Resolution of thrombocytopenia through sodium bicarbonate intervention as an alternative diluent for heparin in an extracorporeal membrane oxygenation patient: a case report

Abstract

Background

Thrombocytopenia is a common and multifaceted complication in patients receiving heparin therapy during extracorporeal membrane oxygenation (ECMO). Its development may result from several mechanisms, such as platelet consumption, hemodilution, shear stress, circuit activation, and heparin-induced thrombocytopenia. Managing thrombocytopenia in this context can be challenging. Therefore, we present a patient who developed thrombocytopenia while receiving a heparin infusion, which resolved after changing the infusion diluent.

Case presentation

We presented a case of a 64-year-old Pakistani male with multiple cardiac comorbidities who was admitted in a critical condition following cardiac arrest, requiring immediate resuscitation and veno-arterial ECMO initiation for circulatory support. During the ICU course, he developed thrombocytopenia while on heparin infusion diluted in D5W. Despite ruling out heparin-induced thrombocytopenia, his platelet counts continued to decline. As an alternative approach, the heparin diluent was switched to sodium bicarbonate, leading to recovery in platelet levels without the need for transfusion.

Conclusions

This case highlights a possible association between using sodium bicarbonate as a diluent for heparin infusion and platelet recovery. While this finding warrants further investigation, additional studies are necessary to determine its broader clinical implications.

Introduction

Thrombocytopenia is a frequent hematologic disorder among critically ill patients and commonly complicates the management of those receiving extracorporeal membrane oxygenation (ECMO). It represents the most prevalent hematologic issue associated with ECMO therapy. [1]. Platelet consumption, hemodilution, shear stress, circuit activation, and heparin-induced thrombocytopenia are recognized factors contributing to platelet count reduction in ECMO patients [1, 2]. Managing thrombocytopenia in ECMO patients is particularly challenging due to their heightened risk of bleeding, necessitating vigilant and specialized care [3].

Studies indicate that up to 40% of ECMO patients experience severe thrombocytopenia, with platelet counts frequently dropping below 50 × 10⁹/L [3]. The use of anticoagulants, such as heparin, further complicates this problem. Clinicians should pursue safer alternatives for thrombocytopenia prevention that do not compromise anticoagulation. Ongoing efforts should prioritize the development of innovative strategies to prevent thrombocytopenia while maintaining optimal anticoagulation. Studies have evaluated various methods to manage platelet loss by adjusting anticoagulant regimens without compromising their efficacy [4]. One promising approach involves using sodium bicarbonate as a diluent for heparin infusions, which may help prevent thrombocytopenia. Evidence suggests that sodium bicarbonate stabilizes anticoagulant proteins by preserving their active conformations, thereby reducing protein degradation and lowering the risk of thrombogenesis [5].

Sodium bicarbonate has traditionally been used to treat metabolic acidosis and maintain buffering during hemodialysis. Recent research indicates it may also confer benefits in ECMO and other cardiopulmonary support settings. Notably, the advantages of diluting heparin with sodium bicarbonate appear to stem primarily from its ability to stabilize heparin, rather than from direct therapeutic effects. By raising pH, sodium bicarbonate helps preserve the structure of coagulation proteins and prevents their premature breakdown [6]. Additionally, it has been used successfully in purging solutions for devices such as Impella, outperforming D5W by improving anticoagulant protein levels and reducing thrombotic events [7]. This case aims to highlight the potential advantages of using sodium bicarbonate as an alternative heparin diluent for managing thrombocytopenia in ECMO patients.

Case presentation report

A 64-year-old Pakistani male with a history of diabetes mellitus, hypertension, and ischemic heart disease. In 2023, he underwent percutaneous coronary intervention (PCI) with placement of two drug-eluting stents (DES) in the right coronary artery (RCA) to treat triple-vessel disease (TVD) and left main coronary artery involvement. His ejection fraction (EF) was 40%. There was no family history of hematologic or hereditary platelet disorders. The patient was presented to the emergency department (ED) with chest pain, shortness of breath, and diaphoresis. On arrival, the patient developed cardiac arrest with pulseless electrical activity (PEA) and was revived after 16 minutes of cardiopulmonary resuscitation (CPR). The electrocardiogram (ECG) showed ST elevation. He was immediately given aspirin 325 mg and clopidogrel. The extracorporeal membrane oxygenation (ECMO) team initiated veno-arterial ECMO (VA ECMO) for cardiac support, and he was urgently transferred to the catheterization lab for PCI to address restenosis, stent implantation in the RCA, and thrombectomy. His vital signs included a temperature of 36.9 °C, a heart rate of 86/min, a blood pressure of 96/68 mmHg, and a respiratory rate of 18/min.

Initial laboratory findings

Laboratory results indicated metabolic acidosis with elevated lactate, markers of cardiac injury, and renal dysfunction. Coagulation studies showed prolonged clotting times, while platelet count remained normal. Liver function tests were within normal limits, except for a slight elevation in aspartate aminotransferase (Table 1).

Table 1.

Admission blood work results

Parameter	Result	Normal reference range
Lactate	5.6 (mmol/L)	0.6–1.8 mmol/L
NT-ProBNP	1731.70 pmol/L	20.0–54.9 pmol/L
Troponin I	3284.70 ng/L	0.0–34.2 ng/L
Creatine kinase	339.00 U/L	30–200 U/L
Serum creatinine	171.6 µmol/L	63.6–110.5 µmol/L
Blood urea nitrogen (BUN)	11.90 mmol/L	3.0–9.2 mmol/L
Uric acid	525.70 µmol/L	210–420 µmol/L
Activated partial thromboplastin time (APTT)	> 180 seconds	26–40 seconds
International normalized ratio (INR)	2.48	0.8–1.2
Prothrombin time (PT)	31.10 seconds	12.3–14.7 seconds
Platelets	258 × 103/uL	150–400 × 103/uL
Hemoglobin	13.20 g/dL	13.5–18 g/dL
Aspartate aminotransferase (AST)	39 U/L	11–34 U/L

ICU course

The patient was transferred to the intensive care unit (ICU) on the same day. A new left bundle branch block (LBBB) was noted on ECG. He was intubated, placed on mechanical ventilation (FiO₂ 90%, PEEP + 5 cmH₂O, tidal volume (VT) 500 mL, respiratory rate (RR) 30 breaths/minutes), and started on norepinephrine (1 mcg/kg/minutes), adrenaline (0.2 mcg/kg/minutes), vasopressin (0.04 unit/minutes), and fentanyl sedation. On day 2, continuous renal replacement therapy (CRRT) in the form of continuous veno-venous hemodialysis (CVVHD) was initiated for 24 hours.

He was treated with insulin infusion, piperacillin/tazobactam (4.5 g intravenous every 6 hours on CVVHD; 3.375 g intravenous every 8 hours off CVVHD), aspirin (100 mg oral daily), ticagrelor (90 mg oral twice daily), omeprazole (40 mg intravenous daily), and heparin (Unfractionated Heparin, 1000 units/mL; Hospira, USA) was administered as a continuous infusion at 100 units/hour, diluted in 5% dextrose water (D5W, 50 mL total volume). He developed wide QRS tachycardia, followed by clinical deterioration with hypotension, bradycardia, and coexisting electrolyte disturbances.. Norepinephrine was increased, sodium bicarbonate, magnesium sulfate, and potassium chloride were administered. The cardiac rhythm subsequently stabilized after electrolyte correction.

On day 6, he developed thrombocytopenia (PLT: 72.9 × 10³/µL; normal: 150–400 × 10³/µL) and hemoglobin drop (7.76 g/dL; normal: 13.5–18 g/dL). No signs of bleeding were identified at the presentation. HIT testing (PF4-heparin ELISA, optical density < 0.4, negative) was performed on Day 6 at the platelet nadir. Following thrombocytopenia onset, the clinical team switched the heparin diluent to sodium bicarbonate (8.4%, 25 mEq in 50 mL). Infusion rate and the dose remained unchanged, leading to PLT count improvement without platelet transfusion. PLT counts from days 2 to 9 were: 77, 125, 153, 155, 158, and 175 × 10³/µL. Hemoglobin levels improved: 7.23, 7.75, 7.79, 7.18, 8.44, and 7.99 g/dL (Table 2).

Table 2.

Laboratory parameters pre- and post-intervention with D5W to sodium bicarbonate dilution

Days	Platelet count (10*3/μL)	Hgb (g/dL)	INR	pH	Pco2	Hco3	Ca +  +	SCr	Urea	Fibrinogen	LDH	Reticulocyte	D-dimer
The dilution as D5W
1	258	13.2	1.42	7.13	65	16.4	1.34	127	10.6	n/a	N/A	N/A	N/A
2	185.00	10.5	2.48	6.94	36	20.4	1.38	113	11.9	3.08	N/A	N/A	> 20
3	137.00	10.1	2.30	7.49	32	21.7	1.37	141	9.2	4.16	N/A	N/A	N/A
4	130.00	9.55	2.21	7.40	33	20.4	1.36	159	9.9	5.81	1567	N/A	N/A
5	93.30	8.65	1.26	7.44	32	22.2	1.31	161	10.5	5.19	N/A	N/A	N/A
6	72.90	7.67	1.44	7.42	44	27.9	1.26	200	12.1	4.44	1029	N/A	> 20
After changing to the sodium bicarbonate dilution
7	76.60	7.23	1.29	7.47	36	24.5	1.28	222.8	13.3	4.99	768	N/A	> 20
8	125.00	7.75	1.37	7.53	39	27.1	1.27	193.6	12.7	6.18	N/A	N/A	> 20
9	153.00	7.79	1.38	7.51	30	21.8	1.33	177.4	13.5	4.45	680	N/A	> 20
10	155.00	7.18	1.21	7.50	28	23.4	1.26	165.7	12.2	3.27	N/A	N/A	> 20
11	158.00	7.99	1.28	7.46	31	25.3	1.29	139	9.3	3.43	559	N/A	> 20
12	175.00	8.44	1.30	7.33	36	25.0	1.23	178.9	9.1	3.12	N/A	N/A	> 20

On day 13, the patient was successfully decannulated from VA ECMO without complications. However, his clinical status continued to decline. He developed ventilator-associated pneumonia (VAP), which progressed to severe septic shock, ultimately resulting in multi-organ failure. He required maximum vasopressor and inotropic support as well as full mechanical ventilation. Bedside echocardiography revealed a low ejection fraction and poor cardiac contractility, indicating a poor prognosis. Subsequently, he developed bradycardia followed by asystolic cardiac arrest and was pronounced deceased.

Discussion

Thrombocytopenia progressed more rapidly than typically observed in ECMO patients, suggesting the presence of an additional contributing factor beyond circuit-induced platelet consumption. Distinguishing between ECMO-associated thrombocytopenia and HIT remains complex, as both conditions can present with similar platelet kinetics, requiring careful interpretation of functional and immunologic HIT assays.

In addition to ECMO-related platelet activation, several other factors may have contributed to the observed thrombocytopenia. The physiological stress associated with prolonged cardiopulmonary resuscitation (CPR) can trigger platelet activation and consumption. Moreover, the potential presence of latent or subclinical sepsis, medications, systemic inflammation, hemodilution due to fluid resuscitation, and early disseminated intravascular coagulation (DIC) could also have played a role. Recognizing these overlapping mechanisms provides a more comprehensive understanding of platelet variability and strengthens the interpretation of the observed clinical improvement after changing the heparin diluent.

Observed platelet recovery post-switch suggests a potential role of D5W in exacerbating platelet dysfunction, possibly through osmolar or endothelial interactions that warrant further investigation. Sodium bicarbonate’s anticoagulant properties may involve modulation of fibrin polymerization or thrombin inhibition, mechanisms that could influence ECMO-associated coagulopathy. These effects include direct interference with thrombin generation, fibrin polymerization disruption, and platelet function modification, resulting in reduced aggregation and adhesion to foreign surfaces [1–3, 5]. The observed prolongation of clotting time in the presence of sodium bicarbonate suggests a coordinated inhibition of fibrin formation, potentially mediated through alterations in calcium ion availability or direct interactions with coagulation factors [8]. However, the precise mechanistic pathways through which sodium bicarbonate modulates heparin activity remain incompletely elucidated, warranting further investigation into the molecular underpinnings of their interaction.

According to previous evidence, sodium bicarbonate is an effective alternative purge solution in patients with contraindications to heparin and its application in mechanical circulatory support devices has been explored in specific clinical scenarios. A case report descriping two patients with cardiac shock who received an Impella device and initially underwent heparin purge solution treatment, which later led to the development of HIT. In response, the healthcare team switched to sodium bicarbonate as an alternative purge solution along with systemic intravenous argatroban for anticoagulation. In the first case, HIT was diagnosed on postoperative day 2, prompting the transition to sodium bicarbonate. This change resulted in recovery of platelet counts without the need for transfusions. The second patient, who required emergency PCI for acute coronary syndrome, also developed HIT after 16 days of heparin use, leading to gastrointestinal bleeding. Following the switch to sodium bicarbonate, platelet levels stabilized without complications. Adopting sodium bicarbonate simplified anticoagulation management and showed a reduced risk of bleeding without affecting Impella flow or pressure over a 13-day observation period [6].

Another case report descriped a 73-year-old female with cardiogenic shock using an Impella CP™ device. Initially, clinicians used a heparin-based purge solution but switched to a bicarbonate-based solution due to severe hemothorax. These interventions helped control bleeding, allowing for device removal after 2 days [9]. In addition, a study demonstrated that purge fluid with sodium bicarbonate significantly increased pH, protein stability, and reduced denaturation compared to heparin or dextrose solutions. Two bicarbonate concentrations (25 mEq/L and 50 mEq/L) raised pH by an average of 0.61 ± 0.03 and 0.81 ± 0.03 units, respectively (p < 0.0001). Additionally, thromboelastography (TEG) showed reduced blood thrombogenicity with sodium bicarbonate, either alone or alongside heparin [1, 8].

Several factors could contribute to thrombocytopenia in ICU patients, including heparin infusion, infections, and medication side effects. In this case, the platelet counts improved after switching the diluent from heparin to sodium bicarbonate, suggesting a potential relationship between the previous diluent and thrombocytopenia. One hypothesis for this improvement is that the heparin diluent might have contributed to platelet suppression. This case allows us to speculate that sodium bicarbonate decreases the severity of thrombocytopenia in patients on ECMO. Once sodium bicarbonate was used as a diluent instead of D5W, platelet count saw a dramatic increase, and the patient did not require a platelet transfusion, which further suggests that sodium bicarbonate might be an alternative strategy to alleviate the impact of thrombocytopenia in patients on ECMO therapy [10]. Although the mechanisms for this remain to be established, modulation of calcium-dependent processes remains prevalent in sodium bicarbonate's action in clotting cascades surrounding future investigation.

This case report highlights the potential of sodium-water dilution for balancing bleeding and thrombosis in ECMO patients, particularly those with thrombocytopenia. Sodium bicarbonate may help maintain heparin stability while reducing thrombocytopenia, which can occur due to ECMO-related platelet consumption, sepsis, or blood disorders. While improvements in platelet counts have been noted, further studies are necessary to validate this approach in larger patient cohorts, incorporate it into ECMO protocols, and determine optimal dosing.

Conclusion

This case report shows potential benefits of using sodium bicarbonate as an alternative diluent for heparin in managing ECMO-associated thrombocytopenia. The successful resolution of thrombocytopenia without requiring platelet transfusions highlights sodium bicarbonate's potential as an effective adjunctive therapy. By helping to stabilize platelet counts in complex clinical scenarios, sodium bicarbonate may offer a promising option for managing thrombocytopenia where traditional interventions are limited. Further research is necessary to validate these findings and determine their wider applicability.

Author contributions

All authors critically revised the manuscript, agreed to be fully accountable for ensuring the integrity and accuracy of the work, and read and approved the final manuscript.

Funding

None.

Data availability

The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

Declarations

Ethics approval and consent to participate

The study was approved on February 24, 2025 by King Saud Medical City Center Institutional Review Board, Riyadh, Saudi Arabia (Ref. H2RI-24-Feb25-01). Participant confidentiality was strictly observed throughout the study by assigning anonymous, unique identifiers and limiting data access to the investigators only. Informed consent was waived due to the research's method as per the policy of the governmental and local research center.

Consent for publication

Written informed consent was obtained from the patient for publication of this case report and any accompanying images. A copy of the written consent is available for review by the Editor-in-Chief of this journal.

Competing interests

No author has competing interests in this study.