Original Article

Journal of Child Neurology

2018, Vol. 33(2) 140-145
Therapeutic Plasma Exchange Use in Pediatric ª The Author(s) 2017
Reprints and permission:
sagepub.com/journalsPermissions.nav
Neurologic Disorders at a Tertiary Care DOI: 10.1177/0883073817749368
journals.sagepub.com/home/jcn
Center: A 10-Year Review

Sonika Agarwal, MD1, Jake R. Keller, BS2, Chloe E. Nunneley, MD2,

Eyal Muscal, MD, MS1,3, Michael C. Braun, MD4,
Poyyapakkam Srivaths, MD4, and Timothy E. Lotze, MD1

Abstract
Pediatric neurologic conditions requiring therapeutic plasma exchange are rare in children and literature is sparse. The study aims
to determine the outcomes, safety, and feasibility of therapeutic plasma exchange treatment in pediatric neurologic disorders.
This retrospective analysis looked at the outcomes and safety of therapeutic plasma exchange in children (n ¼ 50) with neurologic
conditions. Patient age ranged <1 to 19 years old with a mean of 10.35 years. Of the 50 children treated with plasmapheresis, 26
patients received inpatient rehabilitation. At discharge, functional status can be summarized as follows: 24 (48%) with mental
status impairment, 10 (20%) with vision impairment, 19 (38%) with bladder incontinence, and 37 (74%) with motor impairment.
Three-month follow-up: 30% with mental status impairment, 10% with vision impairment, 18% with bladder incontinence, and
52% with motor impairment. Therapeutic plasma exchange is an effective and safe therapy for neurological conditions in the
pediatric population.

Keywords
pediatric, disability, neuroimmunology, rehabilitation, outcome, autoimmune

Received July 27, 2017. Received revised November 18, 2017. Accepted for publication November 28, 2017.

Introduction of the procedure and concern for complications has led TPE to
more often be considered a second-line treatment when other
Therapeutic plasma exchange (TPE) or plasmapheresis is a
immunomodulatory treatments such as steroids and intrave-
procedure that involves exchanging the patient’s plasma with nous immunoglobulin (IVIG) have not shown improvement
either fresh-frozen plasma or albumin solution and is an estab-
in disease symptoms.
lished treatment for a variety of neurologic disorders in adults
Information on acute and long-term outcomes is sparse.
and children.1 The therapeutic effects of TPE in neurologic
We describe the largest single-cohort of pediatric patients
disorders are presumed to be through removal of pathologic
treated with TPE for neuroinflammatory disease over a
substances from the blood (auto-antibodies, cytokines, and
10-year period. Our study describes the safety and
monoclonal paraproteins), replacement of deficient plasma
components, and potential alterations in lymphocyte prolifera-
tion and function that could facilitate the response to other
immunosuppressant and chemotherapeutic agents.2 1
Division of Neurology & Developmental Neuroscience, Baylor College of
TPE is an established first-line therapy for Guillian-Barré Medicine, Texas Children’s Hospital, Houston, TX, USA
2
syndrome and myasthenia gravis. Published individual case Baylor College of Medicine, Houston, TX, USA
3
Division of Immunology, Allergy And Rheumatology, Baylor College of
reports, small case series, and expert opinion papers and Medicine, Texas Children’s Hospital, Houston, TX, USA
best-care practices advocate for the use of TPE in other neu- 4
Division of Nephrology, Baylor College of Medicine, Texas Children’s
roinflammatory conditions, such as acquired demyelinating Hospital, Houston, TX, USA
diseases, severe neuropsychiatric lupus (NPSLE), and NMDA
receptor antibody-mediated encephalitis.3-8 The overall use of Corresponding Author:
Sonika Agarwal, MD, Section of Pediatric Neurology and Developmental
TPE in pediatric populations is limited because of the relative Neuroscience, Nemours A.I. duPont Hospital for Children, 1600 Rockland Rd,
rarity of these diseases and the need for specialized multidisci- Wilmington, DE 19803, USA.
plinary care for these patients. In addition, the invasive nature Emails: [email protected]; [email protected]
Agarwal et al 141

feasibility of TPE as well as acute and long-term outcomes Table 1. Summary of Patient Demographics.
in this population.
Age when TPE (years) administered
Range 0.74-19.7
Methods and Design Mean (SD) 10.35 (5.6)
Diagnosis, n (%)
This was a 10-year retrospective descriptive study of clinical and Acute motor axonal neuropathy 1 (2)
safety outcomes of TPE in children with immune-mediated neurologic ADEM 5 (10)
conditions. The medical records of all children treated with TPE from Autoimmune, NOS 7 (14)
2007 to 2016 were reviewed. Those children undergoing TPE for an CIS 3 (6)
acute neurologic condition were selected for further analysis. GBS 6 (12)
Summary statistics were used to describe the patient characteris- Multiple sclerosis 1 (2)
tics in the study sample. For continuous data, we calculated the NMDAR 6 (12)
range, mean, and standard deviation. For categorical variables, NMO 9 (18)
number and percentages in each study group were calculated. The Transverse myelitis 10 (20)
demographics of this patient population and use of concomitant Tumefactive demyelination 1 (2)
disease-modifying and symptom control treatments (before and after Isaac syndrome 1 (2)
Number of TPE cycles
TPE) was collected. Any direct complications of TPE reported in the
Range 3-42
medical record were recorded.
Mean (SD) 6.80 (5.80)
Length of stay outcomes were compared among various neurologic
Median 6
conditions by calculating the mean and standard deviation for duration of Treatment before TPE, n (%)a
hospitalization and requirement for and length of inpatient rehabilitation. Steroid 40 (80)
Neurologic disability was based on recorded mental status impair- IVIG 21 (42)
ment, vision impairment, bladder impairment, and motor impairment. Rituximab 5 (10)
Disability outcomes posttreatment were studied based on disability at Mycophenolate 2 (4)
discharge, at first clinic follow-up, and at last follow-up visit Cyclophosphamide 1 (2)
abstracted from the medical record. None 2 (4)
All TPE procedures were performed using centrifugation with Spec- Treatment after TPE, n (%)
tra OPTIA and COBE Spectra (Terumo BCT, Lakewood, CO). Steroid 42 (88)
Regional anticoagulation was instituted using 3% citrate solution and IVIG 38 (79)
an anticoagulation ratio of 14:1. Intravenous calcium (2.16 mg/mL Rituximab 18 (38)
elemental calcium) replacement was initiated at the start of all proce- Mycophenolate 16 (33)
dures. Replacement fluid for TPE procedures included 5% albumin, Cyclophosphamide 13 (27)
fresh-frozen plasma, and a combination of both 5% albumin and fresh- Azathioprine 1 (2)
frozen plasma. One to 1.5 plasma volume exchanges were done for TPE None (palliative) 1 (2)
procedures; typically procedures were planned daily or every other day. Complications, n (%)b
Because of the diverse group of clinical conditions looked at, the None 34 (72)
study population was heterogeneous and the decision for plasmapher- Hypotension 3 (6)
Neuropathic pain/paresthesias 3 (6)
esis and timing of the treatment was based on clinical assessment and
Transfusion reaction to FFP 3 (6)
disease course. There was variability in the time to first cycle of
Mild intermittent facial swelling 1 (1)
plasmapheresis and repeat cycles based on the clinical decision and
judgment of the treating physician. Each cycle included 5 treatments Abbreviations: ADEM, acute disseminated encephalomyelitis; CIS, clinically
with TPE over a period of 7 to 10 days. isolated syndrome; FFP, fresh-frozen plasma; GBS, Guillain-Barré syndrome;
IVIG, intravenous immunoglobulin; NMDAR, anti-NMDA receptor
encephalitis; NMO, neuromyelitis optica; Autoimmune, NOS, autoimmune
Results phenotype without antibody–not otherwise specified.
a
Known patient treatment, n ¼ 48.
b
A total of 215 children received TPE treatment from 2007 to Known patient complications, n ¼ 44.
2016. Of these, 50 were identified with an acute neurologic
condition as the principal indication for TPE. Neurologic con- follow-up was at 1 to 12 months after discharge (mean
ditions included acute disseminated encephalomyelitis 3 months). The mean duration of follow-up was 38 months
(ADEM), transverse myelitis, neuromyelitis optica (NMO), (range 18-53 months).
N-methyl-D-aspartate (NMDA) receptor encephalitis, Guil- Table 1 shows details of age and number of cycles of TPE,
lain-Barré syndrome, tumefactive demyelination, multiple concomitant treatments, and complications attributed to the plas-
sclerosis, Isaac syndrome, clinically isolated syndrome (CIS), mapheresis. Patient ages ranged from as young as 8 months to 19
and autoimmune disorder–not otherwise specified (autoim- years old with a mean of 10.35 years. Gender distribution was
mune phenotype without antibody). The study population was equal (males 25, females 25). Transverse myelitis (n ¼ 10) and
heterogeneous and the number of TPE cycles ranged from 3 to NMO (n ¼ 9) were the largest categories comprising more than a
42 (mean 6.8). Because of the small and variable number of third of the patient population undergoing treatment with TPE.
patients in each disease category, the relationship between dis- Other neurologic disorders in the study group included NMDA
ease and number of cycles could not be studied. The first clinic receptor antibody–mediated encephalitis (n ¼ 6), Guillain-Barré
142 Journal of Child Neurology 33(2)

Table 2. Summary of Inpatient Outcomes by Diagnostic Category.

Autoimmune, Transverse
ADEM NOS CIS GBS NMDAR NMOa myelitisb Otherc Total

n 5 7 3 6 6 9 10 4 50
Age, y, mean (SD) 6.0 (2.9) 14.1 (4.9) 15.5 (1.8) 8.46 (4.8) 9.5 (5.4) 13.8 (3.8) 7.3 (6.2) 12.4 (6.2) 10.4 (5.6)
Duration of hospitalization, 33.2 (17.5) 35.8 (22.2) 53.3 (68.1) 43.8 (22.8) 58.5 (40.7) 22.9 (17.2) 33.4 (16.9) 35.5 (28.1) 37.6 (28.2)
d, mean (SD)
Inpatient rehabilitation, n (%) 4 (80) 2 (28.6) 1 (33.3) 4 (66.7) 4 (66.7) 3 (33.3) 6 (60) 2 (50) 26 (52)
Duration of inpatient 24.5 (14.6) 22.2 (9.9) 35 (0) 40.5 (19.3) 36.8 (13.2) 22.0 (14.9) 21.2 (11.2) 41.5 (6.4) 29.6 (14.6)
rehabilitation, d, mean (SD)
Disability at discharge, n (%)
Mental status impairment 4 (80) 6 (86) 1 (33.3) 1 (16.7) 6 (100) 2 (22.2) 1 (10) 3 (75) 24 (48)
Vision impairment 2 (40) 0 (0) 2 (66.6) 0 (0) 0 (0) 5 (55.6) 0 (0) 1 (25) 10 (20)
Bladder impairment 0 (0) 1 (14) 1 (33.3) 0 (0) 4 (66.7) 5 (55.6) 6 (60) 2 (50) 19 (38)
Motor impairment 4 (80) 4 (57) 3 (100) 3 (50) 6 (100) 4 (44.4) 9 (90) 4 (100) 37 (74)
Abbreviations: ADEM, acute disseminated encephalomyelitis; Autoimmune, NOS, autoimmune phenotype without antibody–not otherwise specified; CIS,
clinically isolated syndrome; GBS, Guillain-Barré syndrome; NMDAR, anti-NMDA receptor encephalitis; NMO, neuromyelitis optica.
a
One patient did not have complete hospital records; did not contribute to calculated values.
b
One patient did not have complete hospital records; did not contribute to calculated values.
c
Acute motor axonal neuropathy, tumefactive demyelination, Isaac syndrome, multiple sclerosis (only 1 patient for each separate diagnosis).

syndrome (n ¼ 6), autoimmune encephalitis–not otherwise Table 3 summarizes the cohort characteristics at early and
specified (n ¼ 7), ADEM (n ¼ 5), clinically isolated syn- last clinic follow-up visit. At the first clinic follow-up, 41
drome (n ¼ 3), acute motor axonal neuropathy (n ¼ 1), tume- patients had improvement in functional status compared to
factive demyelination (n ¼ 1), and Isaac syndrome (n ¼ 1). discharge: 15 (30%) with mental status impairment, 5 (10%)
Before TPE, 40 of 50 patients (80%) received concomitant with vision impairment, 9 (18%) with bladder incontinence,
treatment with intravenous methylprednisolone (30 mg/kg/ and 26 (52%) with motor impairment. Disability at last office
dose, maximum 1 g for 5 doses), 21 of 50 (42%) received visit can be summarized as 13 (26%) with mental status impair-
treatment with IVIG (2 g/kg), and 19 of 50 (38%) received ment, 4 (8%) with vision impairment, 8 (16%) with bladder
both (Table 1). After TPE was completed, 42 patients incontinence, and 15 (30%) with motor impairment. Sixteen
received additional intravenous methylprednisolone. Like- (32%) of patients were continuing to receive disease-
wise, 21 of the 22 pre-TPE IVIG group received additional modifying drugs: 7 mycophenolate mofetil, 1 azathioprine, 3
IVIG treatment after TPE, with 2 patients having initial dos- rituximab every 6 months, 5 monthly IVIG.
ing of IVIG after TPE. Corticosteroids (88%) and IVIG (79%)
were the most frequently continued treatments after TPE.
Other treatments started after completion of TPE included Discussion
rituximab (38%), cyclophosphamide (33%), mycophenolate The most common standard practice treatments for the acute
mofetil (27%), and azathioprine (2%). management of neuroimmune disorders in the pediatric neuro-
Procedure-related complications of TPE were hypotension logic population are immunomodulatory or immunosuppres-
(6%), pain/paresthesias (6%), transfusion reaction (6%), and sive: corticosteroids, IVIG, and TPE. 9-12 Because of the
facial swelling (1%). All these complications were self- adverse effects and expense involved with IVIG and TPE treat-
resolving over a period of 1-2 days. The complications were ments, there is no standard protocol or guideline that is fol-
not related to number of cycles of plasmapheresis. lowed uniformly when managing these disorders, except for
Table 2 shows length of stay and discharge outcomes. The Guillain-Barré syndrome and myasthenic gravis. Studies con-
mean length of stay (LOS) was 37.6 + 28.2 days. Fifty-two cerning TPE in the pediatric cohort are lacking, and most are
percent of patients required inpatient rehabilitation. Mean case series of small numbers. In a case series of 5 patients with
length of stay for inpatient rehabilitation was 29.6 + 14.6 ADEM, plasmapheresis was found to be a good therapeutic
days (range 21.2-40.5 days). At discharge, 24 (48%) had men- option and safe when performed in an experienced center.7 A
tal status impairment, 10 (20%) had vision impairment, 19 retrospective study of children with Guillain-Barré syndrome
(38%) had bladder incontinence, and 37 (74%) had motor found TPE to be the best treatment modality and was seen to
impairment (Table 2). reduce the duration of hospitalization and hasten recovery.4
Figure 1 shows the difference in LOS as well as length of TPE was also found to be effective in severe pediatric central
inpatient rehabilitation by diagnostic category. Patients with nervous system demyelination in a retrospective analysis
NMDA receptor encephalitis had the longest acute hospital involving 12 children.6
LOS (mean ¼ 58.5 + 40.7 days), whereas NMO patients had This single tertiary care center retrospective descriptive
the shortest LOS (mean ¼ 22.9 + 17.9 days). study identified the safety, feasibility, and outcomes of TPE
Agarwal et al 143

ADEM
Autoimmune NOS
CIS
GBS
NMDAR
NMO
Transverse Myelitis
Diagnosis

Other

ADEM
Autoimmune NOS Hospitalization
CIS Inpatient Rehabilitation
GBS
NMDAR
NMO
Transverse Myelitis
Other
0

5
10

15

20

25

30

35

40

45

50

55

60

65

70

75
Days

Figure 1. Duration of hospitalization and inpatient rehabilitation.

Table 3. Summary of Outpatient Outcomes by Diagnostic Category.

Autoimmune, Transverse
ADEM NOS CIS GBS NMDAR NMOa myelitisb Otherc Total

n 5 7 3 6 6 9 10 4 50
Age, y, mean (SD) 6.0 (2.9) 14.1 (4.9) 15.5 (1.8) 8.46 (4.8) 9.5 (5.4) 13.8 (3.8) 7.3 (6.2) 12.4 (6.2) 10.4 (5.6)
Seen for first follow-up visit, n (%) 4 (80) 5 (71.4) 3 100) 4 (66.7) 5 (83.3) 7 77.8) 9 (90) 3 (75) 40 (80)
Disability at first follow-up visit, n (%)
Mental status impairment 1 (20) 4 (57.1) 2 (66.7) 0 (0) 5 (83.3) 0 (0) 2 (20) 1 (25) 15 (30)
Vision impairment 1 (20) 0 (0) 1 (33.3) 0 (0) 0 (0) 3 (33.3) 0 (0) 0 (0) 5 (10)
Bladder impairment 0 (0) 1 (14.2) 0 (0) 0 (0) 1 (16.7) 2 (22.2) 5 (50) 0 (0) 9 (18)
Motor impairment 2 (40) 2 (28.6) 2 (66.7) 3 (50) 3 (50) 4 (44.4) 8 (80) 2 (50) 26 (52)
Seen for last office visit, n (%) 4 (80) 5 (71.4) 3 (100) 5 (83.3) 4 (66.7) 7 (77.8) 10 (100) 3 (75) 41 (82)
Disability at last office visit, n (%)
Mental status impairment 1 (20) 3 (42.9) 2 (66.7) 0 (0) 2 (33.3) 1 (11.1) 3 (30) 1 (25) 13 (26)
Vision impairment 0 (0) 0 (0) 1 (33.3) 0 (0) 0 (0) 3 (33.3) 0 (0) 0 (0) 4 (8)
Bladder impairment 0 (0) 0 (0) 0 (0) 1 (16.7) 0 (0) 2 (22.2) 5 (50) 0 (0) 8 (16)
Motor impairment 0 (0) 0 (0) 1 (33.3) 3 (50) 1 (16.7) 4 (44.4) 5 (50) 1 (25) 15 (30)
Taking disease-modifying therapies,d 0 (0) 4 (57.1) 1 (33.3) 0 (0) 2 (33.3) 7 (77.8) 0 (0) 2 (50) 16 (32)
n (%)

Abbreviations: ADEM, acute disseminated encephalomyelitis; Autoimmune, NOS, autoimmune phenotype without antibody–not otherwise specified; CIS,
clinically isolated syndrome; GBS, Guillain-Barré syndrome; NMDAR, anti-NMDA receptor encephalitis; NMO, neuromyelitis optica.
a
One patient did not have complete hospital records; did not contribute to calculated values.
b
One patient did not have complete hospital records; did not contribute to calculated values.
c
Acute motor axonal neuropathy, tumefactive demyelination, Isaac syndrome, multiple sclerosis (only 1 patient for each separate diagnosis).
d
Disease-modifying therapies include azathioprine, intravenous immunoglobulin, solumedrol, mycophenolate mofetil, rituximab.

treatment in a large pediatric cohort with diverse neurologic length of hospital stay was lower than other disease groups,
conditions. Our cohort included 3 children less than 1 year of and it may be inferred that TPE may add benefit in patients
age and 12 in the age group of 2-5 years. Our youngest patient with a more severe and fulminant presentation when steroids or
was 8 months old, suggesting that the procedure may be per- IVIG do not show the expected response.
formed safely in younger populations. TPE has also shown to be of benefit in a case series (n ¼ 6)
This study also shows the successful use of TPE to treat a of NMDA receptor encephalitis with a reported inpatient and
broader range of pediatric neuroimmune disorders. In children rehabilitation period of 2-9 weeks.12 Our data were comparable
with ADEM and central nervous system inflammatory demye- in this disease group with a prolonged hospitalization and
lination (neuromyelitis optica and transverse myelitis), the rehabilitation period. In patients with NMDA receptor
144 Journal of Child Neurology 33(2)

encephalitis, the mental status and psychiatric disturbances diverse group of neurologic disorders, it is difficult to make
are often seen to have an impact on the rehabilitation process conclusions about the efficacy of various treatment methods.
and improvement. Also, in severe debilitating transverse mye- Outcomes of TPE treatment looking at safety and efficacy may
litis, the initial improvement may be slow and thus the reha- be measured better through larger multicenter trials. Future
bilitation process may be longer as compared to other prospective studies should center upon specific outcomes with
neurologic conditions. use of standardized scales to measure pre- and post-treatment
The short-term (20%-74%) and long-term (8%-30%) dis- functional status. Such trials may facilitate timely initiation of
ability outcomes of our patients indicate a continued improve- TPE when indicated and help overcome the perceptions of this
ment over time. Predictors of outcome may relate to several procedure being associated with increased morbidity and
factors: level of disability at nadir, time from clinical onset to adverse effects.
treatment initiation, and duration of hospitalization. Overall,
our data indicate a favorable long-term outcome with continued Author Contributions
improvement after discharge.
SA contributed to acquisition, analysis and interpretation of data,
The complication rate in our study cohort was minimal, and
drafted manuscript, critically revised the manuscript and acted as
TPE was safely tolerated in children in the 0-5-year age group corresponding author. TEL contributed to acquisition, analysis and
(n ¼ 15). In the literature, it has been reported that increased interpretation of data, critically revised the manuscript and gave final
rates of adverse events occur in children compared to adults approval. JRK, CEN contributed to acquisition and analysis of data
when undergoing TPE.13 The usual side effects associated with and reviewed manuscript for final approval. EM contributed to acqui-
TPE are hypotension, anemia, allergic reactions to fresh-frozen sition, analysis and interpretation of data, critically revised the manu-
plasma, hypocalcemia, and line-related thrombosis. The severe script and gave final approval. MCB and PS contributed to acquisition,
complications are mainly related to the presence of central analysis and interpretation of data, critically revised the manuscript
venous catheter and disorders of hemostasis due to anticoagu- and gave final approval.
lant therapy, and hypotension. In our study group, adverse
reactions were mild and none of the children required any Declaration of Conflicting Interests
interventions. However, it is important to note that therapeutic The authors declared no potential conflicts of interest with respect to
plasma exchange is an invasive procedure and, rarely, may lead the research, authorship, and/or publication of this article.
to severe complications. The safety and outcome of this proce-
dure are largely dependent on the availability of multidisciplin- Funding
ary teams (nephrologist, pediatric neurologist, intensive care The authors received no financial support for the research, authorship,
physicians, transfusion medicine teams, and the support staff). and/or publication of this article.
It is also important to consider that the cumulative experience
of the treating center, frequency of exchanges, age and baseline Ethical Approval
status of the patient, and the specific neurologic disease may
have an impact on the clinical and safety outcome. The study was approved by the institutional review board of Baylor
The strength of our retrospective analysis includes a large College of Medicine/Texas Children’s Hospital (IRB H-38483).
cohort highlighting the safety and feasibility of TPE in children
and adolescents. Our study is the single largest cohort reporting References
short-term and long-term outcomes with TPE use in a diverse 1. Cortese I, Cornblath DR. Therapeutic plasma exchange in neurol-
group of pediatric neurologic conditions. Our study was limited ogy: 2012. J Clin Apher. 2013;28:16-19.
by its retrospective design with nonstandardized treatment and 2. Reeves HM, Winters JL. The mechanisms of action of plasma
outcome measure usage across the study population. Thus, the exchange. Br J Haematol. 2014;164:342-351.
individual patients had variability in the use of concomitant 3. Mühlhausen J, Kitze B, Huppke P, Müller GA, Koziolek MJ.
treatments, such as steroids and IVIG, making it difficult to Apheresis in treatment of acute inflammatory demyelinating
assess or compare absolute efficacy of TPE to other treatment disorders. Atheroscler Suppl. 2015;18:251-256.
measures. Because these disease entities are rare, only large 4. Saad K, Mohamad IL, Abd El-Hamed MA, et al. A comparison
multicenter, prospective observational (comparative effective- between plasmapheresis and intravenous immunoglobulin in
ness) or randomized controlled trials may be able to compare children with Guillain-Barré syndrome in Upper Egypt. Ther Adv
TPE to other therapeutic options. Neurol Disord. 2016;9:3-8.
5. Liew WK, Powell CA, Sloan SR, et al. Comparison of plasmapheresis
and intravenous immunoglobulin as maintenance therapies for juve-
Conclusion nile myasthenia gravis. JAMA Neurol. 2014;71:575-580.
Our data are comparable in outcomes to other studies, and is 6. Bigi S, Banwell B, Yeh EA. Outcomes after early administration
the largest cohort of children with neurologic disorders under- of plasma exchange in pediatric central nervous system inflam-
going TPE. Our retrospective study supports the safety and matory demyelination. J Child Neurol. 2015;30:874-880.
feasibility of TPE, with good short- and long-term outcomes. 7. Borras-Novell C, Garcı́a Rey E, Perez Baena LF, Jordan Garcia I,
Because of heterogeneity in assessment and follow-up of a Catella Cahiz D, Cambra F. Therapeutic plasma exchange in
Agarwal et al 145

acute disseminated encephalomyelitis in children. J Clin Apher. 10. Gajdos P, Chevret S, Toyka KV. Intravenous immunoglobulin for
2015;30:335-339. myasthenia gravis. Cochrane Database Syst Rev. 2012;12:
8. Houtrow AJ, Bhandal M, Pratini NR, Davidson L, Neufeld JA. CD002277.
The rehabilitation of children with anti-N-methyl-D-aspartate- 11. McMillan HJ, Darras BT, Kang PB. Autoimmune neuromuscular
receptor encephalitis: a case series. Am J Phys Med Rehabil. disorders in childhood. Curr Treat Options Neurol. 2011;13:590-607.
2012;91:435-441. 12. Houtrow AJ, Bhandal M, Pratini NR, Davidson L, Neufeld JA.
9. Cortese I, Chaudhry V, So YT, Cantor F, Cornblath DR, Rae- The rehabilitation of children with anti-N-methyl-D-aspartate-
Grant A. Evidence-based guideline update: plasmapheresis in receptor encephalitis: a case series. Am J Phys Med Rehabil.
neurologic disorders: report of the Therapeutics and Technology 2012;91:435-441.
Assessment Subcommittee of the American Academy of Neurol- 13. Michon B, Moghrabi A, Winikoff R, et al. Complications of
ogy. Neurology. 2011;76:294-300. pheresis in children. Transfusion. 2007;47:1837-1842.