Tirofiban-induced thrombocytopenia

Abstract

Tirofiban is a small non-peptide ligand-mimetic Glycoprotein (GP) IIb/IIIa inhibitor which can reversibly bind to the arginine-glycine-aspartic acid (RGD) recognition site of GP IIb/IIIa to prevent platelet aggregation. It reduces the incidence of thrombotic cardiovascular events in patients with non-ST-segment elevation acute coronary syndrome (NSTE-ACS). Although generally considered safe, tirofiban has been reported to be associated with thrombocytopenia in several case reports and clinical trials. The pathogenesis for this adverse reaction is not entirely understood, is thought to be due to immune-mediated reaction. This side effect caused by tirofiban is especially concerning given how frequently it is used in the practice of contemporary cardiovascular care. The present review provides an overview of the pathophysiology, clinical presentation, management, and risk factors associated with tirofiban-induced thrombocytopenia.

KEY MESSAGES

• Tirofiban-induced thrombocytopenia usually occurred within the first 24 h of treatment, frequently accompanied by bleeding symptoms. The majority of the time, supportive care is used to manage this adverse event, and the platelet count often returns to normal in a few days.

• Although the exact cause of this adverse response is unknown, it is thought to be due to drug-dependent antibodies that bind to GP IIb/IIIa, presumably after tirofiban-induced conformational change.

• Age ≥ 65 years, white blood cell ≥ 12 × 10⁹/L, diabetes mellitus, congestive heart failure, and chronic kidney disease were identified as the risk factors for tirofiban-induced thrombocytopenia. Further investigations are needed for this.

Keywords:

• Tirofiban
• thrombocytopenia
• GP IIb/IIIa inhibitors
• clinical presentation
• pathogenesis
• risk factors

1. Introduction

Glycoprotein (GP) IIb/IIIa is the dominant integrins on platelets, which can bind to the fibrinogen to crosslink platelets, and is essential for platelet activation and aggregation [1]. By occupying the same pocket to block the binding of GP IIb/IIIa to fibrinogen, GP IIb/IIIa antagonists inhibit platelet aggregation (Figure 1(A)). Tirofiban, a highly selective small non-peptide platelet GP IIb/IIIa antagonist, was approved by the FDA in 1998 to reduce thrombotic cardiovascular events in patients with non-ST-elevation acute coronary syndrome (NSTE-ACS) (Table 1) [2]. It is used globally in patients with acute coronary syndrome or undergoing percutaneous coronary intervention (PCI) [3]. According to previous clinical trials, tirofiban also appeared to have an acceptable tolerance, safety, and efficacy profile for the treatment of ischemic stroke, particularly for patients complicated with atherosclerotic stenosis [4]. The SaTIS (Safety of Tirofiban in acute Ischemic Stroke) trial detected that tirofiban might be safe in acute ischemic stroke without increasing the incidence of cerebral hemorrhage and parenchymal hemorrhage [5]. The safety of tirofiban for acute ischemic stroke patients with large artery atherosclerosis receiving endovascular therapy was verified in a multi-center prospective study, and it also showed a trend of reducing mortality [6].

Figure 1. (A) The mechanism of tirofiban. (B) The Pathogenesis of tirofiban-induced thrombocytopenia. Created with BioRender.com.

Table 1. Characteristics of tirofiban.

Drug name	Tirofiban
Chemical structure
Chemical name	N-(butylsulfonyl)-O-[4- (4-piperidinyl)butyl]-L-tyrosine
Substance type	Non-peptide
Inhibition modality	Selective competitive
Binding modality	Reversible
Route of administration	Intravenous
Onset	>90% inhibition of platelet aggregation seen within 10 min
Distribution	Vdss: 22–42 L
Protein Binding	65%
Plasma Half-life	2h
Platelet function recovery time	4–8 h after discontinuation
Excretion	Urine (65%) and feces (25%) primarily as unchanged tirofiban

Tirofiban was generally well tolerated, with bleeding complications being the most commonly reported events, but there was no significant difference in the rate of major bleeding in tirofiban recipients compared with heparin in clinical trials [7]. Although being regarded as widely safe, tirofiban has been reported to be associated with thrombocytopenia in rates ranging from 0.4% to 5.6% [8]. This condition has been linked to a poor prognosis and higher mortality [9]. The majority of patients who experienced thrombocytopenia following therapy with tirofiban recovered uneventfully, but severe bleeding and fatalities had been reported, which were associated with longer in-hospital times, increased healthcare costs, morbidity, and even mortality [10]. The purpose of this article is to provide an overview of tirofiban-induced thrombocytopenia including its clinical presentations, pathogenetic mechanisms, and available treatments.

2. Thrombocytopenia caused by tirofiban

2.1. Clinical presentation

In most studies, mild thrombocytopenia was defined as the platelet count between 50 and 100 × 10⁹/L, severe thrombocytopenia as the platelet count less than 50 × 10⁹/L, and profound thrombocytopenia as the platelet count less than 20 × 10⁹/L [11,12]. In the clinical trials of tirofiban, the incidence of severe thrombocytopenia ranged from 0.1% to 0.5%, approximately twice the incidence among individuals who were not administered tirofiban [13,14]. Tirofiban-induced thrombocytopenia usually developed within 24 h after treatment [11], but in some cases might develop delayed thrombocytopenia up to 10 days later [15,16]. Strictly monitoring platelet count in patients receiving tirofiban was significant. Monitoring of platelet count at 6, 12, and 24 h after administrating tirofiban would detect most cases of acute thrombocytopenia [17]. Bleeding symptoms were frequently present along the course of thrombocytopenia, and occasionally life-threatening consequences such as alveolar and gastrointestinal system hemorrhages could occur [18,19]. Angina, palpitations, dyspnea, fever, hypotension, and chills may be observed in some situations [20,21]. The reported cases and studies of tirofiban-induced thrombocytopenia were shown in Tables 2 and 3 (table format and score/exclusion criteria refer to https://www.ouhsc.edu/platelets/ditp.html).

Table 2. Reported cases of thrombocytopenia associated with tirofiban.

NO.	CITATION	YEAR	AGE	SEX	DEATH	HOS	BL	SC	EX
1	Blood. 2002;100(6):2071–2076 [18]	2002	65	M	NO	YES	0	3	NA
2	Blood. 2002;100(6):2071–2076 [18]	2002	78	M	NO	YES	0	3	NA
3	Blood. 2002;100(6):2071–2076 [18]	2002	78	F	NO	YES	0	3	NA
4	Blood. 2002;100(6):2071–2076 [18]	2002	64	M	YES	YES	3	5	1
5	Interact Cardiovasc Thorac Surg. 2002;1(2):63–65 [22]	2002	52	NR	NO	YES	0	3	NA
6	J Thromb Haemost. 2003;1(10):2248–2250 [23]	2003	70	M	NO	YES	0	1	NA
7	J Thromb Haemost. 2003;1(10):2248–2250 [23]	2003	57	M	NO	YES	2	1	NA
8	Cardiovasc Drugs Ther. 2004;18(6):503–505 [24]	2004	63	M	NO	YES	0	1	NA
9	Am J Hematol. 2004;77(1):67–71 [25]	2004	47	F	NO	YES	NR	2	NA
10	Am J Hematol. 2004;77(1):67–71 [25]	2004	67	M	NO	YES	NR	2	NA
11	Am J Hematol. 2004;77(1):67–71 [25]	2004	72	F	NO	YES	NR	2	NA
12	Angiology. 2005;56(3):351–355 [26]	2005	80	M	NO	YES	NR	3	NA
13	J Thromb Haemost. 2005;3(5):1109–1110 [16]	2005	67	M	NO	YES	NR	3	NA
14	J Thromb Haemost. 2005;3(5):1109–1110 [16]	2005	80	M	NO	YES	NR	3	NA
15	Acta Cardiol. 2006;61(5):577–579 [27]	2006	57	M	NO	YES	0	2	NA
16	J Thromb Thrombolysis. 2006;22(1):77–78 [28]	2006	58	M	NO	YES	0	2	NA
17	Pharmacotherapy. 2006;26(3):423–427 [29]	2006	51	M	NO	YES	NR	3	NA
18	J Thromb Thrombolysis. 2007;24(1):59–64 [30]	2007	62	M	NO	YES	0	2	NA
19	J Thromb Haemost. 2007;5(5):1068–1070 [15]	2007	57	M	NO	YES	NR	3	NA
20	Ann Thorac Surg. 2009;87(2):629–631 [31]	2009	63	M	NO	YES	3	3	NA
21	Tex Heart Inst J. 2009;36(1):55–57 [32]	2009	58	M	NO	YES	3	3	NA
22	Tex Heart Inst J. 2010;37(1):109–112 [33]	2010	50	M	NO	YES	NR	3	NA
23	Indian J Pharmacol. 2011;43(6):726–728 [34]	2011	44	M	NO	YES	0	2	NA
24	Int J Hematol. 2012;96(3):370–375 [35]	2012	73	M	NO	YES	2	2	NA
25	Int J Hematol. 2012;96(3):370–375 [35]	2012	47	M	NO	YES	NR	2	NA
26	Asian Cardiovasc Thorac Ann. 2013;21(1):74–76 [36]	2013	55	M	NR	NR	NR	3	NA
27	Platelets. 2013;24(4):335–337 [10]	2013	87	F	NR	NR	NR	3	NA
28	Case Rep Med. 2014;2014:190149 [37]	2014	50	M	NO	YES	1	3	NA
29	Platelets. 2015;26(2):197–198 [17]	2015	40	M	NO	YES	0	3	NA
30	Exp Ther Med. 2016;12(2):1177–1180 [38]	2016	68	M	NO	YES	NR	2	NA
31	Int J Cardiol. 2016;206:93–94 [19]	2016	66	M	YES	YES	3	5	1
32	Case Rep Cardiol. 2016;2016:4605139 [20]	2016	69	M	NO	YES	2	3	NA
33	Case Rep Hematol. 2018;2018:4357981 [39]	2018	79	M	NO	YES	0	2	NA
34	Am J Ther. 2019;26(4):e534–e537 [40]	2019	79	M	NO	YES	0	2	NA
35	Am J Ther. 2019;26(5):e659–e661 [41]	2019	64	M	NO	YES	1	2	NA
36	Pharmacotherapy. 2019;39(4):521–525 [42]	2019	59	M	NO	YES	0	2	NA
37	J Investig Med High Impact Case Rep. 2020;8:2324709620947891 [43]	2020	66	F	NO	YES	1	2	NA
38	J Community Hosp Intern Med Perspect. 2020;10(2):171–173 [44]	2020	69	F	NO	YES	1	2	NA
39	Cureus. 2021;13(9):e18217 [45]	2021	68	M	NO	YES	1	2	NA
40	Pan Afr Med J. 2021;38:9 [46]	2021	65	F	NO	YES	3	2	NA
41	Pan Afr Med J. 2021;38:9 [46]	2021	76	F	NO	YES	1	2	NA
42	Am J Ther. 2022;10.1097/MJT.0000000000001553 [47]	2022	42	M	NO	YES	0	2	NA
43	Heliyon. 2023;9(4):e14504 [48]	2023	59	F	NO	YES	0	2	NA

F: female; M: male; HOS: Hospitalized; BL: Bleeding; SC: Score; EX: Exclusion; NR: Not Recorded; NA: Not Applicable.

Table 3. Summary of thrombocytopenia occurrence in studies of tirofiban.

NO.	CITATION	YEAR	AGE	TOTAL TCP	EX TCP	CT TCP	SC	EX
1	Circulation. 1997;96(5):1445–1453 [14]	1997	ADULT	22	12	10	4	NA
2	N Engl J Med. 1998;338(21):1488–1497 [49]	1998	ADULT	21	15	6	4	NA
3	N Engl J Med. 1998;338(21):1498–1505 [13]	1998	ADULT	24	18	6	1	NA
4	N Engl J Med 2001;344:1888–1894 [12]	2001	NR	70	12	58	1	NA
5	J Am Coll Cardiol. 2004;44(1):14–19 [50]	2004	NR	2	1	1	5	1
6	Platelets. 2005;16(8):462–468 [51]	2005	ADULT	11	NR	NR	3	NA
7	JAMA. 2005;293(17):2109–2117 [52]	2005	NR	9	1	8	1	NA
8	JAMA. 2008;299(15):1788–1799 [53]	2008	NR	18	3	15	1	NA
9	Lancet 2008;372(9638):537–546 [54]	2008	ADULT	19	10	9	4	NA
10	Eur J Clin Pharmacol. 2012;68(5):821–832 [55]	2012	ADULT	10	9	1	2	NA
11	Stroke. 2021;52(12):3750–3758 [56]	2021	ADULT	1	1	0	4	NA

Total TCP: Total number of patients with thrombocytopenia; EX TCP: The number of people in the experimental group that developed thrombocytopenia; CT TCP: The number of people in the control group that developed thrombocytopenia; SC: Score; EX: Exclusion; NR: Not Recorded; NA: Not Applicable.

The treatment strategies for tirofiban-induced thrombocytopenia included modification of the drug regimen, other interventions to minimize the risk of bleeding, and supportive care. The risk of bleeding increases when the patient’s platelet count is below 100 × 10⁹/L, and the therapies which affect hemostasis may need to be discontinued [57]. After discontinuation of treatment, the platelet count might normalize within 1–6 days (mean of 2.1 days) [11]. However, platelet count recovery might be slower in individuals with decreased metabolism due to renal or hepatic insufficiency [33]. Platelet transfusions were recommended in the presence of active bleeding associated with profound thrombocytopenia (platelet count < 20 × 10⁹/L) by the European Society of Cardiology [58], but maybe little effect while reversibly binding GP IIb/IIIa inhibitor tirofiban remains in circulation (half-life 2 h) [59]. In patients with persistent major bleeding, fibrinogen supplementation with cryoprecipitate or fresh frozen plasma may be considered [46]. Supportive treatments for profound thrombocytopenia might include corticosteroid and immunoglobulin therapy [58]. Intravenous immunoglobulin (IVIG) was frequently used to successfully treat thrombocytopenia caused by tirofiban [10,15,17], resulting in an immediate increase in the platelet count and rapid return to normal levels. However, due to thrombocytopenia evolving, inadequate antiplatelet therapy and the use of thrombopoietin may increase the risk of thrombosis, occasionally leading to the occurrence of thrombotic events [38].

2.2. Pathogenesis

Currently, the pathophysiological mechanism of tirofiban-induced thrombocytopenia is still uncertain, it is thought to be associated with immune-mediated thrombocytopenia induced by drug-dependent antibodies (DDAbs) [60]. Drug-induced immune thrombocytopenia (DITP) refers to thrombocytopenia due to drug-dependent antibody-mediated destruction of platelets. Evidence suggested that the combination of DDAbs, drug, and target antigen – the drug may be trapped at the antigen-antibody interface to form a three-molecule complex – was the mechanism of DITP associated with drugs like quinine and many others. Then, macrophages recognized the Fc ‘tail’ of DDAbs and cleared antibody-coated platelets from the circulation [21,61]. When the drug is present, it can bind to the complementarity-determining regions (CDR) of the antibody to remodel its structure and increase its binding affinity to the target antigenic, which in the absence of the drug is weak for the antibody response to the target antigen [62]. The most common target antigens include platelet glycoprotein complexes such as GP IIb/IIIa and GPIb/IX [63].

The conformation of GP IIb/IIIa was altered after binding to tirofiban, as demonstrated by numerous studies, enabling it to bind to specific DDAbs. Platelets were removed from circulation as a result of these DDAbs being subsequently recognized by immune-mediated reaction. Bougie et al. [18] demonstrated DDAbs specific for GP IIb/IIIa in nine patients who developed acute and severe thrombocytopenia (platelet count < 25 × 10⁹/L) after treatment with tirofiban (four patients) or eptifibatide (five patients). In the acute phase, each patient’s serum contained a high titer of IgG antibodies that reacted to the GP IIb/IIIa complex only in the presence of the therapeutic drugs. Neither tirofiban- nor eptifibatide-dependent antibodies were found in 100 randomly selected healthy blood donors. Of the 23 patients treated with tirofiban or eptifibatide without thrombocytopenia (platelets less than 100 × 10⁹/L), 21 had no detectable antibodies and only 2 had extremely weak tirofiban-dependent antibodies. The results indicated that acute thrombocytopenia following tirofiban or eptifibatide treatment may be caused by DDAbs induced by prior exposure to the drug or produced naturally. This could provide an explanation for the instances of thrombocytopenia observed a few hours after the initial administration [21]. In 2003 and 2005, Dunkley et al.confirmed that thrombocytopenia was caused by tirofiban by finding DDAbs in the blood of four individuals using flow cytometry [16,23]. In addition, Clofent-Sanchez et al. [15] verified that the DDAbs were indeed exclusively generated in the presence of tirofiban by ELISA and flow cytometry, indicating the specificity of the immune response. It should be noted as well that no tirofiban-depended antibodies were detected by ELISA when the platelet count recovered.

Evidence was obtained that the antibodies might recognize multiple target epitopes on the GP IIb/IIIa-drug complex. The binding of GP IIb/IIIa ­antagonists to the RGD recognition site might induce conformational changes in GP IIb/IIIa, leading to the expression of novel epitopes or ligand-induced ­binding sites (LIBS) that can be recognized by the antibodies already present in the plasma (Figure 1(B)) [61]. Bougie et al. [64] found that antibodies recognize subtle drug-induced structural changes that were mainly distributed around RGD recognition sites in the head region of GP IIb/IIIa heterodimer. After binding to antibodies, platelets are destructed by macrophages in the spleen or liver through Fcγ receptor signaling via spleen tyrosine kinase (Syk) [11,65,66].

Additionally, various studies have demonstrated that tirofiban-induced antibodies can result in platelet activation and/or microparticle formation [67]. This may explain why the adverse ischemic outcomes observed in the EPIC trail that patients occurred thrombocytopenia had an approximately 12-fold higher mortality rate at 30 days and a roughly four-fold increase rate of myocardial infarction [68]. It could be assumed that platelets may be activated in GP IIb/IIIa inhibitor-induced thrombocytopenia, increasing the risk of thrombosis events and accelerated platelet clearance [69]. The evidence of platelet activation caused by tirofiban-dependent antibodies found by Dunkley et al. [23] may further support the hypothesis. They identified the tirofiban-dependent antibodies in two patients who had occurred profound thrombocytopenia by flow cytometry. The results of the serotonin release test (SRA) represented that tirofiban-dependent antibodies detected in the serum of two patients caused platelet activation. They followed cases receiving tirofiban in order to confirm that this was caused by tirofiban-dependent antibodies [51]. 11 of 871 patients treated with tirofiban developed profound thrombocytopenia (platelet count < 20 × 10⁹/L). Tirofiban-dependent antibodies were confirmed in all cases by flow cytometric assay. Thrombotic events occurred in seven of the eleven patients, including six patients with persistent or recurrent cardiac ischemia and one patient with deep vein thrombosis. Platelet activation was analyzed by various methods, including SRA, P-selectin, and annexin V, showed that thrombotic events were significantly associated with platelet activation. Six patients had evidence of platelet activation, significantly related to further coronary ischemic events occurring during acute thrombocytopenia. Nevertheless, the amount of data is still very limited currently, and more data will be needed to verify such a hypothesis. Further research should be undertaken to investigate how platelets activate and what factors influence them.

2.3. Risk factors

Thrombocytopenia may have negative effects on clinical outcomes, yet clinicians have difficulty identifying high-risk patients with thrombocytopenia. Some patients with thrombocytopenia may not be noticed until severe bleeding complications develop. Therefore, it may be crucial to identify patients at high risk of thrombocytopenia for early and timely intervention.

Yi et al. [8] conducted a study to investigate risk factors of tirofiban-induced thrombocytopenia and developed a simple clinical assessment tool that can be used to predict the occurrence of thrombocytopenia before using tirofiban in patients undergoing PCI. Five independent risk factors for tirofiban-induced thrombocytopenia were ultimately identified by univariable and multivariable analysis, involving age ≥ 65 years, white blood cell ≥ 12 × 10⁹/L, diabetes mellitus, congestive heart failure, and chronic kidney disease (CKD). Patients with thrombocytopenia were found to be older than those without it (62.78 ± 10.93 vs. 60.66 ± 10.87) in this study. To date, the majority of patients reported in cases of tirofiban-induced thrombocytopenia were older than 65 years. The results of this study also showed a higher incidence of thrombocytopenia in diabetes mellitus patients (5.51% vs. 2.66%). Diabetes mellitus was a comorbidity in 664 patients (47.6%) in the study. Additionally, Chen et al. [70] found that patients with diabetes mellitus tended to increase the severity of thrombocytopenia. Another significant risk factor was congestive heart failure, which is defined as New York Heart Association Functional Class III to IV. In this research, the rate of thrombocytopenia in patients with congestive heart failure was 6.37%. Thrombocytopenia has also been reported as a clinical complication in heart failure patients. Mondal et al. [71,72] reported that the modulation of GP IIb/IIIa expression and shedding induced by oxidative stress may play a potential role in platelet apoptosis, leading to thrombocytopenia and non-surgical bleeding in patients with heart failure. It is well known that patients with CKD were at high risk of both bleeding and thrombosis [73]. According to the study, 11.1% of patients with thrombocytopenia had CKD, and patients with CKD were more likely to develop thrombocytopenia than those with normal renal function (7.74% vs. 3.28%). Furthermore, the univariate analysis indicated that white blood cell count was an independent risk factor for thrombocytopenia in patients receiving tirofiban. The cut point of 12 × 10⁹/L is further determined by the LOWESS program. Fountain et al. [74] found that 15% of 533 patients diagnosed with clostridium difficile infection (CDI) patients and elevated levels of white blood cell had moderate thrombocytopenia (platelet count < 100 × 10⁹/L at CDI diagnosis). Numerous studies have shown that infection was often accompanied by the activation of platelets, resulting in increased platelet consumption and clearance, leading to the development of thrombocytopenia [75].

For scoring purposes, age ≥ 65 years, DM, and congestive heart failure were counted as two points, CKD, and white blood cell ≥ 12 × 10⁹/L were scored one point (Table 4). Patients with higher risk scores were more likely to develop thrombocytopenia. This model can not only help clinicians identify patients at high risk for thrombocytopenia, but also reduce unnecessary platelet count monitoring in low-risk patients using tirofiban. If a patient is determined to be at high risk of tirofiban-induced thrombocytopenia, prompt treatments such as improving indicator monitoring and altering anticoagulation methods may be necessary. However, further large-scale studies are needed before the scoring system can be used in clinical practice.

Table 4. Scores points for the independent variables [8].

Risk factors	Points
Age
≥65 years	2
50–65 years	0
White blood cell
≥12 × 10^9/L	1
<12 × 10^9/L	0
Diabetes mellitus
Yes	2
No	0
Congestive Heart failure
Yes	2
No	0
Chronic kidney disease
Yes	1
No	0

Total score: ≤2: Low-risk; 3–6: Moderate-risk; ≥7: High-risk.

In addition, Adamo et al. [76] found that the incidence of thrombocytopenia was affected by differences in buffers used in the formulation of tirofiban to maintain a slightly acidic pH in undiluted drugs. In the PRISM trial, two different formulations of tirofiban were used sequentially. The study drug used in the early stage of the research was a phosphate-buffered product, which was later replaced by citrate-buffered formulation due to a report of drug instability. The results indicated that the incidence of thrombocytopenia was significantly higher in the early stage. Thrombocytopenia occurred more frequently in patients treated with phosphate buffer-tirofiban than in patients treated with unfractionated heparin (UFH) during the same period (1.7% vs. 0.5% at platelet nadir < 90 × 10⁹/L, 2.0% vs. 0.7% at platelet nadir < 100 × 10⁹/L), with approximately three-fold increased risk of thrombocytopenia. However, during the later stage, no significant difference was observed between citrate-buffered tirofiban and UFH (0.3% vs. 0.1% at platelet nadir < 90 × 10⁹/L, and 0.7% vs. 0.7% at platelet nadir < 100 × 10⁹/L). When compared to UFH, citrate-buffered tirofiban had a lower 30-day risk of mortality (HR: 0.49; 95% CI: 0.27–0.89; p = 0.019), whereas phosphate-buffered tirofiban did not vary from UFH (HR: 0.87; 95% CI: 0.49–1.54; p = 0.629). Thrombocytopenia is associated with a 5- to 10-fold increased risk of TIMI (Thrombolysis In Myocardial Infarction) major or minor bleeding. Patients with platelet nadir < 100 × 10⁹/L also had a two-fold increased risk of developing net adverse cardiovascular events. It is yet unknown why different buffer formulations of the same pharmacological molecule cause thrombocytopenia in variable degrees. It was speculated that the conformational changes in GP IIb/IIIa induced by phosphate-buffered tirofiban were more frequently recognized by naturally occurring antibodies than citrate-buffered tirofiban, and therefore more prone to thrombocytopenia. Phosphate-buffered tirofiban was associated with a higher incidence of thrombocytopenia compared with citrate-buffered tirofiban and may therefore increase the risk of adverse clinical outcomes. Therefore, comprehensive post-marketing surveillance of the formulations of drugs is necessary.

3. GP IIb/IIIa inhibitors-induced thrombocytopenia

In most countries, currently clinically available GP IIb/IIIa inhibitors include abciximab, eptifibatide and tirofiban, the former being monoclonal antibodies and the latter two being small molecule compounds [1]. Abciximab is the Fab fragment of the mouse-human monoclonal antibody 7E3 that binds to the epitope near the ligand binding site of GP IIb/IIIa, resulting in steric hindrance, thus inhibiting platelet aggregation. Eptifibatide is an 832 Da cyclic heptapeptide ligand mimetic containing a lysine-glycine-aspartic acid (KGD) sequence that reversibly blocks platelet aggregation by binding and inhibiting GP IIb/IIIa receptors.

As with tirofiban, abciximab or eptifibatide can also cause acute, severe thrombocytopenia. The various GP IIb/IIIa inhibitors have different propensities to induce thrombocytopenia [60]. A pooled, patient-level analysis of the CHAMPION trials [77] showed a significantly higher incidence of thrombocytopenia in patients treated with GP IIb/IIIa inhibitors compared with patients who did not (2.0% vs. 0.68%). Thrombocytopenia developed in 18 of 558 (3.2%) patients treated with abciximab, 31 of 2105 patients (1.5%) with eptifibatide, and 9 of 327 patients (2.8%) with tirofiban. The final multivariate model analysis revealed that GP IIb/IIIa inhibitors use, age, hyperlipidemia, diabetes mellitus, and prior coronary artery bypass grafting were independently associated with thrombocytopenia, and GP IIb/IIIa inhibitors use was the strongest predictor. A meta-analysis of 29 randomized trials involving 123,419 patients showed that the use of GP IIb/IIIa inhibitors increased the risk of thrombocytopenia (platelet count < 100 × 10⁹/L) by 63% compared with placebo [9]. The results of the subgroup analysis showed that abciximab [risk ratio (RR) = 2.93, 95% confidence interval (CI) 2.43–3.52] and tirofiban (RR = 2.79, 95% CI 1.17–6.63) significantly increased the risk of thrombocytopenia compared with placebo, while the risk of thrombocytopenia with eptifibatide (RR = 1.05, 95% CI 0.86–1.29) did not increase significantly. The different rates of thrombocytopenia observed with various GP IIb/IIIa inhibitors indicated that the new epitopes produced by drugs were different. Thrombocytopenia occurred more frequently in patients with abciximab. It may be that abciximab is an antibody derived and has a higher antigenicity than small molecule compounds such as tirofiban, and thus is more likely to induce an immune response [11]. In the EPIC trial, lower baseline platelet count (< 200 × 10⁹/L), older age (> 65 years), and lighter weight (< 80 kg) were considered to be important risk predictors of thrombocytopenia [26,68].

Five patterns of GP IIb/IIIa inhibitors-induced thrombocytopenia had been described: (1) acute severe thrombocytopenia (platelet count < 10 × 10⁹/L) within 12 h of first exposure, (2) acute thrombocytopenia within 12 h of the second exposure, (3) delayed thrombocytopenia (in 5–7 days after treatment), (4) anaphylaxis after first or second exposure, and (5) pseudo-thrombocytopenia due to platelet clumping [78].

Given that some patients who developed thrombotic events after drug-induced thrombocytopenia may need further treatment, switching to another GP IIb/IIIa inhibitor with a lower or absent risk of thrombocytopenia is critical. So far, at least four patients with thrombocytopenia during abciximab administration were subsequently treated with tirofiban [79,80] or eptifibatide [81,82]. Clofent-Sanchez et al. [15] reported that a patient with profound and prolonged thrombocytopenia induced by tirofiban was treated with abciximab without thrombocytopenia. Lev et al. [83] preliminarily evaluated the clinical safety of sequential treatment with abciximab after tirofiban or eptifibatide in patients undergoing PCI. Of them, 25 patients were treated with post-tirofiban abciximab (tirofiban-abciximab group), 10 with post-eptifibatide abciximab (eptifibatide-abciximab group), and 15 with abciximab alone (abciximab control group). First, tirofiban and eptifibatide were administered over 20–24 h in the tirofiban-abciximab group and eptifibatide-abciximab group, respectively. Prior to PCI, abciximab was initiated and administered continuously over 12h in all patients after obtaining arterial access, and tirofiban or eptifibatide was discontinued after 5 min of overlap with the abciximab infusion. None of the 50 patients developed severe thrombocytopenia. Only 1 in the 25 patients in the tirofiban-abciximab group, and 2 in the abciximab control group had mild thrombocytopenia. This finding indicated that the administration of abciximab after tirofiban or eptifibatide may be effective and safe. It was speculated that the epitopes recognized by the presumably pathogenic antibodies could be different for the three drugs. In vitro, abciximab has inhibited the antibody reaction to thrombocytopenia induced by the ligand-mimetic agents (eptifibatide and tirofiban). The proposed mechanism for this lack of cross-reactivity was the steric hindrance provided by the molecule after binding to the GP IIb/IIIa receptor [18]. However, abciximab-associated thrombocytopenia had been reported years after previous tirofiban-induced thrombocytopenia [29]. Based on the aforementioned evidence, patients with a history of thrombocytopenia induced by these agents should be cautious when readministered GP IIb/IIIa inhibitors.

4. Conclusion

Although tirofiban-induced thrombocytopenia is self-limiting, it has independent prognostic effects because it is associated with a higher risk of death, ischemic events, and bleeding than patients without thrombocytopenia [11]. This tirofiban-related adverse effect resulting from tirofiban usually develops within 24 h after initial treatment, and the platelet count may return to normal in 1–6 days. Close monitoring is essential during the first few hours of treatment, and if thrombocytopenia is detected and confirmed, the drug should be promptly discontinued. If thrombocytopenia persists or active clinical bleeding complications are observed, the patient should be treated with platelet transfusions. It is reasonable to attempt corticosteroids or immunoglobulins in case of profound thrombocytopenia if there are no contraindications. Advanced age (≥ 65), white blood cell ≥ 12 × 10⁹/L, DM, congestive heart failure, and CKD are considered to be independent risk factors for predicting patients at high risk of thrombocytopenia, and clinicians should pay particular attention to those groups. Tirofiban-induced thrombocytopenia is thought to be an immune-mediated reaction. Since the drug-dependent antibodies may persist for years, patients should avoid using the drug in the future and should be cautious when using other GP IIb-IIIa inhibitors [84].

Ethics approval

Ethics committee approval is not required for narrative reviews.

Authors contributions

JW conceptualized the manuscript and performed the literature search and manuscript drafting. DZ supervised and revised the manuscript. All authors contributed to the article and approved the submitted version.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Data availability statement

Data sharing is not applicable to this article as no new data were created or analyzed in this study.

Additional information

Funding

This work was partially funded by the Special Funding for Clinical Pharmacy Research of Shandong Provincial Medical Association YXH2022ZX01002.