Therapeutic Plasma Exchange as a Bridge to Spontaneous Recovery in Pediatric Acute Liver Failure in a Resource‐Limited Setting: A Prospective Single‐Center Experience in Morocco

ABSTRACT

In resource‐limited settings where liver transplantation is unavailable, pediatric acute liver failure (PALF) carries a markedly high mortality rate. In this context, extracorporeal liver support systems (ELSS), such as therapeutic plasma exchange (TPE), do not serve as a bridge to transplantation, but rather the sole critical bridge to spontaneous liver recovery. We conducted a prospective single‐center study in a pediatric intensive care unit, including seven patients admitted with PALF who underwent TPE. Clinical data, biological markers (including transaminase, bilirubin, INR, and ammonia), and patient outcomes were analyzed to evaluate the biological efficacy of TPE and identify factors influencing survival. All seven patients presented with severe biochemical derangement and advanced hepatic encephalopathy (HE) upon admission. TPE demonstrated significant biological efficacy, consistently reducing hepatic cytolysis and ammonia levels. To further optimize continuous ammonia clearance, one patient underwent continuous venovenous hemodialysis (CVVHD) in conjunction with TPE. Despite these biochemical improvements, clinical outcomes remained poor due to significantly delayed presentations. All patients were admitted at least 2 weeks after symptom onset, largely because jaundice was underestimated and not perceived as a life‐threatening symptom. TPE is a biologically effective intervention and a vital bridge to spontaneous liver recovery for PALF in settings without liver transplant capabilities. However, its clinical success is severely limited by delayed hospital admission. The pre‐existing and irreversible neurological damage caused by prolonged hyperammonemia ultimately drives mortality. Improving survival rates requires a systemic shift toward early recognition of jaundice and immediate intensive care referral. presentation with advanced neurologic injury. The effects of earlier ALF recognition and referral on outcomes require further investigation.

1. Introduction

Pediatric acute liver failure (PALF) is a rare, rapidly progressive, and life‐threatening clinical syndrome characterized by severe hepatocyte injury, encephalopathy, and liver‐derived coagulopathy. Globally, the management of PALF often relies on emergency liver transplantation for patients who fail to demonstrate signs of spontaneous recovery. However, in resource‐limited settings where liver transplantation programs are unavailable, mortality rates remain exceptionally high. In such contexts, management strategies often focus on supporting the liver's potential for spontaneous regeneration while awaiting recovery. Extracorporeal liver support systems (ELSS), particularly therapeutic plasma exchange (TPE), have emerged as crucial interventions. By clearing toxic metabolites, inflammatory mediators, and ammonia while simultaneously replenishing essential coagulation factors, TPE serves to buy critical time for the liver to heal. This study describes our single‐center experience with TPE in seven children with PALF managed in our pediatric intensive care unit. We aim to report the observed changes in biochemical parameters and the clinical outcomes in this resource‐limited setting, highlighting the challenges associated with delayed hospital admission.

2. Methods

2.1. Study Design and Population

This prospective single‐center study was conducted in the pediatric intensive care unit over a two‐year period, from March 2024 to March 2026. This study was conducted in accordance with the declaration of Helsinki and approved by the ethics committee of the faculty of medicine, pharmacy and dental medicine of Fez, Morocco. All pediatric patients admitted with PALF who met the criteria for TPE during the study period were included, with no exclusions.

2.2. Definition of Acute Liver Failure

ALF was defined according to the Pediatric Acute Liver Failure Study Group (PALFSG) criteria:

1. No known evidence of chronic liver disease.

2. Biochemical evidence of acute liver injury (elevated transaminases) with onset within < 8 weeks.

3. Hepatic based coagulopathy defined as a prothrombin (PT) more than 15 s on International normalized ratio (INR) more than 1.5 not corrected by vitamin K in the presence of clinical hepatic encephalopathy (HE), or PT more than 20 s or INR more than 2.0 regardless of the presence of HE

While PALFSG criteria are specifically established for ALF, one patient in our series presented with acute‐on‐chronic liver failure (ACLF). This patient was included based on clinical judgment, as they demonstrated severe hepatic encephalopathy and coagulopathy comparable to the PALF patients, necessitating urgent extracorporeal liver support in the absence of transplant availability.

2.3. Data Collection

We collected demographic data and the primary etiology of liver failure for all patients. Clinical data at admission included neurological status (GCS, pupillary reactivity), respiratory status, and hemodynamic parameters. We also recorded the diagnostic investigations performed to determine the etiology of ALF, as well as details of therapeutic management. Including the treatment of organ dysfunction. In addition, we collected detailed data regarding the TPE procedure, continuous renal replacement therapy (CRRT) parameters when applicable, and extracorporeal therapy‐related complications. Due to the small sample size, no formal statistical significance testing (p‐value) was performed.

2.4. Standard Intensive Care Management

Standard ICU management included individualized hemodynamic resuscitation based on echocardiographic assessment to determine the type of shock and guide our treatment (isotonic saline boluses for hypovolemic shock or continuous norepinephrine for vasoplegic shock) and neurocritical care for patients with hepatic encephalopathy. The latter involved deep sedation, strict prevention of secondary brain injury, regular clinical and transcranial Doppler monitoring, and osmotherapy as needed for intracranial hypertension. Invasive lung‐protective mechanical ventilation was indicated in cases of severe encephalopathy, respiratory failure, or hemodynamic instability.

2.5. Laboratory and Imaging Investigation

Routine laboratory investigations included complete blood count, serum electrolytes, renal and liver function tests, C reactive protein (CRP), procalcitonin, and plasma ammonia levels. The etiological workup included screening for acute viral infections, including hepatitis A, B, C, Epstein Barr virus, and cytomegalovirus. All patients underwent abdominal ultrasonography with Doppler evaluation to assess the liver, spleen, and kidneys.

To evaluate the gravity of liver failure upon intensive care unit admission, scores like Pediatric End‐stage liver disease (PELD) and Liver injury unit (LIU) were calculated at admission. These scoring systems were utilized strictly as indices of initial clinical severity at presentation, rather than tools to quantify post‐procedural clinical improvement, given that plasma exchange inherently alters the biochemical components used in their calculation.

2.6. Therapeutic Plasma Exchange

Therapeutic plasma exchange was initiated based on a combination of biochemical and clinical criteria. Specifically, the procedure was indicated when patients presented with an ammonia level greater than 100 μmol/L associated with neurological deterioration, defined as an altered state of consciousness secondary to hepatic encephalopathy. Vascular access was obtained using a high‐flow double‐lumen catheter inserted into either the right internal jugular vein or the femoral vein under ultrasound guidance by a senior operator, given the increased hemorrhagic risk related to coagulation disorders. A 9 Fr or 12 Fr catheter was selected according to the patient's weight and body size. The TPE procedures were performed using the multiFiltrate acute dialysis machine (Fresenius Medical Care, Bad Homburg, Germany). The exchanged plasma volume was 1.5 plasma volume of the patient; it was calculated using the following formula: 1.5 × (70–80 mL × body weight [kg] × [1—hematocrit]). The exchange flow rate was adjusted according to the blood flow rate and the patient's hemodynamic status in order to maintain the filtration fraction at ~15%. Fresh frozen plasma was used as the primary replacement fluid to replenish essential coagulation factors and mitigate the risk of major bleeding complications. Given the presence of coagulation disorders, anticoagulation was withheld during the procedure. The number and frequency of TPE procedures were tailored to daily clinical and biochemical evolution. TPE was performed daily as long as the ammonia level remained above 100 μmol/L. Once the ammonia level dropped below this threshold, the procedure was withheld for 24 h to monitor the rebound hyperammonemia and evaluate native clearance efficacy.

2.7. Continuous Renal Replacement Therapy

In the cases who underwent CRRT for the management of severe hyperammonemia, the continuous venovenous hemodialysis (CVVHD) modality was utilized. This procedure was performed using the same multiFiltrate device (Fresenius Medical Care, Bad Homburg, Germany) as the TPE. The prescribed dialysis dose ranged from 60 to 100 mL/kg/h, the blood flow rate was set at 5 mL/kg/min and was strictly maintained at more than twice the dialysate flow rate. Net fluid removal was individualized and continuously adjusted based on the patient's degree of fluid overload. Given the presence of coagulation disorders, anticoagulation was withheld during the procedure. For the dialysis circuit, we used MultiBic dialysate solutions containing 2 mmol/L of potassium.

2.8. Laboratory Investigations

Liver and kidney function tests, coagulation profiles, CRP, procalcitonin, and ammonia levels were measured before and 6 h following the completion of each procedure of TPE. They were processed at the central laboratory of Hassan II university Hospital. Blood analysis was performed using point‐of‐case testing. For the purpose of this study, laboratory parameters were classified as pathological or critical based on standardized diagnostic criteria and institutional references, specifically defined as: a serum ammonia level > 100 μmol/L, an international Normalized Ratio (INR) > 1.5, serum transaminases exceeding 150 UI/L, and a total bilirubin level exceeding 2 mg/dL.

2.9. Safety and Complications Monitoring

Procedural safety was systematically monitored throughout the study period. In this series, TPE demonstrated a favorable safety profile: no procedural interruptions occurred, and no bleeding episodes, thrombotic events, or transfusion‐related adverse reactions were observed during or immediately after the procedures.

3. Results

Seven patients with acute liver failure or acute‐on‐chronic liver failure who received therapeutic plasma exchange were included in this case series. The median age was 7 years, ranging from 4 years to 11 years. Five patients were female and two were male. Six patients had ALF and one patient had ACLF. The most common etiology was hepatitis A infection (in five patients), while two cases were of indeterminate origin. Upon admission. All patients exhibited advanced hepatic encephalopathy (grade III) and severe biochemical derangement. Mechanical ventilation was required in all patients. All patients received vasoactive support with norepinephrine at varying doses. One patient underwent renal replacement therapy (CVVHD) (Table 1).

TABLE 1.

Baseline demographics, clinical characteristics, and clinical outcomes of the case series.

	01	02	03	04	05	06	07
Age/gender	11/F	4/M	6/M	7/F	8/F	6/F	5/F
Diagnosis type	ACLF	PALF	PALF	PALF	PALF	PALF	PALF
Etiology	Indeterminate	Hepatitis A	Indeterminate	Hepatitis A	Hepatitis A	Hepatitis A	Hepatitis A
Mechanical ventilation	Yes	Yes	Yes	Yes	Yes	Yes	Yes
CRRT (CVVHD)	No	No	No	No	No	No	Yes
Vasoactive support	Yes	Yes	Yes	Yes	Yes	Yes	Yes
PELD score	23	33	29	28	34	31	43
LIU score	244	286	261	255	318	264	365
Number of sessions of TPE	1	1	3	1	1	7	6
Replacement fluids type	2/3 FFP, 1/3 Albumin 4%	FFP	2/3 FFP, 1/3 Albumin 4%	2/3 FFP, 1/3 Albumin 4%	2/3 FFP, 1/3 Albumin 4%	2/3 FFP, 1/3 Albumin 4%	2/3 FFP, 1/3 Albumin 4%
Final outcome/primary cause of death	Death (Hemodynamic instability)	Death (Cerebral edema)	Death (Cerebral edema)	Death (Cerebral edema)	Death (Cerebral edema)	Survived	Death (Cerebral edema)

The mean baseline total bilirubin level was markedly elevated in all patients (16.8–25 mg/dL). The mean INR at admission ranged from 2.5 to 6. Baseline PELD scores ranged from 28 to 43, and LIU scores ranged from 255 to 365.

A total of 20 TPE sessions were performed. Four patients underwent a single session because they died before the second session could be initiated. One patient received three sessions (case 3), one patient underwent seven sessions (case 6), and the last patient underwent six sessions (case 7). The exchange volumes ranged from 1000 mL to 2500 mL per session and consisted of fresh frozen plasma alone or in combination with albumin 4%. Hemodynamic instability during TPE occurred in two patients and was managed with intravenous fluid resuscitation and escalation of norepinephrine support.

Descriptive analysis of biochemical parameters before and after the initial TPE session showed a general reduction in serum ammonia, transaminases and INR across the case series (Table 2). The mean serum transaminases level declined from 2411 (±1361) UI/L at baseline to 1048 (±643) UI/L post TPE. Ammonia levels, a critical marker for hepatic encephalopathy, decreased from a mean of 177.3 (±40.5) μmol/L to 135 (±35.6) μmol/L. Similarly, mean total bilirubin level decreased from 21.59 (±4.47) mg/dL to 17.33 (±6.64) mg/dL, while the mean INR demonstrated a downward shift from 3.63 (±1.13) to 2.98 (±1.12).

TABLE 2.

Evolution of biochemical parameters before and after Therapeutic Plasma Exchange sessions.

Biological parameter	Baseline mean (pre TPE)	Post TPE (after 1st TPE)	Mean variation (%)	Post 3rd	Mean variation (%)
Total bilirubin (mg/dL)	21.59 (±4.47)	17.33 (±6.64)	−19.7%	7.8 (±5.8)	−65.2%
Transaminases (UI/L)	2411 (±1361)	1048 (±643)	−56.5%	436 (±167)	−83.3%
Ammonia (μmol/L)	177.3 (±40.5)	135 (±35.6)	−23.8%	133.6 (±65.1)	−8.5%
INR	3.63 (±1.13)	2.98 (±1.12)	−17.9%	1.5 (±0.3)	−62.5%

Note: Post‐initial TPE values were calculated across six patients. Subgroup calculations following the third TPE session were strictly restricted to the three patients who received three or more consecutive procedures (Cases 03, 06, and 07).

In the subgroup of three patients who received repeated cycles of TPE (at least 3 sessions), a continuous descriptive decrease in ammonia and transaminase levels was observed over successive sessions, with mean transaminases level dropping from 2620 UI/L to 436 UI/L. The correction of coagulopathy was also observed, with the mean INR decreasing from 4 to 1.5. While the initial session had limited impact on cholestasis, the cumulative effect of three sessions reduced total bilirubin by more than half, from 22.4 mg/dL to 7.8 mg/dL. Finally, ammonia levels showed relative stabilization, averaging 133 μmol/L compared with 146 μmol/L at baseline.

In our case series, six out of seven patients died during hospitalization, while one patient survived with a favorable neurological outcome. Among the non‐survivors, five deaths were attributable to cerebral edema progressing to brain death, and one patient died as a result of refractory hemodynamic instability.

4. Discussion

This case series describes the clinical course and biochemical evolution of seven pediatric patients with acute liver failure managed with therapeutic plasma exchange in a specialized intensive care unit. Crucially, our institution operates without access to an emergency pediatric liver transplantation program. In such resource‐constrained settings, where definitive surgical treatment cannot be offered to patients meeting transplantation criteria, TPE ceases to be merely bridging therapy and instead serves as the primary metabolic support system.

Mortality in acute liver failure is mainly related to cerebral edema with intracranial hypertension and septic shock leading to multiorgan failure. In our case series, 6 patients died: one due to hemodynamic instability, and 5 due to cerebral edema with critical baseline ammonia levels ranging from 170 to 241 μmol/L, indicating a high risk of cerebral herniation. A major confounding factor in our series was the delayed timing of admission due to a local underestimation of early symptoms like jaundice. All patients in this study were admitted late, at least 2 weeks after the onset of symptoms; they often arrive with advanced HE and irreversible neurological damage before extracorporeal support could be initiated. While early neurocritical care and extracorporeal liver support systems (ELSS) are established in the literature as vital strategies to mitigate hyperammonemia and serve as a bridge to spontaneous recovery or transplantation, their efficacy is heavily time‐dependent. Various modalities evaluated for the pediatric population include therapeutic plasma exchange (TPE), continuous renal replacement therapy (CRRT), and molecular adsorption and recirculation system (MARS). The key question is determining the optimal timing for initiating extracorporeal support and selecting the most appropriate modality or combining two or three modalities.

TPE has emerged as a promising extracorporeal supportive therapy in ALF, including in pediatric populations. The procedure involves the removal of the patient's plasma, which contains circulating toxins, inflammatory mediators, and metabolites, followed by its replacement with fresh frozen plasma (FFP) or other replacement fluids. Through this mechanism, TPE may contribute to the clearance of potentially harmful substances implicated in the pathophysiology of ALF while simultaneously replenishing deficient plasma components such as coagulation factors and other essential proteins. They act as a bridge to spontaneous liver regeneration.

While the role of TPE in adults is relatively well established, with both the European association for the study of the liver (EASL), and the American society for apheresis (ASFA) recommending TPE as a first‐line support therapy [1, 2], evidence in the pediatric population remains limited and heterogeneous. Most of the available pediatric data derives from case series or observational cohort studies, with only a small number of comparative trials evaluating its efficacy.

To our knowledge, only two studies have compared TPE with standard medical therapy (SMT). The first study evaluated the impact of TPE in children with PALF, including 65 patients treated with TPE and 65 receiving SMT. By day 4, the TPE group showed significant improvements in biochemical and hemodynamic parameters, and native liver survival at 28 days was significantly higher in the TPE group compared with the control group (46.2% vs. 26.2%) [3]. The second study compared high‐volume plasma exchange (HVPE) vs. SMT in patients with Wilson disease presenting with ALF, and found that HVPE appears to be safe and effective treatment that improves 90‐day transplant‐free survival [4].

In addition to these comparative studies, several studies have also evaluated the effect of TPE on biochemical parameters before and after treatment. They showed significant improvement in INR, total bilirubin, and ammonia levels after TPE sessions [3, 4, 5, 6, 7, 8, 9, 10, 11]. Our findings closely mirror these literature reports. However, our experience strongly suggests that while repeated TPE cycles assist in maintaining a descriptive reduction of blood biomarkers, the final clinical outcome remains strictly dictated by the severity and management of cerebral edema, rather than the correction of conventional liver function markers alone.

In adult ALF, high‐volume plasma exchange has been shown to have a positive impact on transplant‐free survival [12], and is recommended as a first line therapy (category 1, grade 1A) [13]. Evidence of its usefulness in PALF remains limited. A recent systemic review and meta‐analysis reported higher transplant‐free survival with standard‐volume TPE compared with high‐volume TPE, while overall survival was comparable between the two approaches [14].

Coagulopathy is pathognomonic in ALF. TPE treats coagulopathy effectively while avoiding exogenous protein load that might exacerbate HE and has the added benefit of being fluid neutral. FFP is frequently used as the replacement fluid to replenish clotting factors and help correct coagulopathy. However, FFP contains citrate as an anticoagulant, which can lead to citrate accumulation and hypocalcemia when large volumes of plasma are administered during TPE, particularly in high‐volume exchanges. Careful monitoring of ionized calcium levels and prophylactic or therapeutic calcium supplementation is often required during the procedure [15, 16].

Our biological data supports the efficacy of TPE; however, this therapy may lead to improvements in biochemical parameters that do not consistently translate into improved clinical outcomes. Mortality in our series was predominantly driven by cerebral edema. The delayed initiation of TPE meant that while systemic biochemical clearance was achieved, it was not enough to reverse the pre‐existing, catastrophic neurological damage.

In addition to TPE, other extracorporeal liver support strategies may be used in ALF, particularly to control metabolic complications such as hyperammonemia [15, 17]. Continuous renal replacement therapy (CRRT) is therefore frequently employed especially in the presence of severe hyperammonemia and advanced hepatic encephalopathy [11, 15, 18, 19]. Early initiation of CRRT is strongly recommended to reduce ammonia levels, with a target of maintaining concentrations below 100 μmol/L. CRRT is generally initiated when ammonia levels exceed 150 μmol/L [18], although some studies have proposed lower thresholds, such as 75 μmol/L for early initiation [16, 20]. There is no evidence currently available that demonstrates the superiority of one CRRT modality over the other (CVVHD versus CVVH). A retrospective analysis from King's College Hospital involving 165 critically ill children with ALF showed that failure to reduce ammonia levels within 48 h of CRRT initiation was an independent predictor of mortality. Conversely, each 10% reduction in ammonia at 48 h increased the probability of survival by ~50% [21].

The dose of CRRT has a linear correlation with ammonia clearance: the higher the dose of CRRT, the better the clearance of ammonia [22]. The recommended dose is between 60 and 120 mL/kg/h. It should be adjusted to the ammonia clearance, clinical, and biochemical response.

The association between TPE and high‐dose CRRT allows for optimized and continuous ammonia clearance. This combined approach serves a dual purpose: TPE addresses synthetic failure and systemic inflammation, while CRRT stabilizes the internal environment and helps prevent cerebral edema. Recent evidence suggests that a proactive strategy using early high‐volume CRRT may provide better outcomes by serving as an effective bridge to liver spontaneous recovery [15].

Several limitations must be acknowledged in interpreting our findings. First this study is limited by its small sample size (n = 7) and its design as a descriptive, single‐center case series without a control group, which precludes formal statistical comparison and drawing definitive conclusions regarding overall TPE efficacy. Second, the lack of an available pediatric liver transplantation program at our institution represents a major confounding factor that heavily impacted the high mortality rate observed, as TPE alone cannot substitute for definite organ replacement in cases of irreversible parenchymal loss.

5. Conclusion

In this single‐center case series, TPE was feasible and associated with a transient numerical reduction in biochemical markers of liver injury and coagulopathy in pediatric acute liver failure. Nonetheless, mortality was high, reflecting the confounding impact of delayed clinical presentation with advanced, irreversible neurological injury upon admission. These findings underscore that while TPE provides robust temporary metabolic support, its clinical efficacy is strictly limited when initiated late in the course of the disease. Future large‐scale, controlled studies are warranted to evaluate whether earlier recognition, expedited referral, and timely initiation of extracorporeal support can translate into a true survival benefit.

Funding

The authors have nothing to report.

Ethics Statement

This study was conducted in accordance with the declaration of Helsinki and approved by the ethics committee of the faculty of medicine, pharmacy and dental medicine of Fez, Morocco.

Consent

Because the study involved pediatric patients, written informed consent was obtained from the parents during the period of the study.

Conflicts of Interest

The authors declare no conflicts of interest.

Data Availability Statement

The datasets generated and analyzed during the current study are not publicly available to protect patient privacy and confidentiality, but are available from the corresponding author upon reasonable request.