Quetiapine-induced thrombotic microangiopathy in a patient on maintenance dialysis

Abstract

Quetiapine has been reported to cause immune-mediated thrombotic microangiopathy (TMA), although few cases have been reported thus far. A 71-year-old man with autosomal dominant polycystic kidney disease on maintenance dialysis was hospitalized with a hemorrhagic basal ganglia stroke, and was treated with 25 mg quetiapine for delirium from day 4 of admission. There was no worsening of consciousness, fever, diarrhea, or elevated blood pressure during the hospitalization. Gingival bleeding appeared on day 35, and the platelet count on day 38 was 0.5 × 10⁴/μL (13.2 × 10⁴/μL on day 16). The presence of 1% schistocytes, high LDH level, inability to measure haptoglobin, negative direct Coombs test, and normal prothrombin time and activated partial thromboplastin time indicated TMA. We considered an exclusionary diagnosis of drug-induced TMA, because of normal ADAMTS13 activity, no evidence of complement activation and the absence of Shiga toxin or symptoms of collagen disease or cancer. Quetiapine was the most likely causative factor; however, all drugs, including heparin, were discontinued or changed. Due to persistent microbleeding, platelet transfusions were performed several times. After only quetiapine was discontinued, the platelet count recovered smoothly to 3.1 and 7.2 × 10⁴/μL on days 45 and 72, respectively; LDH and fibrinogen levels normalized on day 47. All medications, except quetiapine, were restarted sequentially after day 47, without subsequent thrombocytopenia. Platelet activation predominantly by a drug-dependent antibody might be the etiology of quetiapine-induced TMA. Plasmapheresis may not be necessary for quetiapine, because of its unproven efficacy in drug-induced TMA.

Introduction

Thrombotic microangiopathy (TMA) is a clinicopathological syndrome characterized by microangiopathic hemolytic anemia, consumptive thrombocytopenia, and organ dysfunction due to microvascular thrombi. Thrombotic thrombocytopenic purpura (TTP), a typical example, is triggered by decreased ADAMST13 activity. Many diseases other than TTP have been found to cause TMA, and they are now often classified by etiology, such as infection and complement abnormalities.

Antipsychotics used for the treatment of schizophrenia and delirium are often prescribed not only by psychiatrists, but also by internists. Quetiapine is a widely used atypical antipsychotic due to its low frequency of dopamine-related side effects. Quetiapine rarely causes immune-mediated TMA.

We experienced a case of TMA caused by quetiapine and treated with drug discontinuation alone, without plasma exchange. Only two cases of quetiapine-induced TMA [1, 2] have been reported in the literature to date.

Case presentation

A 71-year-old man who had been on maintenance dialysis for 5 years due to end-stage renal failure secondary to autosomal dominant polycystic kidney disease (ADPKD) presented to our hospital complaining of unsteady gait. A head computed tomography (CT) scan revealed a hemorrhagic basal ganglia stroke and he was hospitalized. He had a history of atrial fibrillation (AF); however, he was not taking an anticoagulant. A head CT performed on hospitalization day 2 showed no rebleeding, and he resumed dialysis. The patient did not lose consciousness before or after dialysis. He had no paralysis of his limbs and was able to speak easily, but communication was slightly difficult due to difficulties with language fluency and auditory comprehension. He was determined to be delirious, removed his intravenous drip tubes more than once and was started on quetiapine (25 mg/day) on hospitalization day 4. His altered behavior improved, though communication remained difficult. The patient continued maintenance dialysis three times a week as well as rehabilitation. His higher brain dysfunction, such as spontaneous speech and auditory comprehension, gradually improved. The patient’s blood pressure was maintained at 120–150 mmHg throughout his hospitalization.

On day 35, a few hours after the end of dialysis, he bled again from the puncture site of vein in the hospital room. He tried to stop the bleeding using a bandage, but slight bleeding continued until the evening of the following day. A large amount of coagulation was found in the oral cavity during dinner on day 37, which revealed continuous bleeding from the upper and lower gums. On hospitalization day 38, the patient’s platelet count decreased to 0.5 × 10⁴/μL from the 13.2 × 10^4/μL count on hospitalization day 16 (Table 1). A physical examination revealed minimal bleeding of the gums, but no purpura of the extremities, fever, change in blood pressure, altered consciousness, or history of diarrhea. There was no increase in megakaryocytes and there were no findings suggestive of myeloproliferative disease on bone marrow examination. Magnetic resonance imaging (MRI) of the head did not reveal any signs of acute infarction and chest and abdomen CT showed no obvious tumors. The blood test revealed negative for anti-platelet factor 4 (PF4)/heparin antibody. The patient also exhibited high lactate dehydrogenase (LDH) levels, undetectable haptoglobin (Hpt), a negative direct Coombs test, normal prothrombin time (PT) and activated partial thromboplastin time (aPTT). Although the diagnostic criteria for DIC were met, there was no causative infection or tumor and, most importantly, PT in the normal range, which was not typical. Finally, he was diagnosed with TMA on an exclusionary diagnostic basis. However, despite the severity of low platelet counts, a schistocyte count of 1% was found on blood smear examination (Fig. 1).

Table 1.

Laboratory findings at the onset of thrombotic microangiopathy

WBC	2550	/μL	Na	132.5	mEq/L
Hb	9.8	g/dL	K	4.2	mEq/L
MCV	96.2		Cl	98	mEq/L
Plt	0.5	104/μL	Ca	8.3	mg/dL
Ret	6.8	‰	P	2.9	mg/dL
PT	13.4	sec	CRP	0.11	mg/dL
PT-INR	1.18		BNP	392.3	pg/mL
APTT	36.1	sec	Haptoglobin	3	mg/dL
Fib	86.7	mg/dL	C3	64	mg/dL
FDP	345.3	μg/mL	C4	28.2	mg/dL
AT-3	98.4	%	CH50	36	U/mL
TP	6.3	g/dL	IgG	1299	mg/dL
Alb	3.4	g/dL	IgA	347	mg/dL
AST	17	IU/L	IgM	74	mg/dL
ALT	35	IU/L	ANA	< 40
LDH	402	IU/L	Anti-cardiolipin-beta2-glycoprotein I complex antibody	< 1.3	U/mL
T.Bil	1.09	mg/dL	Anticardiolipin antibody	11	U/mL
I.BIL	0.4	mg/dL	Lupus anticoagulant test	0.57
ALP	308	IU/L	Direct Coombs test	(–)
r-GTP	68	IU/L	Indirect Coombs test	(–)
BS	112	mg/dL	ADAMTS-13	1.08	IU/ml
UA	3.5	mg/dL	Anti-ADAMTS-13 inhibitor	< 0.5	BU/ml
BUN	44.9	mg/dL	HIT-IgG	< 0.13	U/ml
Cr	10.14	mg/dL	PAIgG	83	ng/107cells

ADAMTS13: a Disintegrin-like and Metalloproteinase with Thrombospondin type 1 motifs 13

HIT heparin-induced thrombocytopenia, PA platelet associated

Fig. 1.

Presence of schistocytes

Normal ADAMTS13 activity and negative Shiga-toxigenic Escherichia coli in the stool ruled out thrombotic thrombocytopenic purpura (TTP) and hemolytic uremic syndrome (HUS) as the cause of TMA. The patient did not have any symptoms suggestive of collagen disease or cancer. Complementary anomalies were ruled out due to a lack of a history of bleeding events. Drug-induced TMA was diagnosed by exclusion. The patient’s medication list included olmesartan, febuxostat, nalfurafine, ferric citrate hydrate, and lanthanum carbonate, all of which he had received for over 2 years. Two months prior to hospitalization, the patient had started treatment with evocalcet; quetiapine was initiated on hospitalization day 4. According to the Oklahoma TTP-HUS Registry, quetiapine can lead to drug-induced TMA. We discontinued all of the patient’s medications, replacing olmesartan with carvedilol and heparin with nafamostat. We chose to monitor the patient after the discontinuation of his medications and did not perform plasma exchange, as there is no clear evidence of the efficacy of plasma exchange in TMA induced by drugs other than ticlopidine. A platelet transfusion was performed on hospitalization day 38 to ensure safe dialysis in the presence of minimal persistent bleeding. The patient had no fever or altered consciousness after the platelet transfusion and four additional transfusions were performed within the following 7 days (Fig. 2).

Fig. 2.

Clinical course of this case. Hpt haptoglobin

On hospitalization day 45, the platelet count improved to 3.1 × 10⁴/μL. The platelet count continued to improve to 5.9 × 10⁴/μL on hospitalization day 47 and reached 7.2 × 10⁴/μL on hospitalization day 72. The patient’s LDH and fibrinogen levels normalized on hospitalization day 47 and his Hpt reached 70 mg/dL on hospitalization day 100 and normalized to 95 mg/dL on hospitalization day 181. Excluding quetiapine, the patient’s medications were resumed sequentially after hospitalization day 47, with no decrease in the platelet count. One year after the onset of TMA, the platelet count remained favorable at 9.0 × 10^4/μL and no new symptoms of collagen disease or tumors were detected. To date, the patient has no neurological sequelae from TMA and has been able to successfully continue dialysis.

Discussion

Although quetiapine is known to cause TMA similar to that induced by quinine, there are very few reports describing this condition and no case presentations reporting the course of the disease as comprehensively as the present report. Herein, we show that quetiapine-induced TMA is also unique in that it features a patient with underlying ADPKD who was undergoing dialysis at the time of diagnosis and was treated with discontinuation of medications alone.

TMA can be congenital, induced by medications, or caused by a variety of conditions, including TTP associated with ADAMST13, atypical HUS related to complement abnormality, HUS caused by Shiga toxin, and HELLP syndrome of pregnancy [3]. TTP has five classic signs: thrombocytopenia, microangiopathic hemolytic anemia, renal dysfunction, fever, and agitated neuropsychiatric symptoms. However, fewer than 10% of patients meet all five diagnostic criteria [4]; therefore, the current practice is to suspect TMA in patients with anemia with elevated LDH, thrombocytopenia, and organ damage and to confirm TMA with the presence of schistocytes (> 1%), low haptoglobin, a negative direct Coombs test, and normal PT and aPTT [5]. Differentiating TMA from DIC is often difficult. Many patients with TMA are known to meet the criteria for a DIC diagnosis, as in this case [6, 7]. Conversely, only 10–15% of patients with DIC meet the diagnostic requirements for TMA [6]. As a way to distinguish TMA from DIC, it has been reported that platelets < 2.0 × 10⁴/μL and PT < upper limit of normal + 5 s are more likely to be indicative of TMA [8]. These laboratory differences may be because fibrinolysis and platelets are activated in DIC, whereas only platelets are activated in TMA [6].

According to the Oklahoma TTP-HUS Registry, the frequency of drug-induced TMA is 13%, making it the second leading cause after idiopathic TMA [9]. Quetiapine is one of 22 drugs with clear evidence of a causal relationship with TMA [10, 11]. Drug-induced TMA can occur via immune-mediated and nonimmune-mediated (toxic) mechanisms [11]. The immune-mediated mechanism involves the drug-dependent antibody-mediated, primary activation of platelets and vascular endothelial cells, resulting in the formation of thrombi in the arterioles or capillaries. Unlike drug-induced immune thrombocytopenia, such as heparin-induced thrombocytopenia (HIT), drug-induced TMA is characterized by having target cells such as neutrophils and endothelial cells besides platelets. In addition, because the antibodies are drug dependent, no new organ damage occurs once the drug is eliminated from circulation [12]. Quinine is a typical drug that causes drug-induced TMA, and quetiapine is considered one of them. Diagnostic criteria for immune-mediated TMA have been proposed [10, 11]. The nonimmune-mediated (toxic) mechanism is thought to be associated with direct tissue impairment by the drug. Some develop with cumulative high doses, while others develop with a single dose. Typical causative agents include vascular endothelial growth factor, (VEGF) inhibitors, anticancer drugs (cyclosporine and tacrolimus), and narcotics (oxycodone and cocaine). In this case, after careful exclusion of events that could exacerbate endothelial damage (worsened hypertension and use of drugs with CYP3A4 inhibitory properties) and causes of TMA, we finally diagnosed quetiapine-induced TMA. Although thrombocytopenia was not detected within 21 days of the initiation of quetiapine, as required by the diagnostic criteria, blood tests were not performed between days 12 and 34 after the administration of quetiapine, and thus thrombocytopenia within the diagnostic time frame cannot be ruled out.

Quetiapine is a multi-acting receptor-targeted antipsychotic (MARTA) that blocks the function of a variety of receptors including dopamine receptors. It is widely used for the treatment of overactive delirium due to its safety and reliable sedative effect. Less than ten cases of quetiapine-induced thrombocytopenia have been reported, while the Oklahoma TTP-HUS Registry has recognized only one case of TMA due to quetiapine [1]. The recognized case developed with hemolytic anemia and thrombocytopenia (minimum 1.4 × 10⁴/μL) on day 5 after quetiapine administration and recovered with drug discontinuation and plasma exchange. The medical records revealed that a similar episode occurred in the patient 2 years earlier. Recently, another possible case of quetiapine-induced TMA has been reported, in which hemolytic anemia and thrombocytopenia (1.0 × 10⁴/μL) occurred after quetiapine administration and did not recover with plasma exchange, but did improve with drug discontinuation [2]. However, the patient in that case had an ADAMTS13 activity < 5%, indicating that TMA may have been a result of a different mechanism. The association with TMA is unclear, but it has been noted that quetiapine has a high risk of acute kidney failure among antipsychotic medications [13].

It is important to determine whether plasma exchange is necessary along with discontinuation of the causative drug for the treatment of drug-induced TMA. According to the American Society for Apheresis (ASFA) guidelines [14], plasma exchange is an effective treatment for ticlopidine-induced TMA, and has unknown significance in the treatment of TMA induced by clopidogrel, while it is ineffective for TMA induced by gemcitabine and quinine. However, George JN et al. reported in a systematic review of published reports that only 22 of the 78 reported to cause TMA had a clear causal relationship [11], but ticlopidine and clopidogrel were not among them. Quetiapine is one of the 22 drugs, and two of the aforementioned cases of quetiapine-induced TMA were treated with plasma exchange, though it was ineffective in one case and the other case had simultaneous plasma exchange and drug discontinuation. In immune-mediated TMA, target cells including platelets are activated by an immune response to a new antigen expressed by the drug and are thought to cause TMA [15]. Hence, we acted on the thought that the TMA would improve with discontinuation of the drug, as with drug-dependent thrombocytopenia such as HIT. However, this case was successful in that approach, but whether plasma exchange is unnecessary needs to be examined in the future. In addition, we subsequently administered five platelet transfusions with no worsening of the disease. Based on the results of an observational study [16], platelet transfusions are generally contraindicated during TMA. However, the results of that study were criticized, as the platelet transfusion group may have been more severely ill. In addition, prospective studies have reported that platelet transfusions did not differ in mortality or neurological sequelae [17]. The half-life of quetiapine is less than 4 h; therefore blood transfusions more than 4 h after the administration of quetiapine are unlikely to cause immunomediated TMA. In our patient, the platelet transfusion reduced the bleeding and did not cause any obvious organ damage.

It should be noted that splenomegaly and end-stage renal disease may have influenced the blood test results in our patient. Our patient had splenomegaly secondary to portal hypertension due to numerous hepatic cysts, which resulted in his baseline platelet count being low (about 9.0 × 10⁴/μL). The prolonged time required for normalization of Hpt levels in our patient may be explained by the consumption of Hpt by splenomegaly [18] and the destruction of red blood cells by dialysis. Secondly, the low number of schistocytes in our patient might be related to the low number of renal vessels due to chronic renal failure. Renal vessels may play an important role in the pathophysiology of TMA, because the most common site of thrombosis in TMA is the kidney (incidence: 80%) [19], recurrence of TMA is reduced after the introduction of dialysis [20], and two reports have shown that bilateral nephrectomy was effective in treatment-refractory TMA [21, 22]. In this regard, patients on dialysis with reduced renal vascularity may have unusual abnormalities, such as reduced schistocytes counts, mild thrombocytopenia, or delayed onset of TMA. However, since the incidence of de novo TMA is only 0.3% of cases in the dialysis population [23], the clinical characteristics in these patients are still unknown.

Over three-quarters (76%) of drug-induced TMAs are caused by nine medications [11]. TMA induced by other drugs, including quetiapine, are scarce. Although the incidence of TMA induced by quetiapine is low, quetiapine is a widely used drug. Therefore, further elucidation of the pathogenesis of the disease through the accumulation of cases is necessary.

Compliance with ethical standards

Conflict of interest

All the authors have declared no competing interest.

Human and animal rights

This article does not contain any studies with human participants performed by any of the authors.

Informed consent

Informed consent was obtained from all individual participants included in the study.