Incidence of adverse events related to intravenous immunoglobulin therapy in children

Abstract

Background

Intravenous immunoglobulin (IVIG) therapy is used in the treatment of pediatric diseases, although data about IVIG‐related adverse events (IVIG‐AEs) are limited. Objectives of this study were to document the incidence of IVIG‐AEs in pediatric hospitalized patients and to identify risk factors for IVIG‐AEs.

Methods

This retrospective cohort study included patients <18 years old who received IVIG therapy while admitted at a Canadian pediatric tertiary care center between 2016 and 2020. Patients and IVIG‐perfusions characteristics were collected, as well as IVIG‐AEs. Bivariate and multivariable logistic regressions were used to explore predictors of IVIG‐AEs.

Results

We included 228 children, totaling 478 IVIG perfusions. Indications included treatment for inflammatory (52.6%), autoimmune disorders (35.5%), and immunoglobulin replacement (11.8%). A total of 213 IVIG‐AEs were reported. Fever (13.6%) and headache (6.7%) were the most frequent IVIG‐AEs. Most IVIG‐AEs were mild (57%) or moderate (31%) in severity, but 12% were severe reactions. The following factors were predictive of IVIG‐AEs in univariate analyses: older age (OR 1.14 [95% CI: 1.07–1.21]), dehydration (OR 2.55 [95% CI: 1.43–4.55]), concurrent allergies (OR 2.87 [95% CI: 1.26–6.56]), first perfusion (OR 1.53 [95% CI: 1.02–2.30]), and higher dosage (OR 2.14 [95% CI: 1.39–3.33]). Concurrent steroids decreased the risk of IVIG‐AEs (OR 0.43 [95% CI: 0.19–0.96]). Older age and higher IVIG dose remained independent predictors of IVIG‐AEs in multivariable analyses.

Conclusions

Mild IVIG‐AEs are frequent in children, and serious reactions may occur. Prospective studies are needed to confirm risk factors for IVIG‐AEs and to evaluate how to best prevent them.

Abbreviations

AE

adverse events

CI

confidence intervals

CMES

Centre Mère‐Enfant Soleil

CTCAE

Common Terminology Criteria for Adverse Events

GEE

generalized estimating equations

IgA

immunoglobulin A

IQR

interquartile range

IV

intravenous

IVIG

intravenous immunoglobulins

IVIG‐AE

IVIG‐related adverse events

OR

odds ratio

TRALI

transfusion‐related acute lung injury

1. INTRODUCTION

Intravenous immunoglobulin (IVIG) is a fractionated blood product of highly purified immunoglobulins G derived from pooled human plasma donors. ¹ IVIG is increasingly used in the pediatric population for both licensed and off‐label indications. ² , ³ The principal uses in children are replacement therapy in immunodeficiencies, as well as an immunomodulatory agent in an increasing number of autoimmune and inflammatory disorders. ⁴ IVIG therapy is considered safe and well tolerated. ⁵ However, adverse events (AEs) still occur with its administration. Studies evaluating IVIG‐related adverse events (IVIG‐AEs) in the pediatric population reported incidence rates ranging from 0.6% to 40%. ⁴ , ⁶ , ⁷ , ⁸ These reactions are typically self‐limited and of mild to moderate severity, ⁴ , ⁸ , ⁹ , ¹⁰ including headache, fever, chills, myalgia, fatigue, nausea, dyspnea, and tachycardia. ⁷ , ⁸ , ¹¹ However, there are more serious and potentially life‐threatening IVIG‐AEs such as acute renal failure, thromboembolic events, aseptic meningitis, hemolysis, anaphylaxis, transfusion‐related acute lung injury (TRALI), and the theoretical potential, albeit low, for viral transmission (e.g., hepatitis B or C virus, human immune deficiency virus). ¹¹ , ¹² Variables potentially affecting the risk of IVIG‐AEs include rate of infusion, initial IVIG perfusion, history of IVIG‐AEs, hydration status, concurrent bacterial infection, specific brand, age, and underlying medical conditions, such as immunoglobulin A (IgA) deficiency, hypertension, autoimmunity, or renal disease. ⁵ , ⁷ , ⁹ , ¹¹ , ¹³ , ¹⁴ Due to lack of data from well‐designed clinical trials surrounding the rate of complications among the pediatric population and the variables potentially affecting the risk and intensity of IVIG‐AEs, ⁷ , ¹³ authors highlight the need for vigilance and expert practice when prescribing IVIG therapy. ² , ⁵ , ⁹ , ¹³ , ¹⁴ , ¹⁵

There is a paucity of robust evidence regarding the incidence and risk factors of IVIG‐AEs in children, despite the constant increase in IVIG use. ¹¹ , ¹⁵ This accelerated expansion appears to be related to the expanding spectrum of “off‐label indications,” ⁵ , ¹⁵ for which existing data are even more limited. ¹¹ , ¹³ In intent to prevent or reduce IVIG‐AEs, understanding the variables potentially affecting the risk of AEs is essential. This study aims to examine the incidence, severity, and risk factors for IVIG‐AEs in pediatric inpatients, based on our practical experience across a wide spectrum of indications.

2. METHODS

2.1. Study design

We conducted a 5‐year observational retrospective cohort study involving all hospitalized children who underwent IVIG therapy at Centre Mère‐Enfant Soleil (CMES), a pediatric tertiary center in Quebec City, Canada. The Institutional Research Ethics Committee approved this study.

2.2. Patients and infusions

Children aged 0–18 years who underwent IVIG for any indication, while being hospitalized in CMES between January 1, 2016, and December 31, 2020, were considered for inclusion. We excluded patients treated exclusively in outpatient settings or in the emergency room, due to limited availability of follow‐up data. Patients who received specific immune globulin, Rho(D) immune globulin, or subcutaneous immune globulin were also excluded.

Patients received IVIG according to the manufacturer's protocol. Typically, infusions commenced at a slow rate (~0.5 mg/kg/min) and were gradually increased to the maximum prescribed rate (around 4.8 mg/kg/min) over an hour. Dosages per treatment course varied based on clinical indication, ranging from 0.5 g/kg for immune replacement to 1–2 g/kg for immunomodulatory therapy. Adjustments to dosages and infusion rates were made as necessary to accommodate individual tolerance. Five commercial IVIG preparations were delivered by our supplier based on their commercial contracts: Gammagard® (Takeda Pharmaceuticals America Inc., 2021), Gamunex® (Grifols Therapeutics Inc., 2019), Hizentra® (CSL Behring LLC Canada, Inc., 2020), Panzyga® (Octapharm Canada, Inc., 2019), and Privigen® (CSL Behring Canada LLC, Inc., 2015). These preparations were supplied by our blood bank based on anticipated product availability. Generally, patients continued to receive the same product throughout their treatment course unless tolerance issues were identified, but patients could be exposed to different products and lots based on IVIG availability.

2.3. Adverse effects (AEs)

We conducted a comprehensive assessment of nonserious and serious AEs ⁷ , ¹⁵ encountered throughout our study period, with each AE's severity graded according to the Common Terminology Criteria for Adverse Events, Version 5.0 (CTCAE v5). ¹⁶ Definitions of IVIG‐AEs and criteria for differentiating the symptoms of underlying conditions from IVIG‐AEs are outlined in Appendix A. Nonserious AEs included fever, anxiety, headache, tachycardia, hypotension, nausea and vomiting, rash, and myalgia. Serious AEs comprised reactions demanding substantial medical attention or bearing potential life‐threatening risks and included anaphylaxis, aseptic meningitis, hemolysis, thromboembolic events, TRALI, and suspected infection transmission. Serious AEs and AEs of grade 3 or higher were considered severe reactions. Additionally, we examined both immediate AEs, defined as reactions occurring during infusion or within 6 h thereafter, and delayed AEs, occurring from 6 h to 10 days after the end of infusion.

2.4. Data collection and statistical analysis

Patients were identified via the Blood Bank database (Traceline software), and their medical health records were then reviewed. We collected information regarding patient demographics, clinical status, and the purpose of IVIG administration. We defined intravenous (IV) hydration as administration of IV fluids during or within 6 h of the IVIG infusion for any medical reason (dehydration, in anticipation of surgery, etc.). Dehydration was defined as documentation in the medical chart of dehydration or significant reduction in oral fluid intake.

Every perfusion administered was reviewed for the emergence of signs and symptoms occurring during and up to 10 days after, as long as the patient remained hospitalized.

Three trained individuals from the research team performed medical chart reviews using a pre‐piloted case report form, using a common extraction manual. Ten percent of charts were reviewed independently by a second research team member; no discrepancies were identified.

We employed bivariate logistic regression to examine the relationships between each characteristic and the incidence of IVIG‐AEs. Multivariate logistic regression was used to explore the predictive roles of potential risk factors and adjust for potential confounding variables, in which IVIG‐AE was the dependent variable. Odds ratio (OR) and 95% confidence intervals (CI) were used to analyze the associations between potential risk factors and occurrence of AEs. Generalized estimating equations (GEE) models were used to handle correlated data from individuals with multiple perfusions. Analyses were performed in SAS software, version 9.4. A, two‐sided p‐value of <.05 was used to establish statistical significance.

3. RESULTS

3.1. Patients and infusion characteristics

Two hundred and ninety‐six hospitalized children underwent IVIG treatment during our study period, and 228 of them (77.1%) met the inclusion criteria. Baseline patient's characteristics are detailed in Table 1. There were 107 females (46.9%) and 121 males (53.1%), with ages ranging from 0.0 to 17.9 years (median 2.2 years). Over half of the patients received IVIG for inflammatory indications (120, 52.6%). Among the underlying diagnoses, Kawasaki disease (58, 25.4%) predominated, followed by neurologic disorders (30, 13.2%), and 103 patients (45.2%) received IVIG for various other diagnoses that could not be classified within our predefined categories. The diagnoses of neurologic and disorders defined as others are described in Appendix B. A subset of 43 patients (18.9%) received three or more IVIG treatments during their hospital stay, but the majority of patients (122, 53.5%) received a single perfusion.

TABLE 1.

Baseline patient's characteristics.

Characteristics	Patients (n=228)
	n	%
Sex (Male)	121	53.1
Age (years), med (range)	2.2 (0.0–17.9)
0–1	84	36.8
1–9	116	50.9
10–17	28	12.3
Allergies	29	12.7
History of IVIG‐AEs	10	4.4
Autoimmunity	3	1.3
Renal disease	8	3.5
Concomitant infection a	91	39.9
Place of hospitalization
Day unit b	2	0.9
Pediatric wards	115	50.4
ICU	48	21.1
NICU	53	23.2
Emergency room b	10	4.4
Indication for receiving IVIG
Autoimmune	81	35.5
Inflammatory	120	52.6
Replacement	27	11.8
Diagnostic
ITP	16	7.0
Neurological disorders	30	13.2
Kawasaki disease	58	25.4
PIMS	5	2.2
Others	103	45.2
Number of perfusions med (range)	1.0 (1.0–18.0)
One	122	53.5
Two	63	27.6
Three to five	31	13.6
Five and more	12	5.3
IV hydration	160	70.2

Abbreviations: ICU, intensive care unit; ITP, immune thrombocytopenic purpura; IV, intravenous; IVIG‐AEs, intravenous immunoglobulin associated adverse events; Med, median; NICU, neonatal intensive care; PIMS, pediatric inflammatory multisystem syndrome.

^a Viral or bacterial infection when receiving their IVIG perfusion.

^b Patients hospitalized before the end of IVIG perfusion.

Throughout the study duration, a total of 478 IVIG perfusions were administered (median: 1.0/patient, interquartile range [IQR]: 1.0–2.0), with 212 (44.4%) representing first‐time exposure. The characteristics of perfusion are listed in Table 2. The median dose administered per treatment course was 1.0 g/kg (interquartile range, IQR: 1.0–2.0). Premedication was employed in 203 perfusions (42.5%), including acetaminophen (132, 27.6%), antihistamines (10, 2.1%), and corticosteroids (81, 16.9%). In certain cases, these medications were administered as part of treatment to the underlying condition, rather than for premedication purposes. IV hydration was provided in 283 perfusions (59.2%). Panzyga® (55.0%) emerged as the predominant IVIG brand, followed by Privigen® (28.0%).

TABLE 2.

Distribution of perfusions and IVIG‐AEs according to potential risk factors.

Characteristics	All perfusions n (%) n = 478	Perfusions with AEs n (%) n = 125	Perfusions without AEs n (%) n = 353	Odds ratio	95% confidence interval	p‐value
Female	247 (51.7)	55 (44.0)	192 (54.4)	Ref	‐	‐
Male	231 (48.3)	70 (56.0)	161 (45.6)	1.44	0.84–2.46	.18
Age (years)						<.0001
0–1	140 (29.3)	17 (13.6)	123 (34.8)	Ref	‐	‐
1–9	285 (59.6)	79 (63.2)	206 (58.4)	3.29	1.61–6.76	.003
10–17	53 (11.1)	29 (23.2)	24 (6.8)	8.26	3.25–21.3	<.0001
Allergies	60 (12.6)	27 (21.6)	33 (9.3)	2.87	1.26–6.56	.01
History of IVIG‐AEs	59 (12.3)	27 (21.6)	32 (9.1)	1.57	0.79–3.13	.20
Comorbidities
IgA deficiency	23 (4.8)	5 (4.0)	18 (5.1)	0.81	0.27–2.46	.71
Hypertension	18 (3.8)	3 (2.4)	15 (4.2)	0.84	0.21–3.44	.81
Autoimmunity	5 (1.0)	2 (1.6)	3 (0.8)	2.70	0.24–30.2	.42
Renal disease	19 (4.0)	6 (4.8)	13 (3.7)	1.53	0.33–7.02	.58
Concomitant infection	177 (37.0)	47 (37.6)	130 (36.8)	1.05	0.63–1.75	.86
Indication for IVIG						.01
Autoimmune	207 (43.3)	46 (36.8)	161 (45.6)	1.17	0.48–2.85	.74
Inflammatory	204 (42.7)	69 (55.2)	135 (38.2)	2.44	1.07–5.57	.03
Replacement	67 (14.0)	10 (8.0)	57 (16.1)	Ref	‐	‐
Concomitant medication
Antibiotics	202 (42.3)	53 (42.4)	149 (42.2)	1.07	0.66–1.74	.78
Antihypertensive	33 (6.9)	4 (3.2)	29 (8.2)	0.58	0.17–2.05	.40
Immunosuppressants	41 (8.6)	11 (8.8)	30 (8.5)	0.98	0.40–2.42	.97
ASA	70 (14.6)	28 (22.4)	42 (11.9)	2.13	1.21–3.73	.008
Medication < 6h a
Acetaminophen	132 (27.6)	54 (43.2)	78 (22.1)	2.12	1.31–3.43	.002
Antihistamine	10 (2.1)	2 (1.6)	8 (2.3)	0.49	0.06–4.28	.52
Corticosteroids	81 (16.9)	12 (9.6)	69 (19.5)	0.43	0.19–0.96	.04
Dehydration	56 (11.7)	27 (21.6)	29 (8.2)	2.55	1.43–4.55	.002
IV hydration	283 (59.2)	81 (64.8)	202 (57.2)	1.10	0.67–1.79	.72
IVIG doses b (g/kg BW) med (IQR)	1.0 (1.0–2.0)	‐	‐	2.14	1.39–3.33	.001
Initial infusion rate c (mg/kg/hour)
Normal	466 (97.5)	121 (96.8)	345 (97.7)	Ref	‐	‐
Slow	11 (2.3)	4 (3.2)	7 (2.0)	1.05	0.27–4.06	.94
Missing	1 (0.2)	‐	‐	‐	‐	‐
First IVIG perfusion d	212 (44.4)	68 (54.4)	144 (40.8)	1.53	1.02–2.30	.04
IVIG brands e						.06
Gammagard®	60 (12.6)	29 (23.2)	31 (8.8)	3.00	1.34–6.73	.008
Gamunex®	12 (2.5)	2 (1.6)	10 (2.8)	0.60	0.12–2.92	.52
Hizentra® f	4 (0.8)	0 (0.0)	4 (1.1)	‐	‐	‐
Panzyga®	263 (55.0)	66 (52.8)	197 (55.8)	1.29	0.66–2.52	.46
Privigen®	134 (28.0)	28 (22.4)	106 (30.0)	Ref	‐	‐
Missing	5 (1.0)	‐	‐	‐	‐	‐

Abbreviations: ASA, acetylsalicylic acid; BW, body weight of patient; IgA, immunoglobulin A; IQR, interquartile range; IV, intravenous; IVIG‐AEs, IVIG‐associated adverse events; Med, median; Ref, reference value.

^a Medication received during or within 6 h prior to IVIG infusion.

^b The value indicate the cumulative dose for one treatment course in one patient.

^c Slow <0.5 mg/kg/min; Normal ≥0.5 mg/kg/min.

^d Patient's initial exposure to IVIG.

^e Takeda Pharmaceuticals America Inc., 2021; Grifols Therapeutics Inc., 2019; CSL Behring LLC Canada, Inc., 2020; Octapharm Canada, Inc., 2019; CSL Behring Canada LLC, Inc., 2015.

^f No adverse reaction was found in this category.

3.2. IVIG‐related adverse events (IVIG‐AEs)

A total of two hundred and thirteen IVIG‐AEs were documented throughout the study duration, occurring in 125 out of 478 perfusions (26.2%, 95% CI 22%–30%), and affecting 89 out of 228 patients (39.0%, 95% CI 33%–46%). Details regarding the frequency and severity of these AEs are shown in Table 3. Immediate IVIG‐AEs (124/213, 58.2%) were more frequent than delayed IVIG‐AEs (69/213, 32.4%). Fever, headache, and tachycardia were the most common AEs encountered, occurring in 13.6% (65/478), 6.7% (32/478), and 6.5% (31/478) of perfusions, respectively.

TABLE 3.

Incidence, grading, and timing of IVIG‐AEs by perfusion.

Adverse events	Incidence a n (% of perfusions b )	Grade a (n)			Timing a (n)
		1–2	3–4	Missing	Immediate	Delayed	Missing
Serious IVIG‐AEs b
Anaphylaxis	3 (0.6)	0	3	0	3	0	0
Aseptic meningitis c	0 (0.0)	0	0	0	0	0	0
Hemolysis	5 (1.0)	4	1	0	0	2	3
Infection	2 (0.4)	2	0	0	0	1	1
Thromboembolic event	4 (0.8)	3	1	0	0	3	1
TRALI	1 (0.2)	0	0	1	0	0	1
Non‐serious IVIG‐AEs
Anxiety	5 (1.0)	4	1	0	4	1	0
Fever	65 (13.6)	56	8	1	41	19	5
Headache	32 (6.7)	28	4	0	17	14	1
Hypotension	20 (4.2)	17	3	0	14	5	1
Myalgia	2 (0.4)	2	0	0	2	0	0
Nausea/Vomiting	28 (5.9)	26	2	0	13	14	1
Rash	15 (3.1)	15	0	0	6	5	4
Tachycardia	31 (6.5)	30	1	0	24	5	2
Total	15/198 d	187	24	2	124	69	20

Abbreviations: IVIG‐AEs, IVIG‐associated adverse events; TRALI, transfusion‐related acute lung injury.

^a On total number of perfusion (n = 478).

^b Percentage after excluding missing values.

^c No patient required a lumbar puncture on account of additional features suggestive of aseptic meningitis.

^d Serious IVIG‐AEs/Non‐serious IVIG‐AEs.

The majority of AEs (121/213, 56.8%) were of mild severity (grade 1) and did not require treatment. However, 11.3% (24/213) were graded ≥ 3, and 7.0% (15/213) were classified as serious AEs (including all grades). Taken together, 16.0% (34/213) were considered as severe reactions (grade ≥ 3 and serious AEs). Approximately 4.6% (22/478) required a reduction in infusion rate, and the same number required a complete discontinuation of the transfusion. Dedicated treatment for IVIG‐AEs was administered in 12.6% (60/478) of perfusions, like acetaminophen (38/478, 7.9%) or antihistamines (13/478, 2.7%).

3.3. Factors associated with the development of IVIG‐AEs

Distribution and bivariate analyses of potential predictive factors and IVIG‐AEs are shown in Table 2. A significantly higher rate of AEs was found in perfusions among patients aged between 1 to 9 years and those aged 10 to 17 years, when compared to patients aged under 1 year. Notably, within the subgroup of perfusions in patients aged over 10 years, 54.7% (29/53) experienced such events. Patients with concomitant documented allergies, whether to antibiotics or other medications, displayed a significantly higher rate of IVIG‐AEs. No significant differences were observed when patients had other comorbidities, such as IgA deficiency, hypertension, autoimmunity, or renal disease. Previous history of IVIG‐AEs did not predict IVIG‐AEs in our cohort.

When exploring factors related to IVIG perfusions, it was found that IVIG infusion intended for inflammatory indications had significantly higher IVIG‐AEs rates compared to those for replacement therapy. Concurrent treatment with acetylsalicylic acid was associated with an increase in IVIG‐AEs, with 40.0% (28/70) of those associated with AEs. Premedication with acetaminophen or antihistamines, whether taken specifically to prevent AEs or for addressing other symptoms (provided they were taken within 6 h prior to IVIG), did not result in a reduced rate of IVIG‐AEs. Corticosteroid usage before treatment was the only medication significantly reducing the AEs rate. When IVIG were administered under conditions of dehydration, a significantly higher occurrence of AEs was observed. IV hydration, either before or during the infusion, did not lead to a reduced rate of IVIG‐AEs. The median dose (g) of IVIG per body weight (kg) for a single treatment course was significantly higher (1.0 g/kg, IQR: 1.0:2.0) for perfusions with AEs, in comparison to perfusions without AEs (1.0 g/kg, IQR: 0.5:1.0). No statistically significant difference was noted when considering a slower (<0.5 mg/kg/min) initial rate of infusion. The first IVIG exposure demonstrated a significantly higher occurrence of AEs compared to subsequent perfusions. Commercial IVIG preparations were not associated with IVIG‐AEs, except for Gammagard® which had significantly more AEs when compared to Privigen®.

Age, IVIG indication, history of IVIG‐AEs, IgA deficiency, concurrent treatment with acetylsalicylic acid, acetaminophen, antihistamines and corticosteroid, dehydration, first IVIG perfusion, and dosage were examined as potential independent risk factors for AEs. The multivariable analysis results are detailed in Table 4. After adjustment for possible confounding factors, AEs displayed significant associations with older age, compared to infants <1 year old (1–9 years: p = .03) (10–17 years: p = .0001) and increased IVIG doses (p = .03).

TABLE 4.

Multivariable regression analysis: Predictors of IVIG‐related adverse events.

Characteristics	aOR	a95% CI	p value
Age (years)
0–1	Ref	‐	‐
1–9	2.95	1.33–6.58	.03
10–17	7.76	2.76–21.8	<.0001
History of IVIG‐AEs	1.64	0.66–4.07	.29
Comorbidities
IgA deficiency	0.95	0.26–3.53	.94
Indication for IVIG
Autoimmune	1.21	0.43–3.35	.72
Inflammatory	1.44	0.48–4.29	.52
Replacement	Ref	‐	‐
Concomitant medication
ASA	0.69	0.31–1.54	.37
Medication < 6h a
Acetaminophen	1.21	0.65–2.27	.54
Antihistamine	0.35	0.04–3.34	.36
Corticosteroid	0.45	0.19–1.08	.08
Dehydration	1.40	0.73–2.70	.31
IVIG dose (g/kg BW)	1.91	1.05–3.44	.03
Initial infusion rate b (mg/kg/hour)
Normal	Ref	‐	‐
Slow	1.51	0.90–2.54	.12

Abbreviations: a95% CI, adjusted 95% confidence interval; aOR, adjusted odds ratio; ASA, acetylsalicylic acid; BW, body weight of patient; IgA, immunoglobulin A; IV, intravenous; IVIG‐AEs, IVIG‐associated adverse events; Ref, reference value.

^a Medication received during or within 6 h prior to IVIG infusion.

^b Slow <0.5 mg/kg/min; Normal ≥0.5 mg/kg/min.

4. DISCUSSION

This retrospective study is, to our knowledge, the largest one to date evaluating IVIG‐AEs in children across various indications for IVIG use and classifying these events using a standardized classification system (CTCAE). We observed two hundred and thirteen IVIG‐AEs in 228 patients who received 478 perfusions, for an incidence rate of 39.0% of patients and 26.2% of perfusions sustaining at least one IVIG‐AE. Fever was the most frequent IVIG‐AE (13.6%), followed by headache (6.7%) and tachycardia (6.5%). While most were mild and self‐limited, 16.0% were severe reactions. Older age and higher IVIG doses were independent predictors of IVIG‐AEs.

The incidence of IVIG‐AEs and their underlying mechanisms are not well understood. Some studies suggest that IVIG‐AEs may be less common in children. ⁴ , ⁶ , ¹³ For instance, Kato et al. ⁶ reported rates of 6.9% in adults and 1.5% in children. However, these studies may have underestimated IVIG‐AEs in children, due to their limited ability to express symptoms. In contrast, a prospective study by Esmaeilzadeh et al. ¹⁷ found similar rates of IVIG‐AEs in both pediatric (41.8%) and adult patients (48.2%) with inborn errors of immunity. Other studies on pediatric populations also show high rates of IVIG‐AEs. Kubota et al. ⁷ documented a 37.5% incidence in 104 Japanese children receiving IVIG for neurological disorders, while Singh‐Grewal et al. ⁴ reported AEs rate of 44.8% in their prospective study of 58 children receiving IVIG for immunodeficiency and immunomodulation, with 23.5% of the 345 perfusions affected. A study by Nosadini et al. ¹⁸ retrospectively studied 196 children receiving IVIG for neurological indications and found a slightly lower incidence of 25.5% in patients. In our present study, which included a larger and more diverse pediatric population (228 patients and 478 IVIG perfusions, for any indications), we observed a high incidence of IVIG‐AEs, affecting 39% of patients and 26.2% of perfusions. These findings align with recent literature, underscoring the need for careful considerations of the substantial risk of AEs when prescribing IVIG to children.

Of note, no episode consistent with aseptic meningitis were observed in our study. Given the retrospective nature of our data, we are tributary to the clinical gestalt and investigations performed by treating physicians. Although it appears to be a rare side effect, ¹⁹ with an estimated prevalence of ~1 case per 200 transfusions in the general population, ²⁰ aseptic meningitis has important clinical consequences and should be considered if a patient presents with a suggestive clinical presentation (fever, nausea, vomiting, headaches, neck stiffness, or seizures).

Our study found that 58% of IVIG‐AEs were immediate reactions, while 32% were delayed reactions (10% undetermined). This contradicts some other pediatric studies, where delayed reactions were more common. For instance, Singh‐Grewal et al. ⁴ observed that delayed reactions IVIG‐AEs occurred more frequently than immediate reactions (41.4% vs. 10.3% of children). However, it is important to note that their definition of delayed reactions included any events occurring after the infusion, which differs from our study's definition (6 h to 10 days after perfusion). Similarly, Kubota et al. ⁷ also used the end of perfusion as the cutoff for immediate versus delayed IVIG‐AEs, and they reported 79.5% of delayed reactions. Sitehm ²¹ defined immediate as occurring within 6 h of an infusion and delayed as occurring 6 h to 1 week after. According to this definition, a ratio of 60% immediate and 40% delayed AEs was suggested, which is similar to our results. In their retrospective study of 77 patients, Palabrica et al. ²² also documented a majority of immediate IVIG‐AEs, with 79% occurring within 0–6 h after the start of perfusion. In conclusion, the variation in the definition of delayed and immediate AEs across studies complicates interpretation. However, when compared with studies using similar definitions, our results are consistent and suggest that most IVIG‐AEs occur within 6 h of the perfusion.

This study is the first, to our knowledge, to identify inflammatory indications as a predisposing factor for IVIG‐AEs in children. Pediatric studies often focus on specific diagnostics, making it challenging to compare different indications and their association with IVIG‐AEs. Even in adult populations, few studies have explored inflammatory conditions as a risk factor for IVIG‐AEs. Daw et al. ²³ studied 16 cases of significant hemolysis reactions following IVIG therapy in adults and identified underlying inflammatory disease and high cumulative IVIG doses as contributing factors. In contrast, Ramirez et al. ²⁴ observed a higher risk of thromboembolic AEs following IVIG in individuals with autoimmune conditions and secondary immunodeficiency in their cohort of 303 patients, which included 15.2% of children. In their retrospective study of 748 patients (36% under 15 years old), Kato et al. ⁶ similarly found that IVIG‐AEs were more common in patients with autoimmune diseases, as well as those receiving high IVIG doses. Since immediate IVIG‐AEs are thought to be inflammatory responses triggered by the IVIG preparation, ⁵ it is likely that an inflammatory state could predispose patients to this type of reaction. Inflammatory conditions also typically require higher dosages than replacement therapy, ¹¹ which is itself another known risk factor for IVIG‐AEs, as previously discussed, and may confound this association. This study expands our understanding of IVIG‐AEs and highlights the need for heightened vigilance when prescribing IVIG for inflammatory conditions.

In our study, previous history of IVIG‐AEs did not predict IVIG‐AEs. However, although side effects appear to occur more frequently in patients naive to immunoglobulins, a retrospective study on 145 children with primary immunodeficiency receiving IV immunoglobulins showed that a previous history of adverse reactions to IVIG infusion increases the risk of adverse reactions in future transfusions and premedication may prevent these reactions. ²⁵ As we found in our cohort a higher occurrence of IVIG‐AEs in first perfusions, compared to subsequent perfusions, it is possible that medical interventions in subsequent perfusions (e.g., reduced rate of perfusion, premedication, and change in IVIG product) prevented further IVIG‐AEs in subsequent perfusions. Our retrospective study design and sample size did not allow to explore the effect of these interventions further.

Different premedication regimens are employed in practice, despite limited robust evidence of their effectiveness in reducing IVIG‐AEs. ²⁶ In their retrospective study involving 420 adults, Souayah et al. ²⁷ reported a significantly lower incidence of IVIG‐AEs in patients with neuroimmunological disorders of patients who received premedication compared to those without premedication (18.2 vs. 29.3%). The most common regimen was acetaminophen with diphenhydramine and dexamethasone, either alone or in combination with acetaminophen or diphenhydramine. However, Liu et al. ²⁸ observed that premedication did not reduce IVIG‐AEs in their retrospective study involving 66 children with Kawasaki disease. Despite 64 of them received premedication, the incidence rate of IVIG‐AEs remained at 25.8%, which was similar to their historical baseline incidence of 20%. Premedication primarily consisted of diphenhydramine or acetaminophen (79% of perfusions), while methylprednisolone was used in only 3 perfusions. In our study, corticosteroids was the only medication that exhibited a significant correlation with a reduced incidence of AEs, indicating its potential as protecting against IVIG‐AEs.

Further prospective study is needed to determine whether corticosteroids might be effective and safe to prevent IVIG‐AEs, especially for pediatric patients who are at high risk of IVIG‐AEs.

4.1. Limitations

This study's primary limitation is its retrospective nature, which could have led to an underestimation of IVIG‐AEs, as some reactions may not have been documented during the hospital stay, and others may have occurred after discharge. The design of our study likely contributed to the higher prevalence of immediate IVIG‐AEs, which are more readily documented in hospital records. Nevertheless, significant reactions occurring during hospitalization were likely reported, implying that the uncollected IVIG‐AEs were likely mild and of limited clinical significance. Second, this study exclusively involved hospitalized patients, which could limit the generalizability of our findings to patients who receive IVIG in different settings, such as daycare units or emergency rooms. Third, we did not collect data on IVIG‐AEs occurring beyond 10 days after the completion of perfusions, even though literature describes more delayed IVIG‐AEs. ¹² IVIGs are recognized for their potential to interfere with vaccination and present a theoretical risk for blood‐borne infections. ²¹ While gathering data on these complications would have been challenging, it is important to note that they are exceptionally rare, accounting for less than 1% of AEs. ²¹

Finally, some adverse events such as anxiety, headache, and nausea were not assessable in certain patients, notably for patients <1 year. The stratification by age group (<1 year, 1–9 years old, and 10–17 years old) was intended to classify patients according to their characteristics (e.g., weight) and to facilitate comparison with equivalent literature. ²² , ²⁹ , ³⁰

5. CONCLUSION

IVIG‐AEs are common in children and can be severe reactions. Proper hydration, reduced IVIG doses, and pretreatment with corticosteroids could potentially minimize IVIG‐AEs, especially in high‐risk children. Prospective studies are required to further assess the effectiveness of these potential protective measures. Recognizing the prevalence of IVIG‐AEs and their potential contributing factors is essential for enhancing the care of children receiving IVIG.

AUTHOR CONTRIBUTIONS

Drs Jade Côté and Mathilde Chaloult‐Lavoie conceptualized and designed the study, designed the data collection instrument, coordinated data collection, collected data, drafted the initial manuscript, and critically reviewed and revised the manuscript. Dr. Élisabeth Poulin collected data, participated in interpretation of data, and revised the manuscript. Dr. Laurence Amélie Hayes participated in the development of the data collection instrument, collected data, and revised the manuscript. Mahukpe Narcisse Ulrich Singbo carried out the initial analyses, participated in interpretation of data, and critically reviewed and revised the manuscript. Dr. Pierre Ouellet critically reviewed and revised the research protocol and the manuscript. Dre Marie‐Claude Pelland‐Marcotte supervised every step of the study and critically reviewed and revised the manuscript. All authors approved the final manuscript as submitted and agree to be accountable for all aspects of the work.

FUNDING INFORMATION

No funding was secured for this study. Dr Pelland‐Marcotte is supported by Fonds de Recherche en Santé du Québec Junior 1 Scholars Award.

CONFLICT OF INTEREST STATEMENT

The authors have no conflicts of interest relevant to this article to disclose.

APPENDIX A. Definition of selected IVIG‐AEs

A.1.

Adverse event	Definition	IVIG‐AEs
Fever	Rectal temperature ≥ 38°C or oral temperature ≥ 37.8°C at least once	Increase of temperature after the start of the transfusion or recurrence of fever during/after the transfusion
Tachycardia	Neonate (up to 96 h of life): >205 beats/min Infant (4 days to <12 months): >180 beats/min Toddler (1–2 years): >140 beats/min Preschooler (3–5 years): >120 beats/min School‐aged child (6–11 years): >118 beats/min Adolescent (≥12 years): >100 beats/min	Abnormal rapid heart rate for age, occurring during up to 1 h after the end of the transfusion
Hypotension	Neonate (up to 96 h): MAP <60 mm Hg Infant (1–12 months): MAP <60 mm Hg Children (≤10 years): MAP <70 mm Hg + (age in years × 2) Children over 10 years of age: MAP <90 mm Hg	Low blood pressure for age, occurring during or up to 1 h after the end of transfusion
Hemolysis	Elevated reticulocyte count Elevated serum lactate dehydrogenase Elevated serum unconjugated bilirubin Low serum haptoglobin Hemoglobinuria, and/or Significant spherocytosis (2+ or more)	Hemolytic episode occurring within 10 days of IVIG administration: decrease in Hb of ≥10 g/L and a positive direct antiglobulin test (DAT) plus ≥2 of the symptoms
Anaphylaxis	Hypersensitivity response characterized by multisystem symptoms: Cutaneous: urticarial rash Angio‐edema Respiratory: dyspnea, cyanosis, hypoxemia Gastro‐intestinal: nausea, vomiting, diarrhea Cardiovascular: dizziness, hypotension, loss of consciousness	Hypersensitivity immune response, presenting with symptoms in at least two systems during the transfusion
Transfusion‐related acute lung injury (TRALI)	Hypoxemia: SpO2 < 90% on room air Clear evidence of bilateral pulmonary edema on imaging No evidence of left atrial hypertension No temporal relationship to an alternate risk factor for ARDS	New acute lung injury that occurs within 6 h of transfusion with all symptoms
Thromboembolic events	A disorder characterized by the occlusion of a vessel by a thrombus, proven radiologically	Occurring up to 10 days after the end of transfusion
Infection	Any infection for which an antibiotic treatment is needed or listed as a complication in the health record summary	Occurring up to 10 days after the end of transfusion
Headache	Clinically significant if treatment or modification of the perfusion needed	Occurring during or up to 6 h after the end of transfusion
Myalgia
Anxiety
Nausea and vomiting
Rash

Abbreviation: MAP, mean arterial pressure.

APPENDIX B. Additional indications for IVIG – Neurologic and other diseases

B.1.

Category of disease	Diagnostic	Number of cases
Neurologic	Acute disseminated encephalomyelitis	2
	Acute encephalopathy	1
	Amyotrophic lateral sclerosis	1
	Autoimmune polyneuropathy	1
	Brainstem damage	1
	Chronic myelitis	1
	Demyelinating neuropathy	1
	Encephalitis	5
	Epilepsy	7
	Guillain‐Barré syndrome	6
	Opsoclonus‐myoclonus‐ataxia syndrome	1
	Progressive dystonia	1
	Refractory infantile spasms	1
	Tuberculous meningitis‐rhombencephalitis	1
Others	ABO incompatibility	16
	Allo‐immunization	9
	Autoimmune encephalopathy	1
	Auto‐immune anemia	2
	Bacteremia	2
	Botulism	1
	Candidemia	1
	Cellulitis	5
	Coagulopathy	1
	Epidermal necrolysis	1
	Fasciitis	5
	Hemolytic uremic syndrome	1
	Hypogammaglobulinemia	1
	IgG deficiency	2
	Immune dysfunction	1
	Macrophage activation syndrome	1
	Acute myeloid leukemia	2
	Myocarditis	10
	Necrotizing fasciitis	3
	Persistent hyperthermia	1
	Pulmonary graft‐versus‐host‐disease	1
	Rh incompatibility with allo‐immune anemia	2
	Septic shock	14
	Severe infection	6
	Toxic shock	5
	Thrombocytopenia	9

Côté J, Chaloult‐Lavoie M, Poulin É, Hayes LA, Singbo MNU, Ouellet P, et al. Incidence of adverse events related to intravenous immunoglobulin therapy in children. Transfusion. 2025;65(1):88–99. 10.1111/trf.18083