou R, Leker RR. Central fever https://hpr-rps.hres.ca/reg-content/summary-safety-review-
in patients with spontaneous intracerebral hemorrhage: pre- detail.php?lang=en&linkID=SSR00269.
dicting factors and impact on outcome. BMC Neurol. 2015; 12. Bromocriptine [prescribing information]. Columbus, OH:
15(1):6. American Health Packaging; Date unknown.
2. Fogelholm R, Murros K, Rissanen A, Avikainen S. Long term 13. Zeltser D, Justo D, Halkin A, et al. Drug-induced atrioven-
survival after primary intracerebral haemorrhage: a retrospec- tricular block: prognosis after discontinuation of the culprit
tive population based study. J Neurol Neurosurg Psychiatry. drug. J Am Coll Cardiol. 2004;44(1):105-108.
2005;76(11):1534-1538. 14. Dahlöf C, Dimenäs E. Side effects of β-blocker treatments
3. Commichau C, Scarmeas N, Mayer SA. Risk factors for as related to the central nervous system. Am J Med Sci.
fever in the neurologic intensive care unit. Neurology. 1990;299(4):236-244.
2003;60(5):837-841. 15. Dario A, Tomei G. A benefit-risk assessment of baclofen in
4. Hocker SE, Tian L, Li G, Steckelberg JM, Mandrekar JN, severe spinal spasticity. Drug Saf. 2004;27(11):799-818.
Rabinstein AA. Indicators of central fever in the neurologic 16. Propranolol tablet [prescribing information]. Hayward, CA:
intensive care unit. JAMA Neurol. 2013;70(12):1499-1504. Impax Generics; 2016.
5. Kang SH, Kim MJ, Shin IY, Park DW, Sohn JW, Yoon YK. 17. Baclofen tablets [prescribing information]. Bridgewater, NJ:
Bromocriptine for control of hyperthermia in a patient with Amneal Pharmaceuticals LLC; 2021.