大韓獸醫學會誌 (2015) 第 55 卷 第 4 號Korean J Vet Res(2015) 55(4) : 263~266http://dx.doi.org/10.14405/kjvr.2015.55.4.263

263

<Case Report>

Drug-induced blood cell dyscrasia associated

with phenobarbital administration in a dog

Han-Byeol Jung, Min-Hee Kang, Hee-Myung Park*

Department of Veterinary Internal Medicine, College of Veterinary Medicine, Konkuk University, Seoul 05029, Korea

(Received: October 31, 2015; Revised: December 7, 2015; Accepted: December 15, 2015)

Abstract : A 13-year-old, spayed, female Chihuahua dog was referred for evaluation of fever, lethargy, and dyspnea.Hematologic evaluation revealed severe neutropenia, thrombocytopenia, and mild anemia. The dog had been undergoingphenobarbital therapy for the past 7 weeks because of generalized seizures due to meningoencephalomyelitis of unknownetiology. After ruling out other possible causes of cytopenias, a tentative diagnosis was made of drug-induced bloodcell dyscrasia. The neutropenia and thrombocytopenia resolved after discontinuation of phenobarbital (8 days and 15days after discontinuation, respectively). This is the first case report in Korea to demonstrate blood dyscrasia associatedwith idiosyncratic adverse effects of phenobarbital.

Keywords : leukopenia, neutropenia, phenobarbital, thrombocytopenia

Phenobarbital (PB) is a long-acting barbiturate that enhances

γ-aminobutyric acid-mediated increases in chloride conduc-

tance by opening chloride channels [11, 12]. PB is a well-tol-

erated and effective anticonvulsant for various seizures and

certain types of clinical epilepsy in dogs and cats [9, 11, 12].

PB has infrequent complications including hyperactivity,

restlessness, excessive sedation, ataxia, polyuria, polydipsia,

and polyphagia; these typically occur during the initial phase

of therapy [9, 10]. In most dogs, these side effects usually are

tolerated after 2 to 4 weeks of therapy [10]. No actual hepa-

tocellular damage during PB treatment has been confirmed;

however, PB administration can contribute to the induction

of serum liver enzyme activities in dogs [2, 8]. In addition,

PB has infrequently been associated with hematologic adverse

drug events (ADEs) in dogs, including reversible neutrope-

nia, thrombocytopenia, and anemia [5, 6]. Long term use of

PB can also lead to bone marrow necrosis [13].

This case report describes the clinical and laboratory fea-

tures of PB-related cytopenia and treatment outcomes. To the

best of authors’ knowledge, this is the first case report dem-

onstrating blood dyscrasia associated with idiosyncratic ADE

of PB in our country (South Korea).

A 13-year-old, spayed female Chihuahua dog, weighing

2.8 kg, was referred to Konkuk University Veterinary Teach-

ing Hospital for evaluation of fever, lethargy, and anorexia.

About 9 months previously, the dog had been admitted for

evaluation of generalized seizure and diagnosed with menin-

goencephalomyelitis of unknown etiology (MUE), based on

magnetic resonance imaging findings (Fig. 1), and cere-

brospinal fluid (CSF) analysis showing mononuclear pleocy-

tosis (208 cells/µL; reference range: 0–5 cells/µL). Further

test results including a polymerase chain reaction test for

canine distemper virus and CSF culture were negative.

Despite the diagnosis of MUE in this dog, the owner was

unable to administer any medication due to personal circum-

stances. Since that visit, the frequency of seizures had

increased, so PB (2.5 mg/kg, per orally [po], quarter [q] 12 h,

Hana Pharm, Korea) had been prescribed by the referring

veterinarian approximately 7 weeks prior to admission to our

hospital. Any other drugs except for PB had not been sup-

plied to the dog by the referring veterinarian. Vaccinations

and preventive parasiticidal agents had been administered 6

months prior to our initial examination.

A physical examination revealed hyperthermia (40.4oC),

delayed capillary refilling time, reduced skin turgor, and low

blood pressure (systolic pressure 94 mmHg). A complete

blood count (CBC) showed leukopenia (1.3 × 103/µL; refer-

ence range: 6–17 × 103/µL) with severe neutropenia (367/µL;

reference range: 3000–11000/µL), lymphopenia (643/µL; ref-

erence range: 1000–4800/µL), and thrombocytopenia (171 ×

103/µL: reference range: 200–500 × 103/µL) (Table 1). Dohle’s

bodies and band cells, with toxic changes, including vacuola-

tion, toxic granules, less condensed chromatin, and blue cyto-

plasm, were detected in a Diff-Quik-stained blood smear

(data not shown). Biochemical analysis revealed elevated serum

alkaline phosphatase activity (275 U/L; reference range: 15–

*Corresponding author

Tel: +82-2-450-4140, Fax: +82-2-450-3037

E-mail: parkhee@konkuk.ac.kr

264 Han-Byeol Jung, Min-Hee Kang, Hee-Myung Park

127 U/L), hypoalbuminemia (2.8 g/dL; reference range: 2.9–

4.2 g/dL), hypoglycemia (57 mg/dL; reference range: 70–118

mg/dL), hypokalemia (3.5 mmol/L; reference range: 3.8–5.0

mmol/L), and an elevated level of C-reactive protein (CRP;

above 210 µmol/L; reference range: < 20 µmol/L). Other

parameters were within the reference ranges. The serum PB

concentration was 12.1 µg/mL (reference range: 15–45 µg/

mL). Serological results and microscopy analysis for tick-

borne diseases (babesiois, ehrlichiosis, anaplasmosis, and Rick-

ettsia rickettsii infection) were all negative. Considering sep-

sis or systemic inflammatory response syndrome (SIRS), a

bacterial culture of a blood sample was performed; the test

results were also negative. No neoplasms or other risk fac-

tors were found on abdominal radiography and ultrasonogra-

phy. The history and several other test results led to a strong

suspicion of PB-induced neutropenia and thrombocytopenia.

The anticonvulsant was replaced with potassium bromide

(KBr; loading dose of 400 mg/kg divided into four doses

over a day; maintenance 40 mg/kg, po, q 24 h; Sigma-Ald-

rich, USA) and PB was discontinued after our initial exami-

Table 1. Profiles of relevant complete blood counts and serum biochemical data and serum phenobarbital (PB) concentrations in a dogwith cytopenia

ParametersDays after admission

Reference rangeDay 1 Day 2 Day 3 Day 5 Day 7 Day 8 Day 15 Day 23

WBC (× 103/µL) 1.3 1.17 1.05 1.87 3.57 9.09 23.8 8.18 6–17

RBC (× 106/µL) 7.23 5.80 5.42 5.82 4.69 4.91 6.11 6.39 5.5–8.5

Hct (%) 39.5 37.1 34.2 36.2 30 31.6 37 42.5 37–55

PLT (× 103/µL) 171 86 64 25 56 51 633 455 200–500

Nphs counts (/µL) 367 443 189 318 2213 6454 19754 6708 3000–11000

ALB (g/dL) 2.8 2.1 – 2.1 1.9 2.1 2.7 3.3 2.9–4.2

CRP (µmol/L) > 210 > 210 – > 210 > 210 > 210 60 – < 20

sPB (µg/mL) 12.1 – – 5.6 – 2.6 2.5 1.6 15–45

WBC, white blood cells; RBC, ted blood cells; Hct, hematocrit; PLT, platelets; Nphs, neutrophils; ALB, albumin; CRP, C-reactive protein;sPB, serum phenobarbital concentration.

Fig. 1. Magnetic resonance imaging of a dog with generalized seizure. The intracranial lesions are shown in the transverse (A-C) and

sagittal images (D-F). Multiple lesions (arrows) are present in the left temporal, parietal and frontal lobes. The lesions are iso-intense

on T1-weighted image (A and D), hyper-intense on T2-weighted image (B and E). In the post contrast T1-weighted image, the lesion

was not enhanced (C and F).

Blood cell dyscrasia induced by phenobarbital 265

nation. The dog entered a sterilized intensive care unit filled

with oxygen, and treatment was initially treated with antibiot-

ics (30 mg/kg, intravenously [iv], q 8 h, cefotaxime; Wooridul

Pharmaceutical, Korea; 5 mg/kg, subcutaneously [sc], q 12 h,

enrofloxacin; Bayer, Korea; 15 mg/kg, iv, q 12 h, metronida-

zole; DAI HAN Pharm, Korea), recombinant human granulo-

cyte colony-stimulating factor (rhG-CSF; 5 µg/kg, sc,

Leukokine Inj 300; CJ Healthcare, Korea), and a balanced

electrolyte solution (0.9% saline solution containing 30 mEq

potassium chloride, 120 mL/kg/day, iv, DAI HAN Pharm) for

hemodynamic stability and prevention of a subsequent

inflammatory response. A 20% dextrose solution (DAI HAN

Pharm) was infused intravenously based on blood glucose

level to treat the persistent hypoglycemia. Immunosuppres-

sive medications for MUE treatment were not prescribed at

that time to allow observation of the therapeutic response.

Blood gases, electrocardiography, and urine production were

also monitored in this dog during its hospitalization.

On the second day after admission, the blood pressure was

improved (systolic pressure 134 mmHg). However, the hypo-

glycemia, hyperthermia (39.8oC), and depressed mentation

continued. In addition, a CBC revealed significant leukope-

nia, neutropenia, and thrombocytopenia (white blood cell

[WBC] 1.17 × 103/µL; Neutrophil 443/µL; PLT 86 × 103/

µL). The reticulocyte index value was extremely low. The

numbers of band cells with toxic changes and Dohle’s bod-

ies remained unchanged. The dog developed metabolic aci-

dosis with compensatory hyperventilation. Decreased venous

pH (7.29; reference range: 7.30–7.45), low levels of bicar-

bonate (14.3 mEq/L; reference range: 18–26 mEq/L), and

low levels of partial pressure of carbon dioxide (29.6 mmHg;

reference range: 33–55 mmHg) were identified. Sodium bicar-

bonate (8.4%; Huons, Korea) was administered for bicarbon-

ate supplementation. Although rhG-CSF had been administered

daily for three days, the neutropenia showed no improve-

ment. Antibiotics and other supportive therapies were main-

tained.

On the fifth day after admission, the electrolyte imbal-

ance, hypoglycemia, metabolic acidosis, fever, and lethargy

were resolved. The other hematologic results, including leu-

kopenia, neutropenia, and thrombocytopenia were unchanged.

The serum PB level (5.6 µg/mL) decreased.

On the eighth day after admission, the WBC (9.09 × 103/

µL) and neutrophil (6,454/µL) counts were within the refer-

ence ranges and no band cells with toxic changes could be

identified. The serum PB level (2.6 µg/mL) was lower than

on the first day of admission (Table 1). The thrombocytope-

nia, anemia, and hypoalbuminemia still persisted, but the dog

was discharged from the hospital at the owner’s request.

The dog was reevaluated at seven and fifteen days after

discharge. The owner was satisfied with the improved appe-

tite and activity of the dog. Fifteen days after discontinuing

PB, a CBC revealed leukocytosis and thrombocytosis. The

hematocrit value (37%) and serum albumin level (2.7 g/dL)

was improved, but still low. The CRP value (60 µmol/L) was

also improved.

Twenty-three days after discontinuing PB, the dog had

completely recovered clinically. All affected blood parame-

ters were within reference ranges, and the serum PB level

(1.6 µg/mL) was lower than before. As the dog recovered, all

medications were discontinued, except the anticonvulsant

drug KBr (40 mg/kg, po, q 24 h). Finally, prednisolone (1

mg/kg, po, q 12 h, Yuhan, Korea) and cyclosporine A (25

mg/dog, po, q 24 h; Novartis, Switzerland) were added to the

treatment to manage the MUE. To date, no blood abnormali-

ties have appeared in the dog that has been well managed

and shown no clinical signs.

In the present case, SIRS was diagnosed using the pro-

posed criteria for the diagnosis of SIRS [4]; the dog showed

two out of four parameters in SIRS criteria including hypoth-

ermia or hyperthermia, leukocytosis or leukopenia, tachycar-

dia, and tachypnea. Alternations supporting the diagnosis

were also found in other biochemical results including hypoal-

buminemia, hypoglycemia, a left shift with toxic changes,

and elevated CRP. Although the cause of SIRS was not fully

identified, PB administrations are thought to the cause of

subsequent modulation of systemic immunity or cytokine-

mediated inflammation as well as destruction of neutrophils,

platelets, and erythrocytes.

PB is considered a safe anticonvulsant and is frequently

used in dogs and cats. Nevertheless, it occasionally causes

life-threatening hematologic disorders [5, 6].

Previous reports have demonstrated that PB intoxication,

accompanied by high PB levels, can induce pancytopenia [6].

However, PB can induce severe blood dyscrasia even at low

dosages, as in the present case, where the PB level was even

lower than the reference range (15–45 µg/mL).

The ADEs affecting the hematologic system have been cat-

egorized as type A (dose-dependent responses) or type B

(idiosyncratic reactions unrelated to pharmacologic effects)

reactions [3, 7, 14]. The representative drugs causing type A

ADEs are chemotherapeutic agents and oxidants [14]. A case

with a dose-related type A ADE is expected to improve with

dose reduction [3]. Although type A ADEs can affect the

quality of life, they are usually reversible and rarely require

withdrawal of medicine [15]. Some drugs, in classes such as

estrogenic compounds, nonsteroidal anti-inflammatory drugs,

antibiotics, antithyroid drugs, anticonvulsants, antiparasitics,

and cardiac drugs, can produce idiosyncratic type B ADEs

[14]. These type B ADEs differ from type A ADEs with respect

to their pathogenesis, which occurs unpredictably and is

apparently unrelated to well-known pharmacologic mecha-

nisms in type B ADEs [15].

The mechanism by which PB induces leukopenia, thromb-

ocytopenia, and anemia as ADE of PB treatment has not

been fully explained [6]. A previous report that evaluated the

bone marrow of dogs with PB-induced pancytopenia showed

the occurrence of myeloid hyperplasia [5]. The study revealed

that the PB-induced blood dyscrasia could be caused by

destruction of mature granulocytes [5, 14]. Another study

266 Han-Byeol Jung, Min-Hee Kang, Hee-Myung Park

suggested that bone marrow necrosis caused by PB exposure

could also result in leukopenia, thrombocytopenia, and/or

anemia [13]. In the present case, no bone marrow specimens

were evaluated. In humans, drug-induced cytopenia is defined

when it is discovered during administration of the drug and

resolves within one month after discontinuation of the medi-

cation [1]. Thus, based on the clinical diagnosis and treat-

ment progress, PB induced blood dyscrasia was conclusive in

the present case.

In the present case, PB administration produced hemato-

logic ADEs, which are unrelated to a PB’s pharmacologic

effects, at a serum PB concentration even lower than the ref-

erence range. Considering dose-independence, unpredictabil-

ity, and irrelevance of PB’s pharmacologic effects, the ADE

in this case was classified as an idiosyncratic type B ADE of

anticonvulsants. Although the pathogenesis of PB in this case

has not been fully identified, peripheral destruction of mature

blood cells is suspected as the mechanism of toxicity based

on time to recovery (8 days after discontinuation); Destruc-

tion of mature granulocytes usually resolves 1 weeks after

discontinuation of treatment, while bone marrow necrosis or

myelofibrosis usually recovers 3 to 8 weeks after stopping

treatment [14].

Despite its rare occurrence, PB-induced blood dyscrasia is

a life-threatening ADE of clinical significance because of the

near impossibility of predicting and preventing its adverse

reactions. However, this critical ADE could be minimized by

careful monitoring of clinical responses and laboratory parame-

ters. The present case report is the first to introduce blood

dyscrasia as an idiosyncratic ADE associated with PB in our

country.

References

1. Benichou C, Solal Celigny P. Standardization of definitions

and criteria for causality assessment of adverse drug reactions.

Drug-induced blood cytopenias: report of an international

consensus meeting. Nouv Rev Fr Hematol 1991, 33, 257-

262.

2. Gaskill CL, Miller LM, Mattoon JS, Hoffmann WE,

Burton SA, Gelens HCJ, Ihle SL, Miller JB, Shaw DH,

Cribb AE. Liver histopathology and liver and serum alanine

aminotransferase and alkaline phosphatase activities in epileptic

dogs receiving phenobarbital. Vet Pathol 2005, 42, 147-160.

3. Greenwood RS. Adverse effects of antiepileptic drugs.

Epilepsia 2000, 41, S42-52.

4. Hauptman JG, Walshaw R, Olivier NB. Evaluation of the

sensitivity and specificity of diagnostic criteria for sepsis in

dogs. Vet Surg 1997, 26, 393-397.

5. Jacobs G, Calvert C, Kaufman A. Neutropenia and

thrombocytopenia in three dogs treated with anticonvulsants.

J Am Vet Med Assoc 1998, 212, 681-684.

6. Khoutorsky A, Bruchim Y. Transient leucopenia, throm-

bocytopenia and anaemia associated with severe acute

phenobarbital intoxication in a dog. J Small Anim Pract

2008, 49, 367-369.

7. Maddison JE, Page SW. Adverse drug reactions. In:

Maddison JE, Page SW, Church D (eds.). Small Animal

Clinical Pharmacology. 2nd ed. pp. 41-52, Saunders, Philadelphia,

2008.

8. Müller PB, Taboada J, Hosgood G, Partington BP,

VanSteenhouse JL, Taylor HW, Wolfsheimer KJ. Effects

of long-term phenobarbital treatment on the liver in dogs. J

Vet Intern Med 2000, 14, 165-171.

9. Podell M. Antiepileptic drug therapy. Clin Tech Small Anim

Pract 1998, 13, 185-192.

10. Vernau KM, LeCouteur RA, Maddison JE. Anticonvulsant

drugs. In: Maddison JE, Page SW, Church D (eds.). Small

Animal Clinical Pharmacology. 2nd ed. pp. 367-379, Saunders,

Philadelphia, 2008.

11. Verrotti A, Coppola G, Parisi P, Mohn A, Chiarelli F.

Bone and calcium metabolism and antiepileptic drugs. Clin

Neurol Neurosurg 2010, 112, 1-10.

12. Verrotti A, Scaparrotta A, Grosso S, Chiarelli F, Coppola

G. Anticonvulsant drugs and hematological disease. Neurol

Sci 2014, 35, 983-993.

13. Weiss DJ. Bone marrow necrosis in dogs: 34 cases (1996-

2004). J Am Vet Med Assoc 2005, 227, 263-267.

14. Weiss DJ. Drug-associated blood cell dyscrasias. Compend

Contin Educ Vet 2012, 34, E2.

15. Zaccara G, Franciotta D, Perucca E. Idiosyncratic adverse

reactions to antiepileptic drugs. Epilepsia 2007, 48, 1223-1244.