Clindamycin and intravenous immunoglobulins in pediatric invasive group A streptococcal disease: what is the evidence?

TL;DR

This review examines the limited evidence for using clindamycin and IVIG in treating severe pediatric group A streptococcal disease, highlighting the need for more pediatric-specific research.

Contribution

The paper emphasizes the lack of strong pediatric-specific evidence for clindamycin and IVIG in iGAS and calls for larger studies.

Findings

Evidence for clindamycin and IVIG in pediatric iGAS is limited and mostly extrapolated from adult studies.

Current treatment decisions rely on expert consensus due to insufficient data.

Ongoing studies like PEGASUS aim to provide clearer guidance on these therapies.

Abstract

Invasive group A streptococcal disease (iGAS) can present with severe manifestations, including sepsis, necrotizing fasciitis, and streptococcal toxic shock syndrome. Although β-lactam antibiotics are considered standard therapy, the role of adjunctive treatments, particularly clindamycin and intravenous immunoglobulin (IVIG), in children with these severe manifestations remains debated. Clindamycin is effective irrespective of pathogen load and is suggested to inhibit toxin production, while IVIG is hypothesized to modulate the immune response through multiple mechanisms, including neutralization of superantigens. The rarity of iGAS in children and the relatively low pediatric mortality rate (2–10%) complicate gathering evidence, with most data derived from small, low-quality studies. Evidence is often extrapolated from adult studies, despite substantial differences in mortality rates (30–80%). In this narrative review, we discuss the available evidence on the use of adjunctive therapies in severe pediatric iGAS and its possibility in guiding clinical management. Conclusion: Current evidence does not allow definitive conclusions regarding the benefit of clindamycin or IVIG in severe pediatric iGAS. Treatment decisions therefore largely rely on expert consensus and individualized clinical judgment. Ongoing multicenter studies, such as PEGASUS, aim to clarify the role of these adjunctive therapies and guide evidence-based practice. 
What is Known:• β-Lactams are first-line treatment for pediatric invasive group A streptococcal disease (iGAS).• Clindamycin and IVIG have been suggested as adjuncts in severe iGAS, but recommendations differ, and many countries have no formal guidance.What is New:• This review shows that evidence supporting clindamycin and IVIG in children with iGAS remains limited and is largely extrapolated from adult studies. It highlights variability in clinical practice, safety and cost concerns, and the need for larger-scale pediatric studies.

What is Known:

• β-Lactams are first-line treatment for pediatric invasive group A streptococcal disease (iGAS).

• Clindamycin and IVIG have been suggested as adjuncts in severe iGAS, but recommendations differ, and many countries have no formal guidance.

What is New:

• This review shows that evidence supporting clindamycin and IVIG in children with iGAS remains limited and is largely extrapolated from adult studies. It highlights variability in clinical practice, safety and cost concerns, and the need for larger-scale pediatric studies.

Full text

Introduction

Group A Streptococcus (GAS) is a Gram-positive, β-hemolytic, facultative anaerobic coccus that spreads via droplets, (in)direct contact, and possibly airborne transmission. About 10% of school-aged children are asymptomatic pharyngeal GAS carriers, compared to 1–5% of adults.

GAS causes a broad spectrum of infections, most commonly non-invasive diseases like pharyngitis and impetigo. Sometimes, GAS causes invasive infections (iGAS), defined as the presence of bacteria in a normally sterile site (e.g., pneumonia, osteomyelitis, bacteremia, and meningitis) or with severe manifestations such as sepsis, necrotizing fasciitis (NF), or streptococcal toxic shock syndrome (STSS). STSS, mediated by superantigens that trigger a massive T-cell response, is defined as iGAS disease with hypotension and organ failure. Why some children develop iGAS disease while others only experience non-invasive disease remains poorly understood. It is likely an interplay involving host factors (e.g., genetic susceptibility and absence of prior immunity), pathogen characteristics (e.g., emm type and other virulence factors), inoculum size, and viral co-infections with varicella-zoster or influenza.

Recently, there has been renewed interest in identifying the optimal treatment strategy for iGAS disease in children. This is due to the post-COVID-19 pandemic surge in iGAS, with a disproportionate burden in children [1, 2]. While β-lactams are the standard treatment, the role of clindamycin and intravenous immunoglobulins (IVIG) as adjunctive treatments remains unclear, even though severe iGAS cases are associated with substantial morbidity and mortality [3, 4]. Some guidelines suggest adjunctive therapy for all children with iGAS disease, whereas others limit it to cases presenting with NF or shock [3]. For many countries, no formal recommendations exist [3]. Here, we provide a narrative review of the evidence regarding the use of clindamycin and IVIG in children with iGAS disease.

Methods

A structured literature search on PubMed was conducted in October 2025 (Box 1). Additional studies were identified through snowballing from reference lists. We reviewed studies reporting outcomes of clindamycin and/or IVIG in children or adults with iGAS. Box 1. Search strategy("Streptococcus pyogenes"[Mesh] OR streptococcus pyogenes[tiab] OR "group A streptococ*"[tiab] OR streptococcal[tiab] OR ("Streptococcal Infections"[Mesh] AND (group A[tiab] OR pyogenes[tiab]))) AND (invasive[tiab] OR bacteremi*[tiab] OR sepsis[tiab] OR "toxic shock"[tiab] OR “STSS”[tiab] OR "necrotizing fasciitis"[tiab] OR “necrotising soft tissue infection”[tiab]) AND ("Clindamycin"[Mesh] OR clindamycin[tiab] OR "Immunoglobulins, Intravenous"[Mesh] OR "intravenous immunoglobulin*"[tiab] OR IVIG[tiab] OR “immunoglobulin G”[tiab])

Results

An overview of the characteristics and results of the included studies can be found in Table 1. Table 1. Overview of the characteristics and results of included studiesAuthor; countryStudy typePopulationSample sizeYear of data collectionYear of publicationComparisonResultsLimitationClindamycinMulla [7]; USARetrospective population-based studyChildren and adults with iGASMedian age 52 years (range: 0–103)Total: 2571996–20002003β-Lactam plus clindamycinvsonly β-lactamMortality: overall 18%, in group < 55 years 12.4%Reduced mortality with clindamycin in patients with NF (OR 0.11, 95% CI 0.01–0,89), but not iGAS without NF (OR 1.01, 95% CI 0.31–3.33)Majority of population likely adults (high median age)Carapetis [8]; AustraliaNationwide prospective surveillance studyChildren and adults with iGASMedian age clindamycin group 56.2 years (range: 3.8–88.1) and no-clindamycin group 70.4 years (range: 2.6–96.2)Total: 333N = 84 with severe iGAS (i.e., STSS, NF, septic shock, or cellulitis with hypotension)2002–20042014β-Lactam plus clindamycinvsonly β-lactamClindamycin in 53 (63%) casesClindamycin-treated patients had more severe disease (i.e., ICU admission, presence of STSS and length of hospital stay) compared to clindamycin-untreated patients, but lower mortality (15% vs 39%) (P = 0.014); OR 0.28 (95% CI 0.10–0.80))Adjusted point estimate of the OR for mortality (adjusted for age and presence of STSS): lower in clindamycin-treated patients (aOR 0.31, 95% CI 0.09–1.12)UnderpoweredNo statistical significanceMajority of population likely adults (high median age)Babiker et al. [9]; USARetrospective multicenter cohort studyAdults with iGAS (and invasive non-group A/B β-haemolytic streptococcal infections)Total: 1956N = 1079 with iGAS2000–20152021β-Lactam plus clindamycinvsonly β-lactamClindamycin in 343/1079 (31.7%) patientsIn-hospital mortality, crude: 6.5% vs 11.0% (P = 0.04)In-hospital mortality, adjusted: aOR 0.44 (95% CI 0.23–0.81)In-hospital mortality, subgroup without vasopressor-dependent shock or necrotizing fasciitis: 2.6% vs 6.1% (P = 0.04; aOR 0.40 (95% CI 0.15–0.91))Adults onlyLinner et al. [10]; SwedenNationwide surveillance studyAdults with STSS672002–20042014β-Lactam plus clindamycinvsonly β-lactamClindamycin in 52/67 (78%) of patientsSurvival: OR 8.6 (95% CI 1.8–40.4, P = 0.007) with clindamycinAdults onlyZimbelman [11]; USARetrospective studyChildren with iGASTotal: 56N = 19 with “deep infection” (= necrotizing fasciitis, bacteraemia, arthritis, osteomyelitis)N = 37 with “superificial disease”1983–19971999β-Lactam plus clindamycinvsonly β-lactamFavorable outcome in83% vs. 14% (P = 0.006) with “deep infection” if clindamycin-treatedSmall sample sizeBetter outcome towards end of data collection period (improved medical care in general?)Inclusion of NF, where clindamycin seems to have a good effectIVIGSenda et al. [18]; JapanNationwide observational studyAdults with iGASMedian age 65 years4812018–20212023β-Lactam plus clindamycin plus IVIGvsβ-lactam plus clindamycinOverall mortality: 31.0%In-hospital mortality: no difference (aOR 0.99, 95% CI 0.93–1.04, P = 0.92)Similar results after propensity score matching (OR: 1.00, 95% CI 0.62–1.61, P > 0.99)Adults onlyLinnér et al. [10]; SwedenNationwide surveillance studyAdults with STSSMedian age IVIG group 60 years (range: 31–87) and no-IVIG group 65 years (range: 32–92)672002–20042014β-Lactam* (+/− clindamycin) plus IVIGvsβ-lactam* (+/− clindamycin)one patient received no β-lactam (group unspecified)*IVIG in 23/67 (34%) of patients. 21/23 (91%) received clindamycin28-day survival: 87% vs 50% (P < 0.01; OR 6.7 (95% CI 1.7–25.7))Factors influencing survival in STSS: Simplified Acute Physiology Score II (OR 1.1; P = 0.007), clindamycin (OR 8.6; P = 0.007), and IVIG (OR 5.6; P = 0.030)Adults onlyMadsen [20]; Scandinavia (1 hospital in Denmark, 3 hospitals in Sweden, 1 hospital in Norway)RCTAdult patients with necrotising soft tissue infection at, or planned to be admitted, to ICU1002014–20162017Standard therapy plus IVIGvsstandard therapy plus placeboStandard therapy* = meropenem, clindamycin and ciprofloxacinMortality, day 28: 12% in IVIG compared to 12% in placebo (6/50 each; RR 1.0, 95% CI 0.35–2.89; P > 0.99)Physical component summary score of Short Form Health Survey (SF-36) at 6 months:36 (IQR 0–43) in IVIG vs 31 (IQR 0–47) in placebo; mean adjusted difference 1 (95% CI − 7 to 10), P = 0.81Pathogen not specifiedDarenberg et al. [21]; EuropePlacebo-controlled RCTAdults with STSS211999–20012003β-Lactam plus clindamycin plus IVIGvsβ-lactam plus clindamycin plus placeboMortality at 28 days: 10% (1/10) vs 36% (4/11)Decreased SOFA score at days 2 (P = 0.02) and 3 (P = 0.04)Increased plasma neutralizing activity against superantigens (P = 0.03)Adults onlyUnderpoweredCarapetis [8]; AustraliaNationwide prospective surveillance studyChildren and adults with iGASMedian age not mentionedTotal: 333N = 84 with severe iGAS (i.e., STSS, NF, septic shock, or cellulitis with hypotension)2002–20042014β-Lactam plus clindamycin plus IVIGvsβ-lactam plus clindamycinClindamycin in 53/84 (63%) cases**IVIG in 14/84 (17%) cases, all of whom also received clindamycin. In the IVIG-group, all had STSS and/or NFMortality: 7% vs 18% (P = 0.24)Adjusted point estimate of the OR for mortality (adjusted for age and presence of STSS): aOR 0.12 (95% CI 0.01–1.29) vs aOR 0.39 (95% CI 0.10–1.46)UnderpoweredNo statistical significanceMajority of population likely adults (high median age)Parks et al. [22]Systematic review and meta-analysisAdults and children with STSS160Includes [5, 7, 11, 13] and Kaul et al. (1999)1980–20172018IVIG vs no-IVIG5 studies individually: administration of IVIG in the clindamycin-treated subgroup associated with lower mortality rates, yet no statistical significancePooled analysis: IVIG associated with reduction in mortality rate (from 33.7% to 15.7%; RR 0.46, 95% CI 0.26–.83; P = 0.01), consistent across nonrandomized studies (RR 0.47; 95% CI 0.25–0.86) and effect size estimate of the RCT (RR 0.42; 95% CI 0.05–3.28)Small sample sizeEffect size estimate has large CIBaseline characteristics probably differingMajority of population adultsAdalat [23]; UKSurvey-based multicenter retrospective studyChildren with streptococcal and staphylococcal TSSTotal: 49n = 29 with streptococcal TSSN = 20 with staphylococcal TSS20142014N.AMortality: 8/29 (27.6%)Clindamycin in 3/8 (38%) of patients who died and in 30/41 (73%) of survivors**IVIG in 0/8 (0%) of patients who died, and in 10/41 (24%) of survivorsUnderpowered study with methodological limitations, limiting reliability and generalizabilityShah [16]; USAMulticenter retrospective cohort studyChildren with STSS1922003–20072009Penicillin or vancomycin (+/− clindamycin) plus IVIGvspenicillin or vancomycin (+/− clindamycin)*Clindamycin in 182/192 (94.8%)**IVIG in 84/192 (44%). Use was stable over time, varied by hospital (29–60%), and given to 30% of non-ICU patients)*Mortality, overall: 6% (IVIG) vs 2.8% (no-IVIG) (P = 0.300). Propensity-matched mortality: 4.5% in both groups (P = 1.000)Length of stay: no significant differenceHospital costs:increased in IVIG group (median difference $6139)Identification of STSS cases relied on discharge ICD codes and billing for IV penicillin, making it likely that some patients without STSS were included

Clindamycin

In pediatric practice, clindamycin is used to treat anaerobic, streptococcal, and staphylococcal infections, including skin, soft tissue, bone, and joint infections. It is a lincosamide antibiotic that inhibits protein synthesis by binding to the 50S subunit of the bacterial ribosome, thereby preventing peptide chain elongation. This not only halts bacterial growth but also inhibits production of virulence factors and exotoxins, including superantigens. While β-lactams rely on active cell wall synthesis, which could be reduced in high-inoculum infections due to altered bacterial physiology and slowed growth—a phenomenon historically associated with the *Eagle effect—*clindamycin remains effective regardless of bacterial growth phase [5]. This makes it particularly valuable in infections with high bacterial load.

Clindamycin in mouse models, adults, and mixed-age populations

For NF, clindamycin significantly reduces skin lesions in mice [6]. This benefit likely transfers to humans, where it possibly also decreases mortality. Mulla et al. performed a retrospective analysis of 228 patients, mostly adults, with iGAS in the USA [7]. Of the 45 patients with NF, treatment with clindamycin was associated with significantly lower mortality (OR, 0.11). No benefit was observed in patients with iGAS without NF. In 2014, Carapetis et al. performed a population-based, prospective, active surveillance study of iGAS in Australia [8]. Data were collected between 2002 and 2004 and included 84 cases with severe iGAS disease. Clindamycin-treated patients had more severe disease but a lower mortality rate (15% vs 39%), though not significantly. It is likely that the number of children was low, as the median ages of the two groups were 56 and 70 years. The retrospective multicenter cohort study by Babiker et al., including 1000 adults with iGAS between 2000 and 2015, showed that clindamycin use was associated with significantly lower in-hospital mortality, even in patients without shock or NF (6.5% versus 11%) [9]. For STSS, one of the main publications is the prospective surveillance study by Linnér et al., including 67 adults. They noted a significant benefit on survival with the addition of clindamycin to penicillin (aOR, 8.6) [10].

Clindamycin in children

In 1999, Zimbelman et al. showed a favorable outcome in 83% of clindamycin-treated children with iGAS, compared to 14% if treatment consisted of β-lactams alone [11]. This was a small, retrospective study, with only 19 children with “deep infection,” including NF. Moreover, children admitted towards the end of the data collection (1983–1997) generally had a better outcome, likely due to improved medical care. The true effect of clindamycin is therefore difficult to isolate.

In 2021, Laho et al. published a review of the benefit of clindamycin in pediatric iGAS, based on the studies by Zimbelman and Carapetis [3]. They concluded that “despite the lack of scientific evidence, the drug is safe and inexpensive” and proposed its use to treat “all serious iGAS infections which need to be treated in a hospital.” This statement on safety contrasts with publications regarding an increase in Clostridioides difficile–associated diarrhea, with clindamycin being a high-risk drug. Furthermore, the increasing antibiotic resistance among GAS isolates, mostly due to erm genes, is concerning. Notably, this resistance is frequently cross-induced by macrolide exposure, as macrolides share an overlapping ribosomal target. Consequently, macrolide-resistant isolates may also exhibit clindamycin resistance, underscoring the importance of antimicrobial stewardship when prescribing these agents in children [12]. Importantly, clindamycin has been shown to decrease toxin production as well as skin lesion size in mice with NF, irrespective of the isolate susceptibility [6].

Other protein inhibitors

An ongoing discussion concerns the use of linezolid, an oxazolidinone that also inhibits protein synthesis via the 50S ribosome subunit, as an alternative to clindamycin. Resistance towards linezolid is lower for GAS, it crosses the blood–brain barrier, short-term use is generally safe in children, and it appears to be as effective as clindamycin in retrospective studies in adults [13–15].

In summary, the clinical effect of clindamycin in severe iGAS has been demonstrated in studies involving adults and mixed-age populations with a median age above 50 years. The only study focusing exclusively on children is small and from 1999.

IVIG

In pediatric practice, IVIG is used to treat immunoglobulin deficiencies, but also to modulate the inflammatory process, for example in Kawasaki disease. IVIG acts by (a) enhancing bacterial clearance through opsonization and phagocytosis; (b) neutralization of exotoxins, including superantigens; (c) inhibition of T-cell proliferation and cytokine production; and (d) exertion of a broader immunomodulatory effect [10, 16, 17]. For iGAS, IVIG could help mitigate the massive “cytokine storm” that occurs when GAS superantigens are at play. When IVIG is administered, it is usually in addition to clindamycin, and limited data exists on the addition of IVIG without clindamycin.

IVIG in mouse models, adults, and mixed-age populations

Few studies have evaluated IVIG in patients with iGAS without STSS, and those that have been conducted did not show reduced in-hospital mortality [18]. Studies on IVIG in NF have concluded that penetration into necrotic tissue might be limited, and mouse models failed to demonstrate an effect on the course of the disease [19]. Linnér et al. observed a positive, albeit non-statistically significant, benefit of IVIG on survival in adult patients with NF [10]. These patients also had STSS, which makes interpretation difficult. A randomized controlled trial (RCT) published in 2018 on 100 adults with NF showed no difference in 30-day mortality nor serious adverse events between IVIG and placebo groups [20]. However, the causative pathogens were not specified. Overall, evidence supporting IVIG in iGAS other than STSS is not convincing. The important question that remains is whether to use IVIG in STSS.

Darenberg et al. designed a European multicenter, double-blind RCT to provide a definite answer on the efficacy and safety of IVIG in STSS [21]. Patients received either IVIG or placebo for 3 days, in addition to clindamycin and benzylpenicillin. Of 21 adults enrolled, 10 received IVIG, before early termination due to low enrollment. Patients who received IVIG demonstrated decreased sepsis-related organ failure assessment (SOFA) score at days 2 and 3 and increased plasma neutralizing activity against superantigens. The 28-day mortality rate was lower in the IVIG group (1/10) compared with placebo (4/11), yet not statistically significant, probably due to the small sample size. The question was therefore left unresolved. Some co-authors subsequently performed a prospective analysis in 2014 of 67 adults with STSS in Sweden [10]. This study showed significantly improved 28-day survival in the IVIG group (87% vs 50%).

Also in 2014, Carapetis et al. evaluated the effectiveness of IVIG in 84 cases of severe iGAS in Australia [8]. It is unclear how many of these were children. IVIG was used in addition to clindamycin and a β-lactam in 14 patients. Of these, 13 had STSS. This study suggested a lower, yet not statistically significant mortality rate among patients receiving IVIG in addition to a β-lactam.

IVIG in children

A 2018 systematic review and meta-analysis including five prospective randomized and non-randomized studies support the previously described findings with a significant reduction in 30-day mortality with IVIG (33.7% vs. 15.7%) [22]. The only pediatric study in this meta-analysis was the multicenter, retrospective study on IVIG in children with STSS by Adalat et al. [23]. They reported eight deaths among 29 children with STSS. All children receiving IVIG survived. This underpowered study collected data via questionnaires without validation against patient records or standardized diagnostic criteria and did not adjust for illness severity and other confounders.

Shah et al. performed a larger, more robustly designed pediatric study, which was not included in the above meta-analysis because data were collected retrospectively [16]. This multicenter cohort study included 192 US children (median age 8.2 years) with STSS between 2003 and 2007. Nearly all received clindamycin (94.8%), and 44.6% received IVIG. The overall mortality rate was 4.2% (6% in the IVIG group, 2.8% in the non-IVIG group). A strength is that the study relied on propensity scores to mitigate the risk of indication bias. First, propensity scores were computed to estimate the probability of receiving IVIG given a set of covariates. Then, IVIG recipients and non-recipients were matched on propensity score to create a pseudo-randomized population. After propensity-score matching, the mortality rate was similar in both groups (4.5%). IVIG use did not reduce the length of stay, but was associated with increased total hospital costs, primarily due to the higher drug costs. IVIG costs 45–60€/g, and the typical dose of 2 g/kg translates to roughly 100 €/kg per patient. The authors mentioned the inclusion criteria of discharge diagnoses as a limitation. This study could therefore have included non-STSS patients, less likely to receive IVIG and less likely to die. To minimize misclassification bias, the discharge codes were combined with billing data for intravenous penicillin.

To summarize, the clinical benefit of IVIG in adults and mixed-age populations did not reach statistical significance. Two pediatric studies have been conducted: one with methodological flaws, and one showing no clinical benefit, only increasing costs.

Clindamycin and IVIG: evidence from real-world practice

There is substantial variability in clindamycin and IVIG use in practice, from which valuable real-world evidence can be derived. In a European multicenter cohort of 320 children, clindamycin was used in 29% (highest in Spain and the UK, with as little as 7% in Switzerland) and IVIG in 4.6% [24]. In contrast, in an Australian surveillance study of 181 children, clindamycin use was much higher: 56% overall, rising to 80% in severe disease and 88.5% in STSS [25]. IVIG was also used more often: 8.8% overall, 20% in severe disease (including STSS), and 38.5% in STSS. Despite variability in management across settings, outcomes were similar, with mortality rates of 2% in the European study and 2.8% in the Australian study.

Discussion

While β-lactams are the cornerstone of therapy for pediatric iGAS disease, clindamycin and IVIG have been suggested as adjuncts, clindamycin mainly for its potential to inhibit toxin production, and IVIG to counteract the cytokine storm occurring in STSS.

Evidence supporting the benefit of clindamycin and IVIG in pediatric iGAS has been limited by small sample sizes, no adjustment for disease severity, and concerns about generalizability. Variations in the choice of β-lactam and the addition of vancomycin further complicate interpretation. To overcome these limitations, well-conducted multinational and multicenter pediatric studies are needed. While an RCT would be ideal, the rarity of iGAS and STSS, along with ethical concerns about withholding potentially beneficial treatments in critically ill children, presents substantial barriers. Well-conducted observational studies which make use of the increasing amount of electronic health records data can provide valuable complementary evidence. This might be the most realistic way forward, especially when RCTs might not be feasible. An example is the ongoing European PEGASUS study, which has enrolled over 2000 children with iGAS to date.

Based on the available data, the need for adjuncts in pediatric patients with iGAS may seem less urgent compared with adults. This is because mortality ranges from 2 to 8% in children with iGAS disease in general and 5 to 10% in those with STSS, versus 30 to 80% in adults. However, many children require pediatric intensive care unit admission, with rates varying between 22 and 50% [1]. This highlights the importance of considering outcomes beyond mortality in upcoming studies, such as the need for intensive care, which may justify the use of additional therapies in pediatric populations despite generally favorable survival rates.

Conclusion

Drawing firm conclusions on the use of clindamycin and IVIG as adjuncts in iGAS in children remains challenging due to limited high-quality evidence. Arguments such as “Are you not going to give everything possible if the child is critically ill?” or “Most experts use these therapies” are understandable, pending further robust evidence. This underscores the importance of larger-scale pediatric studies to guide clinical practice.