Effects of therapeutic plasma exchange (TPE) on efficacy and adverse events in patients with autoimmune neurological diseases: a systematic review and meta-analysis based on real-world studies and single-arm clinical trials.

Introduction

Background

Autoimmune neurological diseases (ANDs) are a group of disorders in which abnormal immune responses involve the central or peripheral nervous system. Their core feature is that the immune system aberrantly attacks components of the nervous system, and this process can trigger inflammation, demyelination, and neuronal damage, leading to various neurological symptoms ( 1 ). These diseases are important causes of acquired neurological deficits, limb paralysis, respiratory failure, and even death in middle-aged and young populations. In recent years, with improvements in disease recognition and advancements in immunological research, the diagnostic spectrum of ANDs has expanded to include Guillain-Barre syndrome (GBS), myasthenia gravis (MG), neuromyelitis optica spectrum disorder (NMOSD), autoimmune encephalitis (AE), and chronic inflammatory demyelinating polyneuropathy (CIDP). The pathogenesis may involve autoantibody-mediated mechanisms, cellular immune abnormalities, complement activation, release of inflammatory cytokines, and breakdown of immune tolerance ( 2 ).

Therapeutic plasma exchange (TPE) is an important immunomodulatory intervention for various antibody-mediated or immune-mediated neurological diseases. As a blood purification technique that separates and replaces plasma by centrifugation or membrane filtration ( 3 ), TPE can remove pathogenic autoantibodies, immune complexes, complement components, and inflammation-related mediators from the circulation ( 4 , 5 ), thereby relieving symptoms and controlling conditions in some neuro-immunological disorders. In clinical practice, TPE, together with intravenous immunoglobulin (IVIG), corticosteroids, and other immunotherapies, constitutes an important treatment strategy for ANDs. However, its appropriate indications and strength of evidence vary by disease type and stage ( 6 , 7 ). According to relevant guidelines, TPE has been recommended for specific clinical scenarios, like first-line treatment for GBS and short-term treatment for crisis or acute exacerbations of MG ( 8 ). Nevertheless, the real-world clinical response and safety of TPE may still be influenced by factors like baseline disease status, timing of treatment, exchange protocol, combined treatment, and procedural management across different disease subtypes, age groups, and regional healthcare settings. In particular, evidence on TPE remains relatively limited in children, old patients, and rare ANDs ( 9 ).

Although TPE is widely used in clinical practice, the currently available systematically integrated evidence regarding its real-world efficacy and safety remains relatively limited. Most existing studies are single-center retrospective or single-arm studies, and considerable heterogeneity was found across studies in terms of disease spectrum, patient age, timing of treatment, exchange protocol, and methods of reporting adverse events. In particular, systematic and comprehensive evaluations are still lacking regarding the efficacy across different disease subtypes, age groups, and populations from different geographic regions, as well as the incidence of common adverse events and death outcomes. Thus, the present study intended to systematically include globally published real-world observational studies and eligible single-arm clinical studies, conduct a systematic review and meta-analysis, evaluate the clinical response rate, mortality rate, and characteristics of adverse events in individuals with ANDs treated with TPE, and further ascertain potential differences through subgroup analyses by disease type, age, and geographic region, thereby providing evidence for the individualized and standardized application of TPE in clinical practice.

Methods

This meta-analysis was implemented as per the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. The protocol was registered with the International Prospective Register of Systematic Reviews (PROSPERO) (registration number: CRD420250652869).

Literature search strategy : PubMed, Embase, and Cochrane Library were retrieved. The initial search covered the period from January 1, 2000 to March 3, 2025. A supplementary search was subsequently performed on January 6, 2026 to include relevant studies published between March 4, 2025 and January 6, 2026, using the same search strategy and databases as the initial search, with the language restricted to English. The search strategy combined medical subject heading (MeSH) and free-text terms, primarily focusing on terms related to TPE and ANDs. Terms related to TPE included “therapeutic plasma exchange,” “plasma exchange,” “plasmapheresis,” and “plasma exchange therapy.” Terms related to ANDs encompassed “autoimmune neurological disease,” “Guillain-Barré syndrome,” “myasthenia gravis,” “neuromyelitis optica spectrum disorder,” “autoimmune encephalitis,” “chronic inflammatory demyelinating polyneuropathy,” and “stiff-person syndrome.” Study types were determined during the screening process based on the inclusion and exclusion criteria. The complete search strategy is provided in Supplementary Table S1 .

Study selection

Studies that satisfied conditions below were included: (1) Study population: individuals diagnosed with ANDs (e.g., GBS, NMOSD, MG, and stiff-person syndrome (SPS)). (2) Intervention: individuals treated with TPE. (3) Outcome measures: clinical efficacy and adverse events (e.g., hypotension, infection). (4) Study design: real-world observational studies primarily included and single-arm clinical trials additionally incorporated to supplement the evidence on the efficacy and safety of single-arm interventions.

Studies meeting conditions below were excluded: (1) Duplicate publications. (2) Non-original studies (e.g., conference abstracts), animal studies, reviews, guidelines, etc. (3) Studies published before 2000 (because early TPE techniques and peri-procedural management differed considerably from modern practice). (4) Studies that did not report the required outcome measures. (5) Studies for which the full English texts could not be obtained.

Literature screening was performed independently by two investigators (JYT and XQY), and a third reviewer (HT) consolidated the screening results.

Data extraction

Data were independently extracted by two investigators (JT and XY), and divergences were addressed by a third reviewer (HT). The following data were acquired from each eligible study: first author, year of study, sample size, geographic region, age, sex composition, disease duration, intervention measures, evaluation criteria for efficacy, definitions of response ( Supplementary Table S2 ), patient response rate, mortality rate, and types of adverse events (encompassing fever and chills, dyspnea, allergic reactions, hypocalcemia, arrhythmia, nausea and vomiting, hypertension, hypotension, catheter-related infections, deep vein thrombosis, and coagulation disorders) as well as the incidence rates of adverse events.

Risk of bias assessment

The methodological quality of the included studies was evaluated utilizing the Methodological Index for Non-Randomized Studies (MINORS) scale ( 10 ). Since most of the included studies were non-comparative, the first eight items of the MINORS scale applicable to non-comparative studies were adopted for scoring, including research with a defined purpose, patient consistency of inclusion, anticipated data collection, outcome indicator response study purpose, objectivity of outcome indicators, adequate follow-up time, lost visit rate below 5%, and estimated sample size. Each item was scored from 0 to 2: 0 for not reported, 1 for reported but inadequate, and 2 for reported and adequate. The total score ranged from 0 to 16. Quality assessment was performed independently by two investigators, and divergences were addressed through discussion. A third investigator was consulted when necessary. The results of the quality assessment are presented as total scores and item scores for each study.

Data synthesis

The primary outcome of the present study was the response rate in disease subgroups, and the secondary outcomes were the mortality rate and the incidence rate of adverse events. Statistical analyses were implemented utilizing R 4.3.3 and RStudio. For single-arm proportional data, a single-arm rate meta-analysis was adopted to pool the response rate in disease subgroups, mortality rate, and the incidence rate of adverse events, and the pooled rates with their 95% confidence intervals (CI) were calculated.

Heterogeneity across studies was appraised utilizing Cochran’s Q test and the I 2 statistic. For I 2 > 50% or the Q test p < 0.10, substantial heterogeneity was considered, and a random-effects model was preferentially adopted. When heterogeneity was low, a fixed-effects model was applied. Given the clinical heterogeneity among the included studies in terms of disease type, age structure, TPE protocol, and definition of outcomes, a random-effects model was preferentially used for the primary outcome and most safety outcomes.

For the primary outcome, subgroup analyses were carried out by disease type, age group, and study region to ascertain potential sources of heterogeneity. Disease subgroups included GBS, MG, AE, and SPS. Age subgroups were classified as <18 years, 18–60 years, and >60 years according to the mean or median ages in the original studies. Region subgroups were categorized as Asia, Europe, the Americas, and the Middle East based on the study location. Studies that included multiple diseases but did not report disease-specific efficacy separately were included only in the overall response rate analysis and not in the disease subgroup analyses.

For adverse events, the incidence rates of adverse events based on the number of patients and TPE sessions were separately extracted according to the reporting methods of the original studies. The primary results of this study were reported using the incidence rates of adverse events based on the number of patients. For rare adverse events or zero-event studies, continuity correction or appropriate transformation methods applicable to single-arm proportional data were adopted for pooled analysis.

A leave-one-out sensitivity analysis was implemented to appraise the robustness of the primary outcome and key safety outcomes. When at least 10 studies were included for a given outcome, funnel plots and Egger’s test were used to evaluate potential publication bias. If publication bias was suggested, trim-and-fill analysis was further performed to evaluate its impact on the pooled results. All statistical tests were two-sided, and p < 0.05 was considered statistically significant.

Results

Study selection

The initial search yielded 822 records, including 253 from PubMed, 464 from Embase, and 105 from Cochrane Library. After automatic duplicate checking using EndNote and manual verification, 184 duplicate records were removed, and 638 records were retained for title and abstract screening. Following title and abstract screening, 523 records that were not directly relevant to the research topic were excluded. The full texts of the remaining 115 articles were then assessed. Among these, 7 articles were excluded because the full texts could not be obtained, and another 74 articles were eliminated due to reasons such as inappropriate study design, language restriction, publication year before 2000, or failure to meet the inclusion criteria regarding study population or outcome measures. We ultimately included 34 studies during the initial search. During the supplementary search, additional 4 eligible studies were incorporated. Thus, a total of 38 studies were included finally ( 3 , 11–47 ). The literature screening process is shown in Figure 1 .

Study characteristics and quality assessment

As shown in Tables 1 , 38 studies were included in this study. All of these were retrospective cohort studies or single-arm clinical trials of non-randomized controlled trials, comprising 2,199 individuals. The age ranged from 8 months to 94 years, and regions included Europe, the Americas, Asia, and the Middle East. These studies mainly involved neuro-immunological diseases like GBS, MG, CIDP, SPS, NMOSD, multiple sclerosis, and AE. The main adverse events encompassed fever and chills, dyspnea, allergic reactions, hypocalcemia, arrhythmia, nausea and vomiting, hypertension, hypotension, catheter-related infections, deep vein thrombosis, and coagulation disorders. The primary outcome measures included the response rate in disease subgroups, mortality rate, and incidence rate of adverse events in individuals after receiving TPE.

The quality of the studies was assessed using the MINORS scale. The results are presented in the table below. The scores of the included studies ranged from 8 to 14, with the majority around 10, indicating that the overall quality of the included literature was acceptable ( Table 2 ).

Clinical response rate

A total of 38 studies were included in this analysis. Among these, 34 studies with 1,722 patients reported clinical response data for overall descriptive pooling; 31 studies with 1,676 patients reported age data for subgroup analysis. Overall heterogeneity was high ( I 2 = 76.4%) ( Supplementary Table S1 ), and thus, a random-effects model was used. Due to the high heterogeneity among studies and the definition of clinical response not completely consistent across studies, this pooled result should be interpreted with caution.

Disease subgroup analysis

Response rates were pooled according to subgroups within the same disease. After stratification by disease type, the GBS subgroup had the highest response rate, while the SPS subgroup had a relatively lower response rate. The response rate was 86% (95% CI: 64–96%, I 2 = 55.6%) for the GBS subgroup, 84% (95% CI: 64–94%, I 2 = 62.8%) for the MG subgroup, 81% (95% CI: 52–94%, I 2 = 65%) for the AE subgroup, and 65% (95% CI: 48–78%, I 2 = 0%) for the SPS subgroup. However, the differences among disease subgroups were not statistically significant ( χ 2 = 4.57, d f = 3, p = 0.2064) ( Figure 2 ).

Age subgroup analysis

Response rates were pooled according to subgroups defined by the mean or median ages in the original studies. Following stratification by age, the response rate was 83% (95% CI: 74–89%, I 2 = 50.0%) for the <18 years subgroup, 80% (95% CI: 74–85%, I 2 = 82.1%) for the 18–60 years subgroup, and 59% (95% CI: 46–71%, I 2 = 55.4%) for the >60 years subgroup. The variations among age subgroups were statistically significant ( χ 2 = 12.44, d f = 2, p = 0.0020). This suggested that the response rate in the subgroup with the mean or median age >60 years was lower than that in the subgroup with the mean or median age <18 years, possibly related to more comorbid conditions in old patients. However, this analysis was based on study-level or subgroup-level data rather than individual patients’ data. Therefore, the age-related results should still be interpreted with caution ( Figure 3 ).

Region subgroup analysis

Response rates were pooled according to subgroups defined by geographic regions in the original studies. After stratification by region, the response rate was 84% (95% CI: 80–88%, I 2 = 64.1%) for the Middle East subgroup, 80% (95% CI: 74–85%, I 2 = 38.8%) for the Asia subgroup, 75% (95% CI: 60–86%, I 2 = 78.7%) for the Europe subgroup, and 72% (95% CI: 59–82%, I 2 = 63.1%) for the Americas subgroup. The differences among region subgroups were not statistically significant ( χ 2 = 5.71, d f = 3, p = 0.1268) ( Figure 4 ).

Incidence rate of adverse events

A total of 11 types of adverse events were analyzed in the present study, encompassing fever and chills, dyspnea, allergic reactions, hypocalcemia, arrhythmia, nausea and vomiting, hypertension, hypotension, catheter-related infections, deep vein thrombosis, and coagulation disorders. Some studies documented adverse events as per TPE sessions and others as per the number of patients. The primary results of this study were reported using the incidence rates of adverse events based on the number of patients.

Overall, the pooled incidence rates of various adverse events were not high, but most outcomes showed moderate to high heterogeneity. Hypotension was the most common adverse event (12, 95% CI: 7–19%, I 2 = 93.3%), followed by coagulation disorders (11, 95% CI: 5–23%, I 2 = 80.9%), catheter-related infections (8, 95% CI: 4–15%, I 2 = 93.9%), allergic reactions (7, 95% CI: 4–14%, I 2 = 83.7%), and hypocalcemia (7, 95% CI: 3–16%, I 2 = 93.5%).

The specific pooled results for various adverse events are as follows.

Incidence rates of adverse events based on the number of patients: For fever and chills, the CI was extremely wide, rendering the estimate of little practical value. Dyspnea was reported in 3 studies with 398 patients, with a pooled incidence rate of 3% (95% CI: 0–16%, I 2 = 88.3%). Allergic reactions were documented in 12 studies involving 1,245 patients, with a pooled incidence rate of 7% (95% CI: 4–14%, I 2 = 83.7%). Hypocalcemia was recorded in 10 studies with 861 patients, with a pooled incidence rate of 7% (95% CI: 3–16%, I 2 = 93.5%). Arrhythmia was identified in 6 studies with 711 patients, with a pooled incidence rate of 2% (95% CI: 1–7%, I 2 = 73.8%). Nausea and vomiting were observed in 5 studies involving 606 patients, with a pooled incidence rate of 3% (95% CI: 1–14%, I 2 = 91.0%). Hypertension was noted in 3 studies involving 510 patients, with a pooled incidence rate of 1% (95% CI: 1–3%, I 2 = 0%). Hypotension was reported in 18 studies with 1,733 patients, with a pooled incidence rate of 12% (95% CI: 7–19%, I 2 = 93.3%). Catheter-related infections were recorded in 14 studies involving 1,644 patients, with a pooled incidence rate of 8% (95% CI: 4–15%, I 2 = 93.9%). Deep vein thrombosis was noted in 7 studies with 801 patients, with a pooled incidence rate of 3% (95% CI: 2–6%, I 2 = 50.8%). Coagulation disorders were documented in 8 studies with 621 patients, with a pooled incidence rate of 11% (95% CI: 5–23%, I 2 = 80.9%).

Incidence rates of adverse events based on TPE sessions: Fever and chills were reported in 5 studies with 1,067 exposures, with a pooled incidence rate of 4% (95% CI: 2–10%, I 2 = 86.4%). Dyspnea was recorded in 4 studies involving 1,171 exposures, with a pooled incidence rate of 2% (95% CI: 0–6%, I 2 = 77.8%). Allergic reactions were documented in 7 studies with 1,511 exposures, with a pooled incidence rate of 4% (95% CI: 2–9%, I 2 = 84.9%). Hypocalcemia was observed in 4 studies involving 982 exposures, with a pooled incidence rate of 3% (95% CI: 1–9%, I 2 = 86.5%). Arrhythmia was identified in 5 studies with 946 exposures, with a pooled incidence rate of 3% (95% CI: 1–13%, I 2 = 93.2%). Nausea and vomiting were noted in 3 studies involving 1,049 exposures, with a pooled incidence rate of 2% (95% CI: 0–9%, I 2 = 84.0%). Hypertension was reported in 3 studies with 602 exposures, with a pooled incidence rate of 3% (95% CI: 1–6%, I 2 = 45.2%). Hypotension was recorded in 9 studies with 1,935 exposures, with a pooled incidence rate of 8% (95% CI: 3–18%, I 2 = 93.3%). Catheter-related infections were documented in 6 studies involving 1,248 exposures, with a pooled incidence rate of 3% (95% CI: 2–5%, I 2 = 21.5%). Deep vein thrombosis was not reported. For coagulation disorders, the CI was extremely wide, rendering the estimate of little practical value. The summary of all adverse events is presented in Table 3 .

Mortality assessment

In total, 2,199 individuals were incorporated in the present study, with 49 deaths reported. Pooled analysis showed that the all-cause in-hospital mortality rate among patients undergoing TPE was 4% (95% CI: 3–5%), with between-study heterogeneity of I 2 = 36.5%, and this value does not represent mortality attributable to TPE. According to the original studies, most deaths were linked to the progression of the primary disease or related complications and were not definitively attributed to the TPE procedure itself. Therefore, this result reflected the overall death status in individuals receiving TPE and should not be directly interpreted as the TPE-related mortality rate ( Supplementary Figure S1 ).

Sensitivity analysis and publication bias

Overall response rate across heterogeneous individual studies

Due to significant heterogeneity in the criteria for defining clinical effectiveness across all included studies, the pooled response rate for all diseases calculated in this study is presented only for descriptive purposes. The leave-one-out sensitivity analysis uncovered that after the removal of any single study, the response rate remained stable at 79–81%, and I 2 fluctuated between 59.8 and 76.7%. This indicated that the overall results were relatively robust. In the publication bias analysis, the funnel plot and Egger’s test suggested potential publication bias ( p = 0.0076). After further adjustment using the trim-and-fill method, the pooled response rate did not change substantially. Thus, the impact of potential publication bias on the overall results might be limited ( Supplementary Figure S2 ).

Mortality rate

The leave-one-out sensitivity analysis for mortality rates demonstrated that after the elimination of any individual study, the pooled mortality rate changed only slightly, and I 2 fluctuated between 20.8–38.2%. This revealed that the overall results were relatively stable. In the publication bias analysis, the funnel plot and Egger’s test indicated potential publication bias ( p = 0.0039). After adjustment utilizing the trim-and-fill method, the pooled mortality rate did not vary considerably. Consequently, the influence of potential publication bias on the mortality results might be limited ( Supplementary Figure S3 ).

Discussion

The present study systematically integrated published real-world observational studies and eligible single-arm clinical trials to evaluate the clinical response rate, mortality rate, and incidence of adverse events associated with TPE in the treatment of ANDs. The results showed that the clinical response rates were relatively high across subgroups after TPE. This study is an exploratory descriptive analysis based on heterogeneous definitions. The pooled response rate is only shown in Supplementary Figure S4 and should not be regarded as an estimate derived from a unified response criterion. and hypotension was the most commonly reported adverse event. Given that the included studies are predominantly single-arm or retrospective in design, and due to variations in disease types, baseline patient characteristics, treatment protocols, and definition of efficacy, the findings should be interpreted as a comprehensive description of the existing evidence and do not allow direct inferences regarding the comparative efficacy of TPE relative to other treatment modalities.

This study identified the pooled response rate in disease subgroups as the primary outcome. The response rates were 86% for GBS, 84% for MG, 81% for AE, and 65% for SPS. This finding suggests that, in real-world and single-arm studies, individuals with ANDs treated with TPE show a relatively high rate of clinical improvement. This result is generally consistent with the findings of Tombak et al. and Korkmaz et al., which demonstrate varying degrees of clinical improvement in individuals with various neuroimmunological diseases following TPE ( 27 , 42 ). Disease subgroup analysis uncovered that the GBS subgroup had a numerically higher clinical response rate, while the SPS subgroup had a relatively lower response rate. However, the variations between disease subgroups were not statistically significant ( p = 0.2064). Previous studies have also indicated that GBS is one of the more common indications for TPE with a relatively robust evidence base ( 40 , 41 ). A possible reason for the good response to TPE in individuals with GBS is that its pathological process is often related to pathogenic antibodies, complement components, and inflammatory mediators in the circulation, and TPE can rapidly remove these plasma components, thereby alleviating immune-mediated nerve injury ( 48 ). For individuals with MG, TPE can clear pathogenic antibodies like acetylcholine receptor antibodies from the circulation ( 49 ) and thus has a certain clinical value in crisis or acute exacerbations of MG. Nevertheless, the differences between disease subgroups in the present study was not statistically significant, and most of the included studies were single-arm in design. Therefore, it cannot be concluded that one disease subtype is necessarily superior to another in terms of efficacy based on these findings.

Age subgroup analysis showed that the pooled response rate in the subgroups with mean or median ages <18 years was higher than that in the subgroups with mean or median ages >60 years, and the variation between subgroups was statistically significant ( p = 0.0020). This finding reflects that age group may be linked to clinical response after TPE. However, because this study was an analysis at the study level or subgroup level rather than at the level of individual patients’ data, it does not prove that age itself is an independent factor influencing the efficacy of TPE. The higher response rate in subgroups with mean or median ages <18 years may be related to factors like composition of the disease spectrum, earlier initiation of treatment, fewer underlying comorbidities, lower burden of chronic neurological damage, and better potential for neurological recovery ( 50 ). In contrast, old patients may have more underlying diseases, higher infection risk, cardiovascular instability, or more severe baseline neurological damage, and all these factors may affect clinical recovery. Thus, the age subgroup findings are more appropriately considered as hypothesis-generating observations and should be further validated in the future based on individual patients’ data or prospective cohort studies.

Region subgroup analysis demonstrated certain numerical differences in response rates across distinct regions, and a relatively higher response rate was identified in the Middle East. However, the differences between region subgroups were not statistically significant ( p = 0.1268). This finding suggests that the efficacy of TPE in different regions may be influenced by variations in disease spectrum, referral patterns, timing of treatment, healthcare resources, operational protocols, and follow-up management. Nevertheless, existing evidence is insufficient to support clear regional differences. Therefore, the region subgroup results should be regarded as trend observations reflecting real-world clinical heterogeneity rather than being interpreted as evidence of superior TPE efficacy in any particular region.

A total of 49 deaths were reported in this study. Pooled analysis showed that the all-cause in-hospital mortality rate among patients undergoing TPE was 4% (95% CI: 3–5%, I 2 = 36.5%), and this value does not represent mortality attributable to TPE. According to the original studies, most deaths were related to disease progression, respiratory failure, infection, or other serious complications, and were not explicitly attributed to the TPE procedure itself. Thus, the mortality result reported in the present study should be interpreted as an indicator reflecting the overall death status among individuals receiving TPE, rather than the TPE-related mortality rate. Due to the general lack of standardized assessment of causes of death in the included studies, this study cannot accurately evaluate the independent effect of the TPE procedure itself on mortality risk.

In terms of safety, the present study found that adverse events related to TPE or reported during treatment were predominantly mild to moderate, including hypotension, allergic reactions, catheter-related infections, hypocalcemia, and coagulation disorders. Among these, hypotension was the most common adverse event, with a pooled incidence rate of 12%. This finding is consistent with results from previous studies, which suggest that hypotension is one of the more common complications during TPE ( 43 ). TPE-related adverse events may be associated with vascular access, blood volume changes, replacement fluid type, anticoagulation regimen, and procedural process ( 39 , 51 ). For example, citrate anticoagulation may lead to hypocalcemia, blood volume fluctuations can induce hypotension, and the use of central venous catheters may increase the risk of infection or thrombosis. With standardized procedures and close monitoring, most adverse events can be promptly identified and managed. However, this does not mean that TPE is entirely without risk.

In the present study, most adverse event outcomes showed moderate to high heterogeneity, which may be attributed to the following factors. First, the included studies were predominantly real-world observational studies and single-arm trials, leading to inconsistencies in patient selection, treatment indications, and procedural protocols. Second, there was a substantial variation in disease subtypes, and individuals with different diseases differed in levels of disease severity, immune activity status, and risk of complications. Third, TPE treatment protocols differed across studies, including the number of sessions, exchange volume, replacement fluid type, anticoagulation method, vascular access, and combined treatment regimens. Fourth, the definitions of clinical response and adverse events, as well as the intensity of monitoring and methods of reporting, varied among studies. Fifth, differences in healthcare resources, level of care, and experience with TPE procedures may exist across geographic regions. In this study, heterogeneity was examined through subgroup analyses by disease, age, region, and statistical approach of adverse events, and the robustness of the results was appraised using leave-one-out sensitivity analysis.

This study has several strengths. First, it systematically integrated real-world evidence from individuals with various ANDs treated with TPE, covering different disease types, age groups, and geographic regions, which helps reflect the application of TPE in clinical practice. Second, the present study simultaneously evaluated the response rate in disease subgroups, mortality rates, and multiple adverse events, providing a relatively comprehensive profile of the efficacy and safety of TPE. Third, subgroup analyses by disease type, age, region, and statistical approach of adverse events, as well as sensitivity analysis and publication bias assessment, enhance the interpretative robustness of the findings. However, because most of the included studies were single-arm or retrospective in design, these strengths cannot replace the causal inference evidence provided by high-quality controlled studies.

The present study also has several limitations. First, most of the included studies were retrospective or single-arm in design, lacking a control group. This precludes direct comparisons of the relative efficacy and safety of TPE and IVIG, corticosteroids, immuno-adsorption, or other treatment strategies. Second, the definition of clinical response varied across studies, and some studies used disease-specific criteria, symptom relief, or physician judgment as measures of efficacy, which may affect the comparability of the pooled response rate. Third, reporting methods for adverse events were inconsistent. Some studies reported adverse events as per the number of patients, while others reported such events as per TPE sessions. Fourth, TPE was often used in combination with corticosteroids, IVIG, immuno-suppressants, or intensive supportive care, and this study could not distinguish the independent contribution of TPE itself and that of combined therapies to clinical improvement. Fifth, there were differences across studies in the timing of TPE initiation, number of sessions, exchange volume, replacement fluid type, anticoagulation method, and choice of vascular access. However, due to insufficient data, adequate subgroup analyses based on these procedural parameters could not be performed. Sixth, publication bias analysis suggested potential publication bias for the primary outcome and mortality rates. Although the pooled results did not change substantially after adjustment by the trim-and-fill method, the findings should be interpreted with caution. Finally, the MINORS scores indicated that most included studies were moderate in methodological quality, which may lower the certainty of the evidence. Thus, it should not be assumed that the moderate quality has no impact on the reliability of the primary outcome.

Conclusion

In conclusion, existing real-world observational studies and eligible single-arm clinical trials indicate that patients with ANDs show a relatively high rate of clinical improvement after TPE. The reported adverse events are predominantly manageable events such as hypotension, suggesting acceptable overall safety. The deaths reported in the included studies are mostly related to the progression of the primary disease or related complications rather than being explicitly attributed to the TPE procedure itself. Given that most of the included studies are single-arm or retrospective in design and inconsistencies exist in efficacy definitions and safety reporting across studies, more high-quality, multicenter studies are needed in the future to further define the appropriate patient population, optimal treatment regimen, and long-term safety of TPE.

Edited by: Marija Mostarica-Stojkovic , University of Belgrade, Serbia

Reviewed by: Sanja Gluscevic , Clinical Center of Montenegro, Montenegro

Ruta Kaladyte Lokominiene , Vilnius University, Lithuania

Data availability statement

The original contributions presented in the study are included in the article/ Supplementary material , further inquiries can be directed to the corresponding authors.

Author contributions

JT: Data curation, Investigation, Visualization, Writing – original draft. XY: Data curation, Investigation, Writing – original draft. HT: Data curation, Formal analysis, Validation, Writing – original draft. JH: Software, Visualization, Writing – original draft. SG: Conceptualization, Funding acquisition, Methodology, Writing – review & editing. JL: Conceptualization, Project administration, Supervision, Writing – review & editing.

Conflict of interest

The author(s) declared that this work was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Generative AI statement

The author(s) declared that Generative AI was not used in the creation of this manuscript.

Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.

Publisher’s note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

Supplementary material

The Supplementary material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/fneur.2026.1916922/full#supplementary-material

References

1. Bobati SS Naik KR . Therapeutic plasma exchange—an emerging treatment modality in patients with neurologic and non-neurologic diseases . J Clin Diagn Res . ( 2017 ) 11 : EC35 – 7 . doi: 10.7860/jcdr/2017/27073.10480 , 28969140 PMC5620780

2. Segal Y Soltys J Clarkson BDS Howe CL Irani SR Pittock SJ . Toward curing neurological autoimmune disorders: biomarkers, immunological mechanisms, and therapeutic targets . Neuron . ( 2025 ) 113 : 345 – 79 . doi: 10.1016/j.neuron.2024.12.006 , 39809275 PMC13102114

3. Moser T Harutyunyan G Karamyan A Otto F Bacher C Chroust V . Therapeutic plasma exchange in multiple sclerosis and autoimmune encephalitis: a comparative study of indication, efficacy and safety . Brain Sci . ( 2019 ) 9 : 267 . doi: 10.3390/brainsci9100267 , 31601005 PMC6827054

4. Ranganathan D John GT . Therapeutic plasma exchange in renal disorders . Indian J Nephrol . ( 2019 ) 29 : 151 – 9 . doi: 10.4103/ijn.IJN_420_17 , 31142960 PMC6521768

5. Pusey CD Levy JB . Plasmapheresis in immunologic renal disease . Blood Purif . ( 2012 ) 33 : 190 – 8 . doi: 10.1159/000334155 , 22269818

6. Lindner M Klotz L Wiendl H . Mechanisms underlying lesion development and lesion distribution in CNS autoimmunity . J Neurochem . ( 2018 ) 146 : 122 – 32 . doi: 10.1111/jnc.14339 , 29574788

7. Rommer PS Milo R Han MH Satyanarayan S Sellner J Hauer L . Immunological aspects of approved MS therapeutics . Front Immunol . ( 2019 ) 10 : 1564 . doi: 10.3389/fimmu.2019.01564 , 31354720 PMC6637731

8. Connelly-Smith L Alquist CR Aqui NA Hofmann JC Klingel R Onwuemene OA . Guidelines on the use of therapeutic apheresis in clinical practice—evidence-based approach from the Writing Committee of the American Society for Apheresis: the ninth special issue . J Clin Apher . ( 2023 ) 38 : 77 – 278 . doi: 10.1002/jca.22043 , 37017433

9. Fleurance A Joseph L Dehoux L Barnerias C Hully M Kossorotoff M . Efficacy and safety of plasmapheresis in children with acute transverse or flaccid myelitis, and Guillain-Barré syndrome . Eur J Neurol . ( 2025 ) 32 : e70406 . doi: 10.1111/ene.70406 , 41251130 PMC12624416

10. Slim K Nini E Forestier D Kwiatkowski F Panis Y Chipponi J . Methodological index for non-randomized studies (minors): development and validation of a new instrument . ANZ J Surg . ( 2003 ) 73 : 712 – 6 . doi: 10.1046/j.1445-2197.2003.02748.x , 12956787

11. Ali T Hkima Abou Fakher F Alawir M Allababidi A Sheikh Hasan A . Management of severe pediatric Guillain-Barré syndrome in a low-income country: efficacy and safety of therapeutic plasma exchange in pediatric patients: a retrospective study . BMC Pediatr . ( 2024 ) 24 : 818 . doi: 10.1186/s12887-024-05298-6 , 39695474 PMC11657456

12. Błasiak M Korohoda P Zachwieja K Drożdż D Gergont A Madej-Świątkowska K . Safety and efficacy of therapeutic plasma exchange in pediatric neuro-immunological diseases . Adv Med Sci . ( 2025 ) 70 : 86 – 93 . doi: 10.1016/j.advms.2025.01.003 , 39824387

13. Bava D Panda S Jain R Mittal S Gupta A Khan A . Therapeutic plasma exchange in autoimmune neurological disorders: a comprehensive evaluation of clinical outcomes, safety, and long-term disability using the modified Rankin scale . Ther Apher Dial . ( 2025 ) 29 : 114 – 22 . doi: 10.1111/1744-9987.14189 , 39053908

14. Carandina-Maffeis R Nucci A Marques JF Jr Roveri EG Pfeilsticker BH Garibaldi SG . Plasmapheresis in the treatment of myasthenia gravis: retrospective study of 26 patients . Arq Neuropsiquiatr . ( 2004 ) 62 : 391 – 5 . doi: 10.1590/s0004-282x2004000300003 15273832

15. Chen J Feng L Li S Wang H Huang X Shen C . Therapeutic plasma exchange in AChR-ab positive generalized myasthenia gravis: a real world study about its early response . J Inflamm Res . ( 2024 ) 17 : 2299 – 308 . doi: 10.2147/jir.S455104 , 38645879 PMC11032135

16. Dogra A Rana K Rathod C Prakash S . Outcome of therapeutic plasma exchange in myasthenia gravis patients . J Family Med Prim Care . ( 2020 ) 9 : 5971 – 5 . doi: 10.4103/jfmpc.jfmpc_1026_20 , 33681028 PMC7928125

17. Dada MA Kaplan AA . Plasmapheresis treatment in Guillain-Barré syndrome: potential benefit over IVIg in patients with axonal involvement . Ther Apher Dial . ( 2004 ) 8 : 409 – 12 . doi: 10.1111/j.1526-0968.2004.00183.x , 15663537

18. Gala-Błądzińska A Mazur K Dębiec A Gargasz K Bartosik-Psujek H . Safety and tolerability of therapeutic plasma exchange in autoimmune neurological diseases - a retrospective single-Centre analysis . Neurol Neurochir Pol . ( 2020 ) 54 : 344 – 9 . doi: 10.5603/PJNNS.a2020.0045 , 32557528

19. Kaya E Keklik M Sencan M Yilmaz M Keskin A Kiki I . Therapeutic plasma exchange in patients with neurological diseases: multicenter retrospective analysis . Transfus Apher Sci . ( 2013 ) 48 : 349 – 52 . doi: 10.1016/j.transci.2013.04.015 , 23619327

20. Kumar R Birinder SP Gupta S Singh G Kaur A . Therapeutic plasma exchange in the treatment of myasthenia gravis . Indian J Crit Care Med . ( 2015 ) 19 : 9 – 13 . doi: 10.4103/0972-5229.148631 , 25624644 PMC4296418

21. Kumawat V Tripathi PP M N Mahadevan A Vittal AV . Therapeutic role of plasma exchange in the management of stiff person syndrome: experience from a tertiary care centre . Hematol Transfus Cell Ther . ( 2024 ) 46 : 443 – 9 . doi: 10.1016/j.htct.2023.11.014 , 38614933 PMC11451340

22. Lehmann HC Hoffmann FR Fusshoeller A Meyer zu Hörste G Hetzel R Hartung HP . The clinical value of therapeutic plasma exchange in multifocal motor neuropathy . J Neurol Sci . ( 2008 ) 271 : 34 – 9 . doi: 10.1016/j.jns.2008.02.022 , 18485370

23. Seyhanli A Yavuz B Selimoglu I Sengun IS Aslan AT Ozsan GH . Therapeutic plasma exchange in neurological diseases: eleven years experience at a tertiary care center in Turkey . Ther Apher Dial . ( 2022 ) 26 : 465 – 70 . doi: 10.1111/1744-9987.13703 , 34173719

24. Shah YD Eksambe P Fomani K Louie J Shefali K Kothare S . Feasibility & safety of plasma exchange in paediatric neuro-immunology: a single center experience . Eur J Paediatr Neurol . ( 2020 ) 27 : 94 – 7 . doi: 10.1016/j.ejpn.2020.04.001 , 32307218

25. Sheckley H Malhotra K Katyal N Narula N Govindarajan R . Clinical experience with maintenance therapeutic plasma exchange in refractory generalized myasthenia gravis . J Clin Apher . ( 2021 ) 36 : 727 – 36 . doi: 10.1002/jca.21923 , 34241920

26. Sinanović O Zukić S Burina A Pirić N Hodžić R Atić M . Plasmapheresis in neurological disorders: six years experience from university clinical center Tuzla . F1000Res . ( 2017 ) 6 : 1234 . doi: 10.12688/f1000research.11841.1 , 28928949 PMC5590081

27. Tombak A Uçar MA Akdeniz A Yilmaz A Kaleagası H Sungur MA . Therapeutic plasma exchange in patients with neurologic disorders: review of 63 cases . Indian J Hematol Blood Transfus . ( 2017 ) 33 : 97 – 105 . doi: 10.1007/s12288-016-0661-3 , 28194064 PMC5280851

28. Yamada C Teener JW Davenport RD Cooling L . Maintenance plasma exchange treatment for muscle specific kinase antibody positive myasthenia gravis patients . J Clin Apher . ( 2015 ) 30 : 314 – 9 . doi: 10.1002/jca.21377 , 25557356

29. Yıldırım M Bektaş Ö Botan E Şahin S Gurbanov A Teber S . Therapeutic plasma exchange in clinical pediatric neurology practice: experience from a tertiary referral hospital . Clin Neurol Neurosurg . ( 2021 ) 207 : 106823 . doi: 10.1016/j.clineuro.2021.106823 , 34304066

30. Zhang Y Huang HJ Chen WB Liu G Liu F Su YY . Clinical efficacy of plasma exchange in patients with autoimmune encephalitis . Ann Clin Transl Neurol . ( 2021 ) 8 : 763 – 73 . doi: 10.1002/acn3.51313 , 33609012 PMC8045938

31. Totzeck A Jahn M Stolte B Thimm A Kleinschnitz C Hagenacker T . Total plasma exchange in neuromuscular junction disorders-a single-center, retrospective analysis of the efficacy, safety and potential diagnostic properties in doubtful diagnosis . J Clin Med . ( 2022 ) 11 : 4383 . doi: 10.3390/jcm11154383 , 35955999 PMC9369332

32. Olsen GM Tormey CA Tseng B Hendrickson JE Sostin N . Therapeutic plasma exchange for peripheral neuropathy associated with trisulfated heparan disaccharide IgM antibodies: a case series of 17 patients . J Clin Apher . ( 2022 ) 37 : 13 – 8 . doi: 10.1002/jca.21944 , 34698404

33. Codron P Cousin M Subra JF Pautot V Letournel F Verny C . Therapeutic plasma exchange in chronic dysimmune peripheral neuropathies: a 10-year retrospective study . J Clin Apher . ( 2017 ) 32 : 413 – 22 . doi: 10.1002/jca.21530 , 28247959

34. Vikrant S Thakur S Sharma A Gupta D Sharma S . Safety and efficacy of therapeutic membrane plasmapheresis in the treatment of Guillain-Barré syndrome: a study from a tertiary care hospital from India . Neurol India . ( 2017 ) 65 : 527 – 31 . doi: 10.4103/neuroindia.NI_907_15 , 28488614

35. Pham ST Nguyen KV Le LTK Khoi NM Trinh TH Nguyen LH . Outcomes of therapeutic plasma exchange among children with autoimmune encephalitis in Vietnam . J Child Neurol . ( 2026 ) 41 : 385 – 95 . doi: 10.1177/08830738251356122 , 40746120

36. Jamal YA Yigitbilek F Yunce M . Therapeutic apheresis for neurological diseases: a five-year retrospective study at a large academic medical center . J Clin Apher . ( 2025 ) 40 : e70035 . doi: 10.1002/jca.70035 , 40491040

37. Gupta S Tiwari AK Aggarwal G Bhardwaj G Pawar S Goyal V . Effectiveness and safety of therapeutic plasma exchange in neurological disorders; measuring the response to treatment using modified Rankin scale—a single-center study . Asian J Transfus Sci . ( 2025 ) 19 : 187 – 92 . doi: 10.4103/ajts.ajts_90_24 , 41244193 PMC12614707

38. Eichinger K Breu M Renken M Siegert S Hilz E Glatter S . Complications of therapeutic plasma exchange in pediatric neuroimmune disorders . Children . ( 2025 ) 12 : 1457 . doi: 10.3390/children12111457 , 41300575 PMC12651235

39. Aljezani MA Althubaiti F Alhamed L Alharthi A Alamoodi A Bakheet Y . Plasma exchange in pediatric neurology patients: a single-center experience . Cureus . ( 2024 ) 16 : e52691 . doi: 10.7759/cureus.52691 , 38384617 PMC10879651

40. Gafoor VA Jose J Saifudheen K Musthafa M . Plasmapheresis in neurological disorders: experience from a tertiary care hospital in south India . Ann Indian Acad Neurol . ( 2015 ) 18 : 15 – 9 . doi: 10.4103/0972-2327.144301 , 25745304 PMC4350207

41. Kaynar L Altuntas F Aydogdu I Turgut B Kocyigit I Hacioglu SK . Therapeutic plasma exchange in patients with neurologic diseases: retrospective multicenter study . Transfus Apher Sci . ( 2008 ) 38 : 109 – 15 . doi: 10.1016/j.transci.2007.11.002 , 18331814

42. Korkmaz G Dagdas S Saltoglu T Ceran F Aydın MS Bektas H . Effectiveness and safety of therapeutic plasma exchange in neurological diseases: an 11-year report from a tertiary care center . Ther Apher Dial . ( 2025 ) 29 : 312 – 20 . doi: 10.1111/1744-9987.14223 , 39500332

43. Fateen T Sultana N Sarwar M Saqlain N . Complications of therapeutic plasma exchange in pediatric patients: an experience at a tertiary care hospital . Pak J Med Sci . ( 2023 ) 39 : 994 – 8 . doi: 10.12669/pjms.39.4.7002 , 37492302 PMC10364278

44. Mercure-Corriveau N Roy S Hu C Crowe EP Zhu X Obando D . Therapeutic plasma exchange in the management of stiff person syndrome spectrum disorders: a case series and review of the literature . Ther Adv Neurol Disord . ( 2023 ) 16 : 17562864231180736 . doi: 10.1177/17562864231180736 , 37529719 PMC10387686

45. Naik A Prakash S Ray GK Mukherjee S . Variable response to therapeutic plasma exchange in pediatric anti-NMDA receptor encephalitis . Transfus Clin Biol . ( 2021 ) 28 : 287 – 90 . doi: 10.1016/j.tracli.2021.03.006 , 33839297

46. Nieto-Aristizábal I Vivas Á J Ruiz-Montaño P Aragón CC Posso-Osorio I Quiñones J . Therapeutic plasma exchange as a treatment for autoimmune neurological disease . Autoimmune Dis . ( 2020 ) 2020 : 3484659 . doi: 10.1155/2020/3484659 32802495 PMC7415086

47. Pagano MB Murinson BB Tobian AA King KE . Efficacy of therapeutic plasma exchange for treatment of stiff-person syndrome . Transfusion . ( 2014 ) 54 : 1851 – 6 . doi: 10.1111/trf.12573 , 24527774

48. Luo MC Wang WF Yin WF Li Y Li BJ Duan WW . Clinical efficacy and mechanism of lymphoplasma exchange in the treatment of Guillain-Barre syndrome . Cell Mol Biol . ( 2017 ) 63 : 106 – 15 . doi: 10.14715/cmb/2017.63.10.17 , 29096750

49. Newsom-Davis J Pinching AJ Vincent A Wilson SG . Function of circulating antibody to acetylcholine receptor in myasthenia gravis: investigation by plasma exchange . Neurology . ( 1978 ) 28 : 266 – 72 . doi: 10.1212/wnl.28.3.266 , 564482

50. Sun F He Z . Neuronal intrinsic barriers for axon regeneration in the adult CNS . Curr Opin Neurobiol . ( 2010 ) 20 : 510 – 8 . doi: 10.1016/j.conb.2010.03.013 , 20418094 PMC2911501

51. Nguyen TC Kiss JE Goldman JR Carcillo JA . The role of plasmapheresis in critical illness . Crit Care Clin . ( 2012 ) 28 : 453 – 68 . doi: 10.1016/j.ccc.2012.04.009 , 22713617 PMC3381605

Glossary

autoimmune neurological diseases

Guillain-Barre syndrome

myasthenia gravis

multiple sclerosis

neuromyelitis optica spectrum disorder

chronic inflammatory demyelinating polyneuropathy

acute inflammatory demyelinating polyneuropathy (subtype of GBS)

acute disseminated encephalomyelitis

transverse myelitis

stiff-person syndrome

multifocal motor neuropathy

autoimmune encephalitis

paraneoplastic syndrome

chronic dysimmune peripheral neuropathy

systemic lupus erythematosus encephalitis

neuromyelitis optica

optic neuritis

myelin oligodendrocyte glycoprotein antibody-associated disease

trisulfated heparan disaccharide IgM antibody-associated peripheral neuropathy

anti-N-methyl-D-aspartate receptor encephalitis

anti-gamma-aminobutyric acid type B receptor encephalitis

anti-leucine-rich glioma-inactivated 1 protein antibody-associated encephalitis

anti-alpha-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid receptor encephalitis

voltage-gated potassium channel antibody diseases

systemic lupus erythematosus-related encephalitis

fresh frozen plasma

human albumin

intravenous immunoglobulin

hydroxyethyl starch

succinylated gelatin