PANDAS With Catatonia: A Case Report.Therapeutic Response to Lorazepam and Plasmapheresis

JOSEPHINE ELIA, M.D., MARY LYNN DELL, M.D., DAVID F. FRIEDMAN, M.D.,

ROBERT A. ZIMMERMAN, M.D., NAOMI BALAMUTH, M.D.,

ASIM A. AHMED, M.D., AND SUSMITA PATI, M.D., M.P.H.

ABSTRACT

This is a report of an 11-year-old, prepubertal boy with acute-onset urinary urgency and frequency, obsessions and com-

pulsions related to urination, severe mood lability, inattention, impulsivity, hyperactivity, and intermittent periods of immo-

bilization. Fever, cough, otitis, and sinusitis preceded neuropsychiatric symptoms. Antistreptolysin O and DNAse B

antibody titers were elevated, and magnetic resonance imaging revealed bilateral diffuse caudate nuclei swelling. Plas-

mapheresis resulted in significant and rapid clinical improvement of obsessive-compulsive disorder symptoms and a

simultaneous decrease in basal ganglia swelling, consistent with an immune-mediated pathophysiological process involv-

ing group A b-hemolytic streptococci. Hyperactivity, impulsivity, and inattention improved with lorazepam, suggesting that

the attention-deficit/hyperactivity disorder symptoms could be manifestations of catatonia. J. Am. Acad. Child Adolesc.

Psychiatry, 2005;44(11):1145–1150. Key Words: PANDAS, catatonia, obsessive-compulsive disorder, attention-deficit/hy-

peractivity disorder, lorazepam, plasmapheresis, magnetic resonance imaging, basal ganglia.

Obsessive-compulsive disorder (OCD) with a uniqueclinical course is identified by the acronym PANDAS(pediatric autoimmune neuropsychiatric disorders asso-ciated with streptococcal infections; Swedo et al., 1998).Five criteria define PANDAS: (1) the presence of OCDand/or tic disorder, (2) prepubertal onset; (3) episodiccourse characterized by acute, severe onset, and dra-matic symptom exacerbations; (4) adventitious move-ments (choreiform) present during the symptomexacerbation; (5) temporal association between group

A b-hemolytic streptococcal infection and onset orexacerbation of symptoms (Swedo et al., 1998).Additional neuropsychiatric conditions reported in

PANDAS include mood lability, attention-deficit/hy-peractivity disorder (ADHD), overanxious disorder,separation anxiety, tactile/sensory defensiveness, andenuresis (Perlmutter et al., 1998, 1999; Swedo et al.,1998). Catatonia, a nonspecific syndrome not previ-ously described in PANDAS, is characterized by affec-tive (intense and uncontrollable emotions), behavioral(perseverations, stereotypies), and motoric symptoms(immobility, posturing, or excessive purposeless move-ments; Gelenberg, 1976; Northoff, 2002). This stateof hyperactivity could be misinterpreted as ADHD.Lorazepam, an effective treatment in 60% to 80% ofcases (Bush et al., 1996;Northoff et al., 1995) potentiatesg-aminobutyric acid A receptors, which are thought tobe decreased in catatonia (Northoff et al., 1999).

CASE HISTORY

An 11-year-old boy developed fever (103�F), cough,and otitis. One week after onset of symptoms, he was

Accepted June 14, 2005.Dr. Elia is with the Department of Child and Adolescent Psychiatry, Drs.

Friedman, Balamuth, and Pati are with the Department of Pediatrics, andDr. Zimmerman is with the Department of Radiology, The Children’s Hospitalof Philadelphia and the University of Pennsylvania School of Medicine, Phila-delphia; Dr. Dell is with the Department of Psychiatry and Behavioral Science,Emory University, Atlanta; and Dr. Ahmed is with the Department of Pediatrics,Children’s Hospital Boston and Harvard Medical School.

Correspondence to Dr. Josephine Elia, The Children’s Hospital of Philadel-phia, Science Center, 3440 Market Street, Suite 200, Philadelphia, PA 19104;e-mail: elia@email.chop.edu.

0890-8567/05/4411–1145�2005 by the American Academy of Child

and Adolescent Psychiatry.

DOI: 10.1097/01.chi.0000179056.54419.5e

J. AM. ACAD. CHILD ADOLESC. PSYCHIATRY, 44:11, NOVEMBER 2005 1145

hospitalized because of increased sleepiness, atypicalbehavior (curling up with a blanket on the dog cushion),intermittent periods of decreased speech, difficultyspeaking, staring episodes, tearfulness, and a constant,subjective need to urinate even when the bladder wasempty. Sinusitis, revealed by computed tomography(CT), was treated with Augmentin. Three additionalhospitalizations occurred in the following 2 weeks toevaluate the incessant need to void. Laboratory tests,abdominal CT scan, obstruction series, bladder scan,and sonogram were normal.The patient was adopted at 2 days of age. His psy-

chiatric history, with the exception of ADHD symptoms(five inattentive, two impulsive, and two hyperactive), wasessentially negative. The biological family history wasunavailable. Medical history was significant for beestingallergy, occasional otitis media, and a single seizureafter his measles, mumps, and rubella vaccination at15 months of age (EEG at that time was normal). Mi-graine headaches, diagnosed at age 7, had decreased infrequency from four times per year to once per year.At the time of the fourth hospitalization, 21 days after

the onset of symptoms, physical examination was unre-markable. Mental status examination showed a veryrestless 11-year-old boy who was constantly movingfrom place to place. He was impulsive, unable to focushis attention, and distracted by minimal stimuli. Dur-ing several brief periods, he stared and became immo-bilized, maintaining his limb position in mid-motion.His cognition, comprehension, and receptive languagewere always intact. Expressive language was difficult, withdecreased amount of speech. His voice was squeaky, thin,and barely understandable. At times, he used gestures(thumb up or down) to answer questions. His moodwas markedly labile, with unprovoked brief crying spellsimmediately followed by unexplained calmness. Hefrequently reported the need to void and needed con-tinual redirection not to constantly use the bathroom.There was no evidence of psychosis.On a parent-structured interview (Schedule for

Affective Disorders and Schizophrenia for School-AgeChildren (K-SADS; Ambrosini, 2000), he met criteriafor ADHD (symptom count on K-SADS: nine inatten-tive, five impulsive, and four hyperactive symptoms)and OCD. On the Children’s Yale-Brown Obsessive-Compulsive Scale (Scahill et al., 1997), he scored34 out of 40, significant for OCD. The National Insti-tute of Mental Health (NIMH) emotional lability scale

showed a score of 4 (0 indicating no irritability and4, extreme irritability; Perlmutter et al., 1999).

Normal laboratory studies included complete bloodcount with differential, erythrocyte sedimentation rate,C-reactive protein, hepatic and thyroid function tests,calcium, magnesium, phosphorus, heavy metal screen,toxoplasma immunoglobulin M (IgM), Lyme enzymeimmunofluorescence assay IgG and IgM, Epstein-Barrand cytomegalovirus titers, and urine toxicology. Theantinuclear antibody titer was elevated at 1:640 (nor-mal 1:20), with a homogeneous pattern suggestiveof PANDAS. Serum ammonia was elevated at 67.2(normal 9–33 mmol/L). Urinalysis showed trace pro-tein. Cerebrospinal fluid (CSF) had trace white bloodcells but was colorless and clear; Venereal DiseaseResearch Laboratory (VDRL) and herpes simplex viruswere negative; and bacterial, viral, and fungal cultureshad no growth. CSF and serum electrophoresis patternswere identical. Electrocardiogram and EEG were un-remarkable. Urine copper and ceruloplasmin werewithin normal limits, and ophthalmological examina-tion was negative for Kaiser-Fleischer rings, rulingout Wilson’s disease.

Twenty-three days after onset of symptoms, anti-streptolysin O (ASO) and DNAase B antibodies were282 (reference 0–240 IU/mL) and 1:680 (referencerange 1:170), respectively. Throat culture was negative.Twenty-one days after symptom onset, the first mag-netic resonance imaging (MRI) showed significantswelling of both putamina and caudate nuclei, with in-creased signal intensity on T1-weighted images and onaxial fluid-attenuated inversion recovery (FLAIR). Theremainder of the brain was normal (Table 1, Fig. 1Aand B). MRI before gadolinium injection includedsagittal, axial, coronal T1-weighted images, and FLAIRimages. After gadolinium injection, axial, coronal, andsagittal T1-weighted images as well as axial diffusionweighted images were performed. FLAIR was usedfor volumetric measurement, the volumes calculatedfrom all axial images with abnormal signal in the basalganglia. These were added together in a volume andcubic centimeters calculated. FLAIR uses an inversionpulse to null the signal of gross CSF, such as is foundin the ventricles and sulci, so that only the abnormalincreased water content that affects the tissue is seenas high signal intensity. This allows the calculation ofthe volume of affected tissue. In our patient, diffusionimaging indicated rapid movement of water within the

ELIA ET AL.

1146 J . AM. ACAD. CHILD ADOLESC. PSYCHIATRY, 44:11, NOVEMBER 2005

basal ganglia that is caused by a disturbed blood-brainbarrier producing focal vasogenic edema.

TREATMENTS

The time line for the lorazepam and plasmapheresistreatments is included in Table 1. The catatonia was

treated with lorazepam at an initial dose of 1 mg every6 hours for 5 days, tapered to 0.5 mg/day by discharge,11 days later, and discontinued 10 days after discharge.Peripheral venous access via the antecubital fossa was

judged to be potentially hazardous because of the young-ster’s impulsive behavior and hyperactivity. Therefore,a double-lumen catheter was placed in the subclavian

Fig. 1 A: Axial FLAIR image shows symmetric hyperintense signal within the basal ganglia. B: Axial T1-weighted image after gadolinium injection shows no

evidence of abnormal contrast enhancement. C: Axial FLAIR image, 17 days later, shows decrease in the areas of involvement (basal ganglia have decreased to

normal size allowing expansion of ventricles to their normal size).

TABLE 1MRI Volumetric Changes and Treatment Time Line

Onset ofSymptoms (d)

MRI Volumetric Measurements of Caudate and PutamenLorazepamTreatment

Plasmapheresis Treatments

Right Caudate Left Caudate Right Putamen Left Putamen 1 2 3 4 5

21 9.8 cm3 8.4 cm3 13.2 cm3 15.6 cm3

2223 1 mg q 6

24 1 mg q 625 1 mg q 626 1 mg q 627 1 mg q 6

28 Tapered to 0.5 mg/d XX2930 XX

313233 XX

34 XX3536

37 5.5 cm3 5.8 cm3 9.7 cm3 10.5 cm3 XX38 Discharged

Note: q 6 = every 6 hours.

PANDAS WITH CATATONIA

J. AM. ACAD. CHILD ADOLESC. PSYCHIATRY, 44:11, NOVEMBER 2005 1147

vein to facilitate plasmapheresis. Automated plasma-pheresis was performed with the COBE Spectra cell sep-arator, with 5% albumin as the replacement fluid andno red cell priming. A course of five plasmapheresis pro-cedures over 10 days was carried out, the first treating1.3 plasma volumes and the rest treating 1.0 volumes.No sedation or physical restraints were used, but 1:1supervision and vigilant protection of the access siteby the plasmapheresis operator were necessary, espe-cially for the first procedure.

RESULTS

Within 24 hours of starting lorazepam, the ‘‘freezing’’and ‘‘staring’’ episodes resolved, expressive language im-proved, andmotor hyperactivity, impulsivity, and inatten-tion decreased. The first plasmapheresis treatment wasdifficult because of the boy’s disruptive behavior. Imme-diately after the procedure, brief periods of increased co-operation, compliance with instructions, and improvedspeech were noted.The patient was generally cooperative with the sec-

ond plasmapheresis treatment, although some disrupt-ive behavior persisted. After the treatment, he was ableto follow three-step commands and accept redirection.His mood was stable and there was no lability. Somehyperactivity persisted, but his movements were slowerand less impulsive and unpredictable. His speech wasfluent, at a normal volume, and easily understood.One-to-one supervision was no longer necessary at alltimes.Improvement continued with the subsequent third,

fourth, and fifth treatments. By the end of the fifth treat-ment, his overall activity level was much reduced,speech was nearly normal, interactions with others wereappropriate, and his mood was stable. One-to-one su-pervision was discontinued. Plasmapheresis was gener-ally well tolerated. Mild leg cramps were noted after thefirst and second treatments; abdominal pain, headache,and emesis occurred 3 days after the third treatment.The second MRI completed 2 days after initiating

lorazepam treatment showed a slight reduction in basalganglia swelling but no significant overall change. Athird MRI completed 12 days later, after the fifth plas-mapheresis treatment, showed marked reduction inswelling (46% reduction in the size of right caudate,31% left caudate, 27% right putamen, and 33% leftputamen) and a reduction in the hyperintensity seen

on T1-weighted images and FLAIR in the involvedstructures (Table 1, Fig. 1C). The basal ganglia andventricles appeared to have returned to normal size.

The patient was discharged home 10 days afterinitiation of plasmapheresis. The catatonia was fullyresolved, and he no longer met criteria for ADHDor OCD on K-SADS. He scored 0 on Children’sYale-Brown Obsessive-Compulsive Scale and on theNIMH emotional lability scale. Discharge medica-tions were amoxicillin 250 mg twice daily and taper-ing doses of lorazepam. Within 3 weeks, he was backat school full-time, and 4 months later, he made thehonor roll.

DISCUSSION

The acute onset of neuropsychiatric symptoms aftera fever, cough, otitis, and sinusitis, with concomitantASO and anti-DNAse B titer elevations, suggests groupA b-hemolytic streptococci involvement. Acute dissem-inated encephalomyelitis (Dale et al., 2001), anotherpoststreptococcal autoimmune spectrum disorder, wasexcluded because of the absence of white matter lesions.

Catatonia, a nonspecific symptom (e.g., fever), hasnot been reported to occur in PANDAS. Our patienthad commonly recognized manifestations of catatonia,including somnolence, posturing, staring, and expres-sive language difficulties. Cognition and receptive lan-guage were not affected, ruling out the possibility ofdelirium. We cannot explain the elevated serum ammo-nia level. A canine model of hyperammonemia, inwhich all seven beagles with a portocaval shunt becamelistless and four developed catatonia, suggests that itcould play a role (Watanabe et al., 1997); however, cat-atonia has also been reported in a case with documentednormal ammonia levels (Tsai et al., 2003). Our patientwas also hyperactive, inattentive, and impulsive. Weinitially attributed the excessive activity to ADHD(running/climbing and often ‘‘on the go’’ being thetwo preexisting hyperactive symptoms), but in retrospect,this was most likely a manifestation of catatonia.ADHD symptoms have been associated with PANDAS(Perlmutter et al., 1998, 1999; Swedo et al., 1998;Waldrep, 2002). Of the NIMH case series (Swedoet al., 1998), 50% presented with motor hyperactivity,and in 40% symptoms were severe enough to warrant anADHD diagnoses; however, the significance of comor-bid ADHD symptoms is unclear and potentially not

ELIA ET AL.

1148 J . AM. ACAD. CHILD ADOLESC. PSYCHIATRY, 44:11, NOVEMBER 2005

ADHD at all. Peterson et al. (2000) demonstrated a sig-nificant association between ASO and DNAase B anti-bodies and ADHD (in comparison with chronic ticdisorders and OCD) in patients who were unselectedwith regard to history for streptococcal exposure. Beforethe current illness, our patient did not meet criteria forany psychiatric disorder but had subclinical, nonimpair-ing ADHD symptoms. After the onset of PANDAS, theADHD symptom count increased; however, the rapiddecrease with lorazepam treatment led us to reconsiderthe increased symptom count as a manifestation ofcatatonia.This is the second case report documenting basal gan-

glia volumetric changes corresponding with improve-ment possibly caused by plasmapheresis. In the firstcase (Giedd et al., 1996), a 12 year old with PANDAShad significant volumetric MRI changes after plasma-pheresis (decrease of 24% caudate, 28% globus pallidus,12% putamen). Improvement in OCD and tics oc-curred concomitantly with volumetric changes. A studycomparing basal ganglia volumes in 34 children withPANDAS and 82 age- and sex-matched controls dem-onstrated increases in caudate volume (mean 8%), puta-men (mean 5%), and globus pallidus (mean 7%) in theaffected patients (Giedd et al., 2000). However, theseMRI changes of the PANDAS cases were not judgedclinically significant by a neuroradiologist. This suggeststhat there may be a spectrum of severity in CNS involve-ment resulting from the autoimmune process. Plasma-pheresis was considered to be effective in our case andthe case of Giedd et al. (1996), both with clinical basalganglia abnormalities.Lorazepam, started 5 days before beginning plasma-

pheresis, resulted in marked improvement of posturing,staring, expressive language, and motor activity but didnot decrease urgency and compulsion to void or moodlability, which dramatically and immediately improvedwith plasmapheresis. However, we cannot conclude thatimprovement resulted solely from plasmapheresis, giventhat lorazepam was continued, albeit at lower doses,after plasmapheresis was started.The obsessive urgency and compulsion to void, not

common OCD symptoms, were not initially inter-preted as neuropsychiatric manifestations in our pa-tient, leading to multiple tests and hospitalizations.These have also been noted in 58% of cases reportedby Murphy and Pichichero (2002), suggesting thatthese may have some particular relevance to PANDAS.

The role of the immune system in PANDAS is notclearly understood. IgA dysgammaglobulinemia hasbeen reported in 13 patients with Tourette disorderand OCD (Hansen and Bershow, 1997). Quantitativeimmunoglobulins were not obtained from our patient,although his medical history suggested an overactivesystem. Three months before the onset of the currentillness, he had an anaphylactic reaction to a beesting,consisting of wheezing and swelling of lips and facewithin 1 hour of being stung. He had received approx-imately eight desensitization treatments before the onsetof PANDAS. There is a previous report of bilateralpallidal/striatal necrosis and neuropsychiatric symp-toms in an adult that occurred after a beesting (Laplaneet al., 1981), but it is doubtful that this was relevantin our patient’s case.The rapidity of the deterioration in function and the

severity of the neuropsychiatric symptoms in this pa-tient led to the use of plasmapheresis, an interventionbased on only a few case reports (Giedd et al., 1996;Perlmutter et al., 1999; Tucker et al., 1996) based onthe hypothesis that symptom reduction results fromremoval of the antistreptococcal antibodies thought tocross-react with neural tissue (Husby et al., 1976; Swedoet al., 1994). The clinical consequences of treating ornot treating are not fully known. Controlled studies de-fining appropriate diagnostic tests, evaluating the safetyand efficacy of treatments, and identifying predictors oftreatment response are necessary.

Disclosure: The authors have no financial relationships to disclose.

REFERENCES

Ambrosini PJ (2000), Historical development and present status of theschedule for affective disorders and schizophrenia for school-age children(K-SADS). J Am Acad Child Adolesc Psychiatry 39:49–58

Bush G, Fink M, Petrides G, Dowling F, Francis A (1996), Catatonia. II.Treatment with lorazepam and electroconvulsive therapy. Acta PsychiatrScand 93:137–143

Dale RC, Church AJ, Cardoso F, Goddard E, Cox TC, Chong WK et al.(2001), Poststreptococcal acute disseminated encephalomyelitis withbasal ganglia involvement and auto-reactive antibasal ganglia antibodies.Ann Neurol 50:588–595

Gelenberg S (1976), The catatonic syndrome. Lancet 1:1339–1341Giedd JN, Rapoport JL, Garvey MA, Perlmutter S, Swedo SE (2000), MRI

assessment of children with obsessive-compulsive disorder or tics associ-ated with streptococcal infection. Am J Psychiatry 157:281–283

Giedd JN, Rapoport JL, Leonard HL, Richter D, Swedo SE (1996), Casestudy: acute basal ganglia enlargement and obsessive-compulsive symp-toms in an adolescent boy. J Am Acad Child Adolesc Psychiatry 35:913–915

Hansen CR Jr, Bershow SA (1997), Immunology of TS/OCD. J Am AcadChild Adolesc Psychiatry 36:1648–1649

PANDAS WITH CATATONIA

J. AM. ACAD. CHILD ADOLESC. PSYCHIATRY, 44:11, NOVEMBER 2005 1149

Husby G, van de Rijn I, Zabriskie JB, Abdin ZH, Williams RC Jr (1976),Antibodies reacting with cytoplasm of subthalamic and caudate nucleineurons in chorea and acute rheumatic fever. J Exp Med 144:1094–1110

LaplaneD,Widlocher D, Pillon B, BaulacM, Binoux F (1981), Obsessional-type compulsive behavior caused by bilateral circumscribed pallidostria-tal necrosis. Encephalopathy caused by a wasp sting [in French]. RevNeurol (Paris) 137:269–276

Murphy DL, Pichichero ME (2002), Prospective identification and treat-ment of children with pediatric autoimmune neuropsychiatric disorderassociated with group A streptococcal infection (PANDAS). Arch PediatrAdolesc Med 156:356–361

Northoff G (2002), Catatonia and neuroleptic malignant syndrome: psy-chopathology and pathophysiology. J Neural Transm 109:1453–1467

Northoff G, Steinke R, Czcervenka C, Krause R, Ulrich S, Danos P, Kropf D,Otto H, Bogerts B (1999), Decreased density of GABA-A receptors in theleft sensorimotor cortex in akinetic catatonia: investigation of in vivo ben-zodiazepine receptor binding. J Neurol Neurosurg Psychiatry 67:445–450

Northoff G, Wenke J, Demisch L, Eckert J, Gille B, Pflug B (1995), Cat-atonia: short-term response to lorazepam and dopaminergic metabolism.Psychopharmacology (Berl) 122:182–186

Perlmutter SJ, Garvey MA, Castellanos X, Mittleman BB, Giedd J,Rapoport JL et al. (1998), A case of pediatric autoimmune neuro-psychiatric disorders associated with streptococcal infections. Am JPsychiatry 155:1592–1598

Perlmutter SJ, Leitman SF, Garvey MA, Hamburger S, Feldman E,Leonard HL et al. (1999), Therapeutic plasma exchange and intravenousimmunoglobulin for obsessive-compulsive disorder and tic disorders inchildhood. Lancet 354:1153–1158

Peterson BS, Leckman JF, Tucker D, Scahill L, Staib L, Zhang H et al.(2000), Preliminary findings of antistreptococcal antibody titers andbasal ganglia volumes in tic, obsessive-compulsive, and attention defi-cit/hyperactivity disorders. Arch Gen Psychiatry 57:364–372

Scahill L, Riddle MA, McSwiggin-Hardin M, Ort SI, King RA,GoodmanWK et al. (1997), Children’s Yale-BrownObsessive CompulsiveScale: reliability and validity. J Am Acad Child Adolesc Psychiatry 36:844–852

Swedo SE, Leonard HL, Garvey M, Mittleman B, Allen AJ, Perlmutter Set al. (1998), Pediatric autoimmune neuropsychiatric disorders associ-ated with streptococcal infections: clinical description of the first 50 cases.Am J Psychiatry 155:264–271

Swedo SE, Leonard HL, Kiessling LS (1994), Speculations on antineuronalantibody-mediated neuropsychiatric disorders of childhood. Pediatrics93:323–326

Tsai H, Kuo P, Yang P (2003), Fever, consciousness disturbance, and musclerigidity in a 68-year-old man with depressive disorder. Chest 124:1598–1601

Tucker DM, Leckman JF, Scahill L, Wilf GE, LaCamera R, Cardona L et al.(1996), A putative poststreptococcal case of OCD with chronic tic dis-order, not otherwise specified. J Am Acad Child Adolesc Psychiatry35:1684–1691

Waldrep DA (2002), Two cases of ADHD following GABHS infection:a PANDAS subgroup? J Am Acad Child Adolesc Psychiatry 41:1273–1274

Watanabe M, Ozaki T, Mushiroi T, Ukai Y, Ueda F, Kimura K, Itoh et al.(1997), Behavioral and electroencephalographic studies of beagles withan Eck’s fistula: suitability as a model of hepatic encephalopathy. Phar-macol Biochem Behav 57:367–375

ELIA ET AL.

1150 J . AM. ACAD. CHILD ADOLESC. PSYCHIATRY, 44:11, NOVEMBER 2005