Evaluation of the effect of therapeutic plasma exchange on serum cytokine levels in pediatric intensive care unit

Abstract

Introduction

Therapeutic plasma exchange (TPE) is crucial for saving lives when used appropriately. This study aimed to assess TPE's impact on tumor necrosis factor‐like weak inducer of apoptosis (TWEAK) protein and IL‐6 levels in critically ill pediatric patients.

Methods

Conducted between May 2022 and December 2022, the study observed pediatric intensive care unit (PICU) patients undergoing TPE, recording demographics, lab results, TWEAK, and IL‐6 levels pre‐ and post‐procedure.

Results

Of 41 patients, 53.7% were male, 51.2% had underlying conditions, and 39% showed sepsis symptoms. IL‐6 levels significantly rose post‐TPE (p: 0.006), while TWEAK protein levels dropped (p: 0.030). Positive correlations were found between interleukin‐6 (IL‐6) pre‐TPE and ventilation duration, ferritin levels (p <0.05), and TWEAK pre‐TPE and organ failure indicators, D‐Dimer levels (p <0.05). Prognosis showed no significant difference in IL‐6/TWEAK levels (p >0.05).

Conclusion

Serum IL‐6 and TWEAK can indicate disease severity and inflammation level, but may not predict prognosis accurately.

Clinical Trial Registration

Our study has clinical study registration number B.10.1.TKH.4.34.H.GP.0.01/154 of the University of Health Sciences Ümraniye Training and Research Hospital.

1. INTRODUCTION

Therapeutic plasma exchange is a life‐saving method when performed in appropriate indications. Although it is effectively and safely used in adults, its use in the pediatric population remains limited due to factors such as vascular access restrictions, side effects, low blood volume, procedure‐related complications, patient compliance, technical challenges, and the absence of universally accepted indications. However, there is growing experience indicating that therapeutic plasma exchange performed for suitable indications is also safe in pediatric patients. In recent times, an increase in the utilization of therapeutic plasma exchange has been observed in pediatric intensive care units across many medical centers.

IL‐6 is a multifunctional cytokine with a wide range of activities. It is produced and released by various cell types in response to infections, injuries, or other inflammatory stimuli. Some of the most significant sources of IL‐6 include T cells, macrophages, monocytes, dendritic cells, endothelial cells, and fibroblasts. IL‐6 plays a crucial role in regulating both the immune system and the nervous system. Additionally, it is involved in processes like liver regeneration and metabolic control within the body.

TPE assists patients in regulating their own immune systems by removing inflammatory cytokines, antibodies, and toxins from their circulation. In recent years, the use of TPE has become increasingly widespread in pathological conditions characterized by elevated IL‐6 levels. Such conditions include COVID‐19, Multisystem Inflammatory Syndrome in Children (MIS‐C), Macrophage Activation Syndrome (MAS), Hemophagocytic Lymphohistiocytosis (HLH), and sepsis, where reducing IL‐6 levels can be particularly beneficial [1, 2].

TWEAK, a member of the tumor necrosis factor (TNF) ligand family, is a multifunctional cytokine widely distributed in cell types and tissues [3].

TWEAK, a cytokine from the TNF family, stimulates the release of proinflammatory cytokines by binding to its receptor, factor‐inducible 14 (Fn14). Initially recognized as an apoptosis inducer, TWEAK's interaction with Fn14 has been linked to various pathological processes, including angiogenesis, carcinogenesis, and inflammation. Its expression increases during vascular and renal injuries, where transient activation of the TWEAK‐Fn14 axis aids tissue repair and regeneration. However, excessive activation can lead to harmful inflammation and cell death [4]. Recent studies suggest that the TWEAK‐Fn14 axis may play a role in cancer, chronic autoimmune diseases (like rheumatoid arthritis and systemic lupus erythematosus), multiple sclerosis, and acute stroke. Inhibiting this axis could slow disease progression across various conditions. Notably, anti‐TWEAK therapy has shown benefits in preclinical models of autoimmune diseases, and lower serum TWEAK levels correlate with better outcomes, indicating its potential as a therapeutic target in COVID‐19 and similar infections, although these findings remain speculative. In the literature, there are very limited studies on TWEAK protein in pediatric patients.

The aim of this study was to investigate the risk factors determining prognosis in critically ill pediatric patients undergoing therapeutic plasma exchange for various reasons and to examine the effect of therapeutic plasma exchange on TWEAK protein and IL‐6 levels.

2. MATERIALS AND METHODS

The study was conducted prospectively and single‐center in patients hospitalized in the pediatric intensive care unit (PICU). Informed consent forms were obtained from all participants. Forty‐one patients who underwent TPE between May 2022 and December 2022 were included in the study. Inclusion criteria for the study were being between 1 month and 17 years of age, staying in the PICU for at least 24 h, having sufficient file data, undergoing at least one session of TPE, and not receiving immunosuppressive therapy. Patients who stayed in the PICU for less than 24 h had insufficient file data, did not provide consent to participate, received immunosuppressive therapy, or had congenital or acquired immunodeficiency were excluded from the study.

Patients' age (months), gender, body weight (kg), comorbid status, pediatric mortality risk score (PRISM), pediatric logistic organ dysfunction score (PELOD), organ failure index (OFI), and indication for the procedure within the first 24 h of hospitalization, complications, replacement fluid used during the procedure, number of sessions, additional treatments administered, presence of infection, number of PICU days, mechanical ventilator requirement and number of days, and need for inotropic drug infusion and prognosis were recorded. The diagnosis of sepsis was made according to the criteria defined in the “International Guidelines for the Management of Septic Shock and Sepsis‐Associated Organ Dysfunction in Children” from the Surviving Sepsis Campaign [5]. Pre‐ and post‐procedure laboratory data complete blood count, blood urea nitrogen (BUN), creatinine, serum calcium, lactate dehydrogenase (LDH), alanine aminotransferase (ALT), aspartate aminotransferase (AST), lactate, c reactive protein (CRP), procalcitonin, ferritin, international normalized ratio (INR), IL‐6, TWEAK levels were recorded. Laboratory data before the first plasmapheresis session were defined as baseline, and laboratory data after the last plasmapheresis session were defined as endpoint.

IL‐6 and TWEAK levels were measured using enzyme‐linked immunosorbent assay (ELISA) kits, specifically designed as 96‐well plates. Blood samples were collected and centrifuged at 1500 g for 10 min at 4°C to separate the plasma, which was subsequently stored at −80°C until analysis. The IL‐6 levels were assessed using the Human Interleukin 6 (IL‐6) ELISA Kit, which operates within a measurement range of 2–600 ng/L and possesses a sensitivity of 0.092 ng/L. This kit is suitable for serum, plasma, and other biological fluid samples, and includes detailed user application information. The product has a shelf life of at least 6 months and is delivered under appropriate conditions by the importing company. Furthermore, the ELISA kits were utilized in the manufacturer's laboratory, where they provided researchers access to an ELISA reader, ELISA washer, 8‐channel pipette, single‐channel pipette, and all necessary supplies and equipment. The laboratory was open to researchers, ensuring assistance from at least two staff members for optimal working conditions and support in calculating results. TWEAK levels were measured using the Human TWEAK ELISA Kit, also designed as a 96‐well plate and capable of reading at a wavelength of 450 nm. The TWEAK kit has a measurement range of 10–4000 ng/mL and a sensitivity of 5.51 ng/mL. Similar to the IL‐6 kit, it is intended for use with serum, plasma, and other biological fluids, features comprehensive user instructions, and maintains a minimum shelf life of 6 months, with appropriate handling during delivery.

For therapeutic plasma exchange, the internal jugular vein, subclavian vein, or femoral vein was used for venous catheterization. The procedure was conducted at the bedside using the PrismaFlex© device (Gambro, Lund, Sweden). Fresh frozen plasma (FFP) was the preferred replacement fluid, and 5% human albumin was administered to patients experiencing allergic reactions to plasma. The objective was to replace 1–1.5 times the estimated plasma volume. In small children (typically weighing less than 10 kg) whose extracorporeal volume exceeded 10% of the total blood volume, the circuit was primed with red blood cells containing 0.9% normal saline before commencing the TPE procedure. The blood flow rate was adjusted within the range of 10–150 mL/min based on the patient's body weight. The formula “70 × weight (kg) × [1‐Hct]” was utilized to estimate the plasma volume. In the study, TPE was performed using membrane filtration‐based plasma exchange. This method involved the use of a semipermeable membrane to filter out plasma from the blood.

2.1. Statistical analysis

Power analysis, performed with the G*Power (v3.1.9) program, determined the sample size. The study's power was expressed as 80%, with β representing the probability of a type II error. Assuming an effect size (d = 0.278) based on Cohen's effect size coefficients, it was determined that a minimum of 40 samples is required. IBM Statistical Package for the Social Sciences 22 (IBM SPSS, Turkey) was utilized for the statistical analysis of the study's findings. The normal distribution of parameters was assessed using the Kolmogorov–Smirnov test (with Lilliefors correction) and the Shapiro–Wilk test. Descriptive statistical methods, including minimum, maximum, mean, standard deviation, median, and frequency, were employed. For the comparison of quantitative data between groups with normally distributed parameters, the One‐way ANOVA test was applied. Student's t‐test was used for comparisons between two groups with normally distributed parameters, while the Mann–Whitney U test was used for comparisons between two groups with non‐normally distributed parameters. Intragroup comparisons of normally distributed parameters were performed using the paired‐sample t‐test, and intragroup comparisons of non‐normally distributed parameters were conducted with the Wilcoxon signed‐rank test. Qualitative data were compared using Fisher's Exact Chi‐Square test and Continuity (Yates) Correction. Pearson's correlation analysis examined relationships between parameters conforming to normal distribution, and Spearman's rho correlation analysis explored relationships between parameters not conforming to normal distribution. Significance was assessed at the p <0.05 level.

3. RESULTS

The study was performed with a total of 41 patients, 53.7% of whom were male, aged between 2 and 191 months. The mean age was 91.07 ± 67.74 months and the mean body weight was 28.46 ± 20.09 kg. The number of days of PICU hospitalization ranged between 3 and 170 days, with a mean of 25.68 ± 29.79 days (Table 1).

TABLE 1.

Demographic characteristics of patients.

	Minimum	Maximum	Mean ± standard deviation
Age (month)	2	191	91.07 ± 67.74
Weight (kg)	5	90	28.46 ± 20.09
Hearth rate/minute	50	180	111.41 ± 34.33
Systolic blood pressure (mmHg)	40	149	98.85 ± 23.72
Diastolic blood pressure (mmHg)	20	96	61.83 ± 17.96
Respiratory rate/minute	14	42	28.17 ± 6.95
Days of PICU hospitalization	3	170	25.68 ± 29.79
Gender n (%)
Male	22	53.7%
Female	19	46.3%

Abbreviation: PICU, pediatric intensive care unit.

When the hospitalization diagnoses of the patients were evaluated, it was observed that MIS‐C patients were the first, followed by sepsis (24.4%), MAS/HLH (22%), and neurological diseases (17.1%), respectively. The mean PRISM score of the patients was 8.10 ± 10.70. There was no statistically significant change in PELOD and OFI scores before and after TPE (p <0.05), but the vasoactive‐inotropic score (VIS) after TPE was significantly higher than before TPE (p = 0.028) (Table 2). Twenty‐one (51.2%) of the patients had underlying diseases, 16 (39%) had sepsis clinic, and 15 (36.6%) were receiving Non‐Invasive Ventilation/Invasive Mechanical Ventilation (NIV/IMV) support. The survival rate of the patients was 68.3%, evaluated only until the time of discharge (Table 3).

TABLE 2.

Comparison of clinical characteristics and laboratory parameters of patients according to prognosis.

		Alive	Dead
		Mean ± SD (median)	Mean ± SD (median)	p
Age (month)		85.04 ± 62.94 (69.5)	104.08 ± 78.19 (155)	0.458 a
Number of PICU days		24.11 ± 34.5 (10)	29.08 ± 16.2 (26)	0.066 a
Number of organ failure		1.11 ± 1.29 (1)	3.69 ± 1.6 (3)	0.001* a
PRISM		4.46 ± 4.5 (4)	15.92 ± 15.46 (10)	0.001* a
VIS	Before TPE	7 ± 21.88 (0)	71.46 ± 74.87 (50)	0.001* a
	Post TPE	0.29 ± 1.51 (0)	172.54 ± 97.72 (230)	0.001* a
PELOD	Before TPE	3.75 ± 4.59 (1.5)	17.69 ± 15.74 (11)	0.001* a
	Post TPE	1.75 ± 5.89 (0)	24.62 ± 12.67 (23)	0.001* a
OFI	Before TPE	0.82 ± 1.02 (0.5)	3.23 ± 1.54 (3)	0.001* a
	Post TPE	0.11 ± 0.42 (0)	4.15 ± 1.46 (4)	0.001* a
IL 6 (ng/L)	Before TPE	353.24 ± 188.11 (347.9)	372 ± 155.72 (372.7)	0.756 b
	Post TPE	404.57 ± 170.78 (427.7)	449.36 ± 182.23 (441)	0.449 b
TWEAK (mg/L)	Before TPE	2945.34 ± 1216.82 (3828.3)	2949.61 ± 1204.24 (3793.7)	0.992 b
	Post TPE	2747.66 ± 1110.79 (2803.5)	2635.86 ± 1039.23 (2664.3)	0.761 b
ALT (U/L)	Before TPE	420.61 ± 834.35 (71)	265.46 ± 389.23 (49)	0.758 a
	Post TPE	49.21 ± 56.7 (27.5)	141.85 ± 205.7 (51)	0.034* a
AST (U/L)	Before TPE	707.14 ± 1341.37 (118.5)	294.08 ± 431.79 (81)	0.844 a
	Post TPE	33.5 ± 18.91 (31.5)	325.54 ± 702.74 (121)	0.001* a
Urea (mg/dL)	Before TPE	39.25 ± 39.42 (23.1)	41.61 ± 27.62 (36)	0.327 a
	Post TPE	23.48 ± 14.53 (19.4)	27.44 ± 20.18 (25)	0.575 b
Creatinine (mg/dL)	Before TPE	0.58 ± 0.65 (0.4)	0.74 ± 0.52 (0.7)	0.120 a
	Post TPE	0.36 ± 0.28 (0.3)	0.49 ± 0.33 (0.4)	0.054 a
CRP (mg/L)	Before TPE	39.62 ± 69.04 (15.2)	58.98 ± 73.24 (17.5)	0.484 a
	Post TPE	4.56 ± 6.45 (1.5)	82.03 ± 80.26 (30.6)	0.001* a
Procalcitonin (ng/mL)	Before TPE	18.94 ± 32.76 (1.5)	32.81 ± 43.36 (6.4)	0.222 a
	Post TPE	21.03 ± 67.17 (0.2)	23.92 ± 32.06 (4.4)	0.001* a
WBC (Leukocyte)	Before TPE	12498.93 ± 7510.74 (11 200)	12636.92 ± 13442.46 (5620)	0.538 a
	Post TPE	9047.5 ± 4748.32 (7565)	11589.23 ± 10674.29 (9970)	0.737 a
Hemoglobin (g/dL)	Before TPE	10.52 ± 2.11 (10.6)	9.98 ± 1.73 (9.4)	0.422 b
	Post TPE	10.16 ± 1.42 (10.3)	9.98 ± 1.77 (10.1)	0.729 b
Hematocrit (%)	Before TPE	31.62 ± 6.17 (30.8)	30.2 ± 4.27 (29)	0.460 b
	Post TPE	30.34 ± 3.75 (30.7)	29.42 ± 5.73 (27.5)	0.604 b
PLT	Before TPE	208 250 ± 141247.78 (205 000)	118846.15 ± 119255.92 (60 000)	0.039* a
	Post TPE	238535.71 ± 125034.66 (255 000)	65461.54 ± 58663.75 (48 000)	0.001* a
INR	Before TPE	1.42 ± 0.52 (1.2)	2.41 ± 2.03 (1.5)	0.070 a
	Post TPE	1.17 ± 0.15 (1.2)	1.48 ± 0.38 (1.5)	0.005* a
Lactate	Before TPE	2.89 ± 3.04 (1.7)	5.77 ± 4.64 (3.8)	0.025* a
	Post TPE	1.46 ± 0.5 (1.4)	5.73 ± 4.45 (4.4)	0.001* a

Abbreviations: ALT, alanine aminotransferase; AST, aspartate aminotransferase; CRP, c reactive protein; INR, international normalized ratio; OFI, organ failure index; PELOD, pediatric logistic organ dysfunction score; PICU, pediatric intensive care unit; PRISM, pediatric mortality risk score; TPE, therapeutic plasma exchange; VIS, vasoactive‐inotropic score.

^a Mann–Whitney U Test.

^b Student t‐test.

TABLE 3.

Clinical characteristics of the patients.

		n	%
Underlying disease	No	20	48.8
	There is	21	51.2
Diseases (n = 21)	Rheumatologic disease	8	38.1
	Cardiac disease	5	23.8
	Neurological disease	5	23.8
	Other	3	14.3
Sepsis	No	25	61
	There is	16	39
Non‐invasive ventilation (NIV)/invasive mechanical ventilation (IMV)	No	26	63.4
	There is	15	36.6
Blood product required	No	4	9.8
	There is	37	90.2
Blood product (n = 37)	Erythrocyte suspension (ES)	13	35.1
	Intravenous immunoglobulin (IVIG)	5	13.5
	ES + platelet suspension (PS)	6	16.2
	ES + PS + IVIG	4	10.8
	ES + PS + fresh frozen plasma (FFP)	6	16.2
	IVIG + PS + ES + FFP	1	2.7
	IVIG + PS	1	2.7
	ES + FFP	1	2.7
Inotropic treatment requirement	No	21	51.2
	There is	20	48.8
Survival	Alive	28	68.3
	Dead	13	31.7
		Min–Max	Mean ± SD (median)
Number of NIV/IMV days		4–65	24.54 ± 18.70 (22)
Number of organ failure		0–6	1.93 ± 1.83 (1)
Number of plasmapheresis sessions		2–16	5.71 ± 3.21 (5)

The mean number of TPE sessions was 5.71 ± 3.21. IL‐6 levels significantly increased after TPE compared to before TPE (p: 0.006; p <0.05). TWEAK protein levels also significantly decreased after TPE compared to before TPE (p: 0.030; p <0.05). ALT, AST, urea, creatinine, ferritin, D‐dimer, procalcitonin, INR, and leukocyte levels, all showed a statistically significant decrease after TPE. Age, gender, IL‐6, and TWEAK protein levels before TPE did not exhibit any statistically significant differences (p >0.05) (Table 4).

TABLE 4.

Comparison of laboratory parameters before and after therapeutic plasma exchange (TPE).

	Before TPE		After TPE
	Min–max	Mean ± SD (median)	Min–max	Mean ± SD (median)	p
IL 6 (ng/L)	72–887.7	359.19 ± 176.74 (351.8)	124.9–862	418.77 ± 173.48 (429.8)	0.006 a
TWEAK (mg/L)	682.1‐4089.4	2946.69 ± 1197.7 (3793.7)	686.5–4687.1	2712.22 ± 1076.86 (2764.8)	0.030 a
ALT (U/L)	6–3662	371.41 ± 721.58 (64)	9–694	78.59 ± 129.49 (33)	0.001 b
AST (U/L)	12–5984	576,17 ± 1143,82 (82)	8–2628	126,1 ± 409,05 (35)	0.002 b
Urea (mg/dL)	0.4–192	40 ± 35.76 (26.9)	4.1–81	24.74 ± 16.38 (21)	0.011 b
Creatinine (mg/dL)	0.2–3.2	0.63 ± 0.61 (0.4)	0.2–1.4	0.4 ± 0.3 (0.3)	0.009 b
Calcium (mg/dL)	5.7–10	8.24 ± 1.01 (8.3)	7.4–10.6	9.06 ± 0.73 (8.9)	0.001 a
Ferritin (ng/mL)	38–100 000	7297.36 ± 18 928.27 (850)	68–15 960	2075.54 ± 3700.88 (527)	0.038 b
Dimer (ng/ml)	400–4400	3064.63 ± 1720.84 (4400)	300–4400	2303.22 ± 1711.43 (1590)	0.001 b
CRP (mg/L)	0.2–334	45.76 ± 70.07 (16.3)	0.3–230.4	29.13 ± 57.38 (3.8)	0.086 b
Procalcitonin (ng/mL)	0–100	23.34 ± 36.49 (2.4)	0–263	21.94 ± 57.93 (0.6)	0.040 b
Leukocyte count (WBC)	210–46 660	12542.68 ± 9606.86 (11 000)	70–31 000	9853.41 ± 7129.87 (7630)	0.02 b
Hemoglobin (g/dL)	6.2–14.5	10.35 ± 1.99 (10.3)	6.7–13.7	10.1 ± 1.52 (10.2)	0.502 a
HTC (Hematocrit) (%)	19.2–43	31.17 ± 5.62 (30.3)	19.4–39.7	30.05 ± 4.42 (30.6)	0.291 a
Platelet count (PLT)	5000–607 000	179902.44 ± 139669.4 (177 000)	3000–487 000	183658.54 ± 135031.96 (139 000)	0.811 b
Prothrombin time (INR)	1–7.8	1.73 ± 1.28 (1.3)	0.8–2.1	1.27 ± 0.28 (1.2)	0.005* b
aPTT (second)	21.7–173.4	39.87 ± 27.84 (31)	24.8–56	32.33 ± 8.17 (29.5)	0.255 b
Lactate	0.9–13.6	3.8 ± 3.81 (1.9)	0.6–16.6	2.81 ± 3.19 (1.7)	0.061 b

Abbreviations: ALT, Alanine aminotransferase; AST, aspartate aminotransferase; CRP, c reactive protein; INR, international normalized ratio.

^a Paired samples t‐test.

^b Wilcoxon sign test.

IL‐6 and TWEAK protein levels of the patient groups before and after TPE were compared concerning hospitalization diagnoses, the presence of sepsis, NIV/IMV support, and the need for blood products. No statistically significant differences were observed (p >0.05). A positive, moderate, and statistically significant correlation was found between the pre‐TPE IL‐6 level, the number of NIV‐IMV days, and serum ferritin level (p <0.05) (Table 5).

TABLE 5.

Correlations with interleukin‐6 (IL‐6) and TWEAK protein levels before therapeutic plasma exchange (TPE).

		Before TPE
		IL 6 (ng/L)	TWEAK (mg/L)
Number of organ failure	r	0.296	0.313
	p	0.060	0.046
Number of days on NIV‐IMV	r	0.504	−0.100
	p	0.009*	0.628
Number of days in NICU hospitalization	r	0.293	0.139
	p	0.063	0.387
PRISM	r	0.191	0.298
	p	0.232	0.058
PELOD	r	0.285	0.213
	p	0.071	0.181
OFI	r	0.287	0.328
	p	0.069	0.036
VIS	r	0.203	0.218
	p	0.203	0.171
ALT (U/L)	r	0.092	−0.024
	p	0.569	0.884
AST (U/L)	r	0.104	−0.066
	p	0.517	0.683
Urea (mg/dL)	r	0.006	0.106
	p	0.969	0.508
Creatinine (mg/dL)	r	0.003	0.009
	p	0.987	0.954
Calcium (mg/dL)	r	0.141	−0.069
	p	0.379	0.666
Ferritin (ng/mL)	r	0.370	0.042
	p	0.017*	0.793
Dimer (ng/mL)	r	0.064	0.348
	p	0.691	0.026
CRP (mg/L)	r	−0.045	0.045
	p	0.782	0.782
Procalcitonin (ng/mL)	r	−0.145	0.182
	p	0.364	0.254
WBC (leukocyte)	r	0.051	0.103
	p	0.752	0.521
Hemoglobin (g/dL)	r	−0.062	−0.076
	p	0.702	0.637
HTC (hematocrit) (%)	r	−0.083	−0.049
	p	0.605	0.76
PLT (platelet)	r	−0.088	−0.228
	p	0.586	0.151
INR (prothrombin time)	r	0.009	0.046
	p	0.957	0.777
aPTT (second)	r	0.146	0.148
	p	0.364	0.355
Lactate	r	−0.125	0.175
	p	0.436	0.275

Abbreviations: ALT, alanine aminotransferase; AST, aspartate aminotransferase; CRP, c reactive protein; OFI, organ failure index; PELOD, pediatric logistic organ dysfunction score; PRISM, pediatric mortality risk score; VIS, vasoactive‐inotropic score.

A positive, moderate, and statistically significant correlation was found between the TWEAK protein level before TPE and the number of organ failures, OFI, and D‐Dimer levels (p <0.05). No statistically significant correlation was found between the difference levels of IL‐6 and TWEAK after TPE and the number of PICU hospitalization days, the number of organ failures, prognosis, VIS, PELOD, and OFI scores before and after TPE (p >0.05). Regarding prognosis, there was no statistically significant difference in age, gender, the number of PICU stay days, WBC count, hemoglobin, hematocrit, serum IL‐6, TWEAK, urea, and creatinine levels (p >0.05). The number of organ failures, PRISM, PELOD, OFI, and VIS scores for deceased patients were statistically significantly higher than those for survivors (p <0.05). While there was no statistically significant difference in ALT, AST, CRP, procalcitonin, and INR levels before TPE based on prognosis (p >0.05), the levels of ALT, AST, CRP, procalcitonin, and INR after TPE were higher for deceased patients than for survivors (p: 0.034, p: 0.001, p: 0.001; p <0.05). Platelet levels for deceased patients before and after TPE were statistically significantly lower (p: 0.039; p: 0.001; p <0.05), and lactate levels were higher (p: 0.025; p: 0.001; p <0.05) than for survivors. Respiratory support with NIV/IMV, the need for inotropic drugs, and the sepsis rate were significantly higher in deceased patients compared to survivors (p: 0.001; p <0.05) (Table 6).

TABLE 6.

Mortality assessments.

		Survivors	Deceased patients
		n (%)	n (%)	p
Gender	Male	14 (50%)	8 (61.5%)	0.724a
	Female	14 (50%)	5 (38.5%)
NIV/IMV		13 (46.4%)	13 (100%)	0.001*b
Sepsis		6 (21.4%)	10 (76.9%)	0.002*a
Blood product needs		24 (85.7%)	13 (100%)	0.288b
Inotropic treatment requirement		8 (28.6%)	12 (92.3%)	0.001*a

4. DISCUSSION

Lyu et al. [6] suggested that the centrifugal method may better eliminate smaller protein molecules than membrane filtration, though their extravascular distribution and high production rate can limit clearance.

In 2001, Nakae et al. conducted a study assessing cytokine levels in patients undergoing therapeutic plasma exchange (TPE) and continuous hemofiltration (CRRT). In patients with liver failure, those receiving only TPE were designated as Group 1, while those undergoing CRRT were Group 2. The study found no significant changes in TNFα, IL‐8, or IL‐6 levels in Group 1 after TPE, but Group 2 showed significant decreases in TNFα and IL‐8, with no change in IL‐6 levels [7].

In our study, patients who underwent simultaneous CRRT and TPE were excluded. The serum IL‐6 level was significantly higher after TPE than before TPE (359 vs. 418 ng/L, p: 0.006), while the TWEAK level significantly decreased after TPE (2946 vs. 2717 ng/L, p: 0.030). Additionally, we observed a decrease in ferritin, D‐dimer, procalcitonin, ALT, AST, urea, creatinine, WBC, and INR levels, along with an increase in serum calcium levels post‐TPE. This increase in serum calcium is attributed to the regular calcium infusion administered during the procedure to mitigate the risk of hypocalcemia.

We hypothesize that the observed elevation in IL‐6 levels during the TPE procedure may be due to the nature of IL‐6, which predominantly exists in a bound form to plasma proteins. While small, freely circulating molecules can be more readily eliminated during plasmapheresis, the removal of protein‐bound IL‐6 is more challenging. Furthermore, given that IL‐6 is continuously produced by the immune system, its levels may rapidly rise again after the procedure. This resurgence, particularly in the context of ongoing inflammatory processes, may prevent IL‐6 levels from returning to baseline. Consequently, to achieve a more effective reduction in IL‐6 levels, plasmapheresis is often combined with additional treatment modalities, such as CRRT. The molecular weight cut‐off limits of the hemofilters used in CRRT range from 30 to 50 kilodaltons, allowing for the filtration of cytokines like IL‐6 (which has a molecular weight of 16 kilodaltons). This study suggests that certain inflammatory cytokine levels may only be adequately reduced through the combination of TPE and CRRT.

Faqihi et al. conducted a randomized trial with 87 ICU patients suffering from life‐threatening COVID‐19. Half received standard treatment, while the other half also received therapeutic plasma exchange (TPE). The TPE group had fewer mechanical ventilator days and lower mortality rates (p: 0.007 and p: 0.02, respectively). Additionally, TPE was linked to increased lymphocyte counts and ADAMTS13 activity, along with decreased levels of serum lactate, LDH, ferritin, D‐Dimer, and IL‐6 [8]. Hassaniazad et al. administered 3 cycles of TPE 24–36 h apart in 22 patients with severe COVID‐19 infection. They reported that the TPE application provided a significant improvement in both clinical and laboratory parameters. A significant decrease in IL‐1, IL‐6, IFN‐γ, and IL‐17 levels was observed following the procedure in the TPE group. They also reported a negative correlation between IL‐1, IL‐6, IFN‐γ, and IL‐17 levels and oxygen saturation before and after TPE, emphasizing that TPE can be added to standard treatment protocols in severe cases of COVID‐19 [9]. In our study, a moderate, positive correlation was found between the baseline IL‐6 level and the number of days on NIV/IMV, as well as the ferritin level. Additionally, we observed a positive correlation between the baseline TWEAK level and the number of organ failures, OFI score, and D‐dimer level. However, no correlation was found between the difference in IL‐6 and TWEAK levels after TPE and prognosis.

In a study published in 2020, it was reported that there was a significant decrease in IL‐6, IL‐10, TNFα, and CRP levels after TPE application in adult COVID‐19‐positive patients on mechanical ventilation. The study emphasized that TPE and continuous hemofiltration applications provided clinical improvement by reducing the levels of IL‐6 and other cytokines in cases of fulminant liver disease, autoimmune inflammation, neurological conditions, and infectious diseases [10]. In a similar study conducted in adult patients with COVID‐19‐induced Acute Respiratory Distress Syndrome (ARDS), it was reported that there was a significant decrease in acute‐phase reactants, inflammatory mediators, liver enzymes, and bilirubin levels after TPE application. Although a decrease in T helper cells was observed after TPE, it was reported that their levels returned to baseline values after 1 week [11]. In our study, we observed a statistically significant decrease in ALT, AST, procalcitonin, ferritin, and TWEAK levels after TPE. However, CRP levels also decreased, but this reduction was not statistically significant (45.7 vs. 29.1 mg/L, p: 0.086).

For patients experiencing a cytokine storm, such as those with primary and secondary Hemophagocytic Lymphohistiocytosis (HLH), TPE offers hematological support and proves beneficial in treatment. Songs et al. reported that plasmapheresis treatment reduced plasma IL‐6 concentration in patients with fatal HLH [12].

A 2020 review highlighted TWEAK [13] protein as a potential novel treatment for systemic and pulmonary arterial hypertension, obstructive sleep apnea, asthma, idiopathic pulmonary fibrosis, and chronic obstructive pulmonary disease. In vivo and in vitro studies suggest that the TWEAK‐Fn14 axis regulates neuroinflammation in various central nervous system inflammatory diseases [14, 15, 16]. In an experiment conducted on mice with anaphylaxis, it was demonstrated that the TWEAK/Fn14 axis accelerated anaphylactic reactions. Deletion of this axis or antibodies blocking TWEAK prevented Histamine/PAF (Platelet Activating Factor)‐related vascular subcutaneous permeability. It was emphasized that inhibiting the TWEAK/Fn14 axis may be effective in the treatment of anaphylaxis [17].

The role of the TWEAK/Fn14 axis has been investigated in various disease groups [18, 19, 20]. In a study that compared 67 patients with Crohn's disease (CD) and 33 healthy individuals, higher serum TWEAK levels were observed in individuals with CD, and it was reported that the TWEAK/Fn14 axis played an important role in CD‐related intestinal fibrosis [21]. In 2017, a study involving 40 patients with Hashimoto's thyroiditis found that serum TWEAK levels were lower than those with Graves' disease and healthy individuals. These levels were negatively correlated with antithyroid peroxidase antibody levels [22]. In our study, different disease groups were compared in terms of serum IL‐6 and TWEAK levels before and after TPE, but no significant difference was found.

Moloi et al. [23] investigated urinary levels of monocyte chemoattractant protein‐1 (MCP‐1) and TWEAK in patients with confirmed active lupus nephritis through renal biopsy, both at baseline and after completion of induction therapy. They observed that urinary MCP‐1 levels decreased from 1092.7 to 314.5 pg/mg (p: 0.06), and urinary TWEAK levels decreased from 159 to 36 pg/mg (p: 0.03) at the end of treatment.

The study revealed that both biomarker levels in urine decreased significantly in patients who achieved complete or partial remission in the early period, while this decrease was not observed in patients without remission. In another study, it was reported that the use of urinary TWEAK had 70% sensitivity and 63% specificity in predicting a complete response in the 3rd month of induction treatment.

In a 2022 study involving critically ill adult patients with COVID‐19, it was observed that the admission TWEAK level was lower in the group of surviving patients compared to the deceased patient group (179.3 vs. 247.2 pg/mL, p: 0.037). The study noted that TWEAK protein increased 1 week after admission but started to decrease again in the second week. However, in the deceased patients, while there was an increase in the first week, the expected decrease was not observed in the second week (247.2 pg/mL at admission, 284.2 pg/mL in the first week, and 293.8 pg/mL in the second week, p: 0.016). The study also reported a positive correlation between serum TWEAK level and CRP (C‐Reactive Protein) and hsTnI (High‐sensitive Troponin‐I) levels. Furthermore, it was emphasized that the admission TWEAK level served as a valuable marker for predicting hospital mortality [24].

Kerget et al. reported that IL‐6 levels showed similar fluctuations, and the baseline and day 5 IL‐6 levels of patients who developed MAS and ARDS were higher than those of patients who did not develop MAS and ARDS [25].

We excluded patients who received both CRRT (Continuous Renal Replacement Therapy) and TPE (Therapeutic Plasma Exchange) treatment because extracorporeal blood purification therapies, such as CRRT and TPE, can remove circulating cytokines and other inflammatory mediators. This could potentially confound the results by affecting the levels of inflammatory markers, thereby obscuring the individual effects of each treatment. Studies have demonstrated that CRRT can effectively eliminate proinflammatory cytokines like IL‐6, which may influence outcomes in cases of cytokine release syndrome (CRS) that are refractory to other treatments such as tocilizumab and corticosteroids [25].

In our study, we compared serum IL‐6 and TWEAK levels of survivors and deceased patients before and after TPE, and no significant difference was observed between the two groups. The deceased patients had a higher number of organ failures, PRISM, PELOD, OFI, and VIS scores compared to the survivors. Similarly, ALT, AST, PCT, and INR levels of the deceased patients after TPE were higher than those of the survivors, and platelet counts were lower before and after TPE. We also observed a positive correlation between the baseline TWEAK level and the number of organ failures, OFI score, and D‐dimer level. TWEAK levels were 2946 mg/dL before TPE and 2712 mg/dL after TPE, and this difference was statistically significant (p: 0.030). However, we did not detect a decrease in IL‐6 levels before and after TPE.

The data obtained at the end of the study revealed that higher serum TWEAK levels were associated with a higher number of organ failures, higher OFI score, and D‐dimer level, while higher serum IL‐6 levels were associated with longer NIV/IMV days and higher ferritin levels in patients who underwent TPE. We found a significant decrease in TWEAK protein levels after the TPE procedure, while IL‐6 levels did not decrease. We believe that serum IL‐6 and TWEAK protein can be used as markers to determine the severity of the disease and the level of inflammation, but they may not be sufficient to predict the patient's prognosis.

FUNDING INFORMATION

The financial support for our study was provided by the Scientific Research Projects Coordinatorship with the application approval number 2023/016.

CONFLICT OF INTEREST STATEMENT

The authors declare no conflicts of interest.

ETHICS STATEMENT

Ethical approval dated 21.04.2022 and numbered B.10.1.TKH.4.34.H.GP.0.01/154 was obtained from the Clinical Research Ethics Committee of Health Sciences University (SBU), Istanbul Ümraniye Training and Research Hospital.

PATIENT CONSENT STATEMENT

A consent form was obtained from the patients for voluntary participation in the study. Patient identity information was kept confidential. In our study, no unnecessary procedures and examinations that harmed the patients were performed. Patient confidentiality was protected under all circumstances. The study does not contradict the Declaration of Helsinki and Good Clinical Practice Guidelines.