INTRODUCTION — Thrombotic microangiopathies (TMAs) are potentially life-threatening conditions caused by small-vessel platelet microthrombi. Patients have the characteristic clinical features of microangiopathic hemolytic anemia (MAHA) and thrombocytopenia, and they may have acute kidney injury (AKI), neurologic abnormalities, and/or cardiac ischemia.
Drug-induced TMA (DITMA) is especially challenging to diagnose because specific laboratory tests to identify a drug etiology may not be available, and the role of a potentially implicated drug or other ingested substance may not be clear. Some of the implicated substances are illegal and may not be volunteered by the patient, and quinine exposure often is not recorded in the medication history when quinine is obtained without a prescription or ingested in a beverage such as tonic water.
The principal treatment of DITMA is withdrawing the suspected drug. However, management of a patient with suspected DITMA may be challenging because the clinical diagnostic criteria for DITMA overlap with other primary TMAs (eg, thrombotic thrombocytopenic purpura [TTP], complement-mediated TMA) that do require specific interventions (therapeutic plasma exchange [TPE] and anti-complement therapy, respectively).
Here we discuss our approach to the evaluation and management of a patient with a suspected DITMA.
Separate topic reviews present an overview of our approach to the evaluation of a patient with unexplained MAHA and thrombocytopenia as well as related TMAs including TTP:
●General approach – (See "Diagnostic approach to suspected TTP, HUS, or other thrombotic microangiopathy (TMA)".)
●TTP diagnosis – (See "Diagnosis of immune TTP".)
●TTP treatment – (See "Immune TTP: Initial treatment".)
●Complement-mediated TMAs – (See "Complement-mediated hemolytic uremic syndrome in children" and "Thrombotic microangiopathies (TMAs) with acute kidney injury (AKI) in adults: CM-TMA and ST-HUS".)
TERMINOLOGY — Terminology for the thrombotic microangiopathy (TMA) syndromes is evolving as disease understanding increases and diagnostic testing allows identification of specific causes. (See "Diagnostic approach to suspected TTP, HUS, or other thrombotic microangiopathy (TMA)", section on 'Terminology'.)
Although some case reports and series use the term drug-induced thrombotic thrombocytopenic purpura (TTP), we prefer to avoid this term and to reserve TTP for the TMA caused by ADAMTS13 deficiency. (See "Diagnostic approach to suspected TTP, HUS, or other thrombotic microangiopathy (TMA)", section on 'Overview of primary TMA syndromes'.)
We divide drug-induced TMA (DITMA) into immune-mediated syndromes, which can occur after exposure to any amount of the drug and are due to an idiosyncratic, antibody-dependent mechanism, and non-immune syndromes, which do not involve drug-dependent antibodies; the non-immune syndromes generally are dose related [1,2].
There are major implications of this distinction, including differences in presentation, management, and prognosis; however, for many reports, the mechanism (immune versus non-immune) is uncertain. These differences are discussed below. (See 'Clinical manifestations' below and 'Management' below and 'Prognosis/expected recovery' below.)
PATHOPHYSIOLOGY
Overview of pathophysiology — DITMA is an acquired condition resulting from exposure to a drug (eg, medication, other substance) that induces formation of drug-dependent antibodies or causes direct tissue toxicity that results in the formation of platelet-rich thrombi in small arterioles or capillaries.
As with other TMAs, thrombocytopenia results from platelet consumption in microthrombi, and microangiopathic hemolytic anemia occurs as red blood cells (RBCs) become fragmented when they pass across these thrombi. Organ ischemia and infarction may also occur as a result of small-vessel thrombosis. The kidney appears to be especially susceptible in DITMA syndromes.
The mechanism of DITMA can be immune (antibody-mediated) or non-immune (toxicity related to cumulative dose). However, initial reports of a new cause of DITMA often do not specify (and have not determined) the mechanism, which requires testing for drug-dependent antibodies. Thus, the mechanism for a specific drug may be uncertain.
Additional information about the pathogenesis of microangiopathic hemolysis, thrombocytopenia, and pathologic lesions in TMA is presented separately. (See "Pathophysiology of TTP and other primary thrombotic microangiopathies (TMAs)".)
Immune-mediated mechanism — In immune-mediated DITMA, the drug induces formation of antibodies that react weakly with multiple cells including platelets, neutrophils, and endothelial cells in the absence of the drug; strong binding only occurs in the presence of the drug (or drug metabolite) [3-8]. These antibodies are described as drug-dependent antibodies. Quinine is the classic example (figure 1). (See 'Quinine' below.)
The binding of drug-dependent antibodies to platelets is similar to drug-induced immune thrombocytopenia (DITP), but in DITP, platelets are the only target of the antibodies. In patients with DITMA, there may be multiple cellular targets that result in microvascular damage and formation of platelet microthrombi. Thrombocytopenia is thought to be due to platelet consumption in microthrombi.
As with other immune-mediated drug reactions, the generation of antibodies requires previous or ongoing exposure to the drug, and the severity of the reaction is independent of the drug dose. Thus, severe reactions may occur rapidly following extremely small drug exposures (eg, the quinine in one sip of tonic water) (see 'Quinine' below). Antibodies in immune-mediated DITMA may be highly specific for the structure of the implicated drug [9].
Antibodies in immune-mediated DITMA require the presence of the drug (or a drug metabolite) to cause cellular damage. Thus, once the drug is cleared from the circulation (hours to days for most drugs), no new organ injury occurs. However, tissue damage (especially kidney injury) may recover slowly and/or incompletely, and therefore MAHA and thrombocytopenia may be slow to resolve. (See 'Prognosis/expected recovery' below.)
Non-immune mechanism — Non-immune (non-antibody-mediated) DITMA may develop by multiple mechanisms. DITMA may only occur with high cumulative doses of a drug given for a long period of time, or it may occur with a single exposure to a drug. Often the TMA develops with rapid onset, such as with intravenous injection of oxymorphone extended-release tablets intended for oral use (Opana ER). (See 'Drugs of abuse' below.)
Cytotoxic cancer chemotherapies such as vascular endothelial growth factor (VEGF) inhibitors are common examples of the former (see 'Cancer therapies' below). Intravenous rather than oral administration of Opana ER can cause DITMA with a single administration. (See 'Drugs of abuse' below.)
Non-immune DITMA generally results from direct tissue injury from the implicated agent. For three drugs, the mechanism has been defined [10]:
●VEGF inhibition in glomerular endothelial cells impairs the unique permeability characteristics of these cells and promotes microvascular injury [11].
●Type I interferon can cause a dose-dependent systemic TMA [12].
●The toxicity of extended-release oxymorphone (Opana ER) and oxycodone (OxyContin) is mediated by the high molecular weight inert ingredient polyethylene oxide (PEO; molecular weight, approximately 7 million Daltons).
PEO was added to oxymorphone tablets in 2012 to resist crushing and solubilization and thereby to decrease the risk of intravenous abuse. However, PEO causes increased blood viscosity, hemolysis, and hemoglobinuric kidney injury [13]. Subsequently, DITMA was reported with intravenous abuse of oxycodone that had similarly been modified by the addition of PEO [10,14,15]. (See 'Drugs of abuse' below.)
Although uncommon, there appears to be a non-immune mechanism of DITMA in which excess activation of clotting factors results in MAHA and thrombocytopenia, similar to that seen in disseminated intravascular coagulation (DIC) or with certain rare genetic causes of TMA. The hemophilia drug emicizumab appears to cause DITMA in this manner. (See "Evaluation and management of disseminated intravascular coagulation (DIC) in adults", section on 'Pathogenesis' and "Pathophysiology of TTP and other primary thrombotic microangiopathies (TMAs)", section on 'Regulators of clotting or fibrinolysis'.)
DRUGS ASSOCIATED WITH DITMA
Overview of drugs and criteria — A number of drugs have been implicated in causing DITMA (sometimes reported in the literature as drug-induced thrombotic thrombocytopenic purpura [TTP] or drug-induced hemolytic uremic syndrome [HUS]). However, we limit associated drugs to those that have evidence for a definite or probable association with DITMA. The evidence for the role of these drugs was evaluated in a systematic review of all published reports describing DITMA, drug-induced TTP, or drug-induced HUS, based on stringent criteria for the level of supporting evidence [1,2,16]. The most common drugs with the strongest evidence for a causal association with DITMA are listed in the table (table 1); these are drugs for which at least one report has documented definite evidence for a causal association with TMA, or two reports documenting probable evidence. Many additional drugs have been reported to cause DITMA without data supporting a definite or probable association.
●For drugs with an assumed immune-mediated mechanism, probable causal association required exclusion of other causes of TMA (eg, TTP), elimination of the possibility that another drug could be responsible, and occurrence of symptoms within 21 days of daily exposure to a new drug or within 24 hours of exposure to a drug taken occasionally. Definite causal association for an immune mechanism required occurrence of systemic symptoms with previous exposure to the same drug or documentation of drug-dependent antibodies.
●For drugs with an assumed non-immune mechanism, probable causal association required exclusion of other causes of TMA and elimination of the possibility that another drug could be responsible. Definite causal association required resolution or improvement of the TMA when the suspected drug was stopped or the dose was reduced. These criteria were less stringent than criteria for an immune mechanism because the clinical course of these reactions can be more variable and there are no available laboratory tests.
Gemcitabine is the only drug for which we believe both immune and non-immune-mediated mechanisms have been documented; however, as with other drugs, most case reports for gemcitabine lack evidence sufficient for a probable or definite causal association.
In contrast, evidence is lacking for a role of the thienopyridine antiplatelet agents that block the platelet P2Y12 receptor (eg, clopidogrel, ticlopidine) in DITMA, despite multiple published reports [17,18]. Patients receiving these agents were also receiving multiple other medications, or their TMA recurred without re-exposure to the drug. Our opinion is that the reported occurrence of TMA in patients taking clopidogrel or ticlopidine was coincidental rather than drug induced.
The monoclonal antibody moxetumomab has been reported to cause hemolytic uremic syndrome (HUS) and has a Boxed Warning about this in the United States prescribing information, but the report does not provide definite or probable evidence. The diagnostic criteria used to make the diagnosis of HUS and the clinical course of the patients were not described.
For drugs that have not been reported but that have a presentation consistent with an immune-mediated mechanism (eg, abrupt onset, history of previous drug exposure associated with systemic symptoms only appreciated in retrospect), testing for drug-dependent antibodies is important for establishing the diagnosis, for advising the patient whether the drug should be avoided for life, and for reporting of new DITMA associations. A finding of drug-dependent antibodies in a patient with DITMA is considered definite evidence of an association and confirms the diagnosis; however, absence of drug-dependent antibodies does not eliminate the possibility of an association, because some antibodies may not be detected and some may be directed against a drug metabolite rather than the parent compound.
All published reports of DITMA and the grades of evidence supporting a causal association are listed on the first author's (JNG) website, which is regularly updated (http://www.ouhsc.edu/platelets/DITMA.htm).
Drugs (immune mechanism)
Quinine — Quinine is the most common cause of immune-mediated DITMA. In a systematic review of all published reports describing drugs and other substances as a suspected cause of TMA, we found that quinine was responsible for 34 of 104 cases in which there was definite evidence for a causal association (33 percent) [1]. Our experience with the Oklahoma Thrombotic Thrombocytopenic Purpura-Hemolytic Uremic Syndrome (TTP-HUS) Registry found quinine-associated TMA in 19 of 509 patients (4 percent) referred for a possible TMA over a 25-year period and found quinine as the cause of DITMA in 20 of 23 patients (87 percent) for whom a drug could be implicated as having a definite or probable causal association with the TMA [2,19]. A 2017 report describing the 19 individuals included in our registry found the following features [19]:
●All 19 were White people. This is distinctly different from TTP, in which approximately one-third of cases occur in Black people (sevenfold higher incidence than the demographics of the study population).
●Eighteen (95 percent) were women. This is greater than the increased frequency of women (75 percent) among patients with TTP.
●Eight (42 percent) had a prior history of quinine-related symptoms (nausea, vomiting, fever, chills, headache, confusion, ataxia).
●Thirteen (68 percent) could recall the precise timing between quinine ingestion and symptom onset (all ≤4 hours).
●Eighteen (95 percent) were caused by a quinine tablet; one was caused by quinine in tonic water of a vodka/tonic drink [6].
●Eighteen (95 percent) had evidence of quinine-dependent antiplatelet (or antineutrophil) antibodies.
●All had acute kidney injury; 17 of 18 required dialysis; three developed end-stage kidney disease; and two underwent kidney transplantation.
●One died from complications of central venous catheter insertion. Of the remaining 18, eight died a median of nine years following diagnosis, five from cardiovascular disease or stroke that may have been related to the TMA.
Exposure to quinine requires specific questioning because it is often omitted from the routine medical history. Quinine tablets may be borrowed from a friend or family member, or the exposure may occur from a beverage (eg, tonic water, bitter lemon) (table 2) [4,6,20,21]. In the United States, the only available quinine tablet (Qualaquin) requires a prescription, and the only approved indication is for malaria treatment. This restricted availability of quinine tablets may explain why we have not seen a patient with quinine-induced TMA since 2009 [19]. However, some individuals may be using the antimalarial drug for leg cramps despite a Boxed Warning regarding TMA as a potential complication [22]. There are also several over-the-counter tablets and herbal remedies for leg cramps available in the United States that may contain quinine, and quinine tablets can be purchased over-the-counter in Canada and other countries. Quinine may also be added to drugs of abuse such as cocaine.
Importantly, TMA from quinine can be triggered either by a single ingestion (eg, one quinine tablet, one quinine-containing beverage) occurring many months or years after a previous exposure, up to 10 years in our experience. This is because the drug-dependent antibodies can persist for many years, but they cannot react with target cells in the absence of the drug (figure 1). Acute immune-mediated tissue damage can occur within hours of re-exposure. It is not known whether the homeopathic doses of quinine present in remedies for leg cramps in the United States can trigger TMA, but in principle, immune-mediated DITMA can occur with extremely low levels of re-exposure.
Chronic kidney disease is common following quinine-induced TMA [19]. (See 'Clinical features of immune DITMA' below.)
Antimicrobials — The antibiotic with the best evidence is trimethoprim-sulfamethoxazole. A number of other antibiotics have evidence for causing immune-mediated DITMA in single case reports, including sulfisoxazole, ciprofloxacin, penicillin, and metronidazole, as well as the antimalarial drug mefloquine and the antiviral agent famciclovir [1].
Gemcitabine — Although several anticancer therapies, including gemcitabine, have been reported to cause non-immune DITMA (see 'Cancer therapies' below), there is also a report of immune-mediated DITMA from gemcitabine [23]. Immune-mediated DITMA from gemcitabine may be less common than the non-immune mechanism.
Oxaliplatin — One report has described definite evidence for DITMA caused by oxaliplatin with the clinical features of an immune-mediated etiology (although the authors of this report attributed the TMA to gemcitabine) [24]. The laboratory of the Versiti Blood Center of Wisconsin (Milwaukee) has also identified oxaliplatin-dependent, platelet-reactive antibodies in a patient with clinical features of TMA [2].
Quetiapine — The antipsychotic agent quetiapine has been associated with DITMA that has recurred with repeated exposures, suggestive of an immune mechanism [25].
Immunomodulatory agents — The anti-T cell monoclonal antibody muromonab-CD3 (OKT3) and the anti-tumor necrosis factor (TNF) monoclonal antibody adalimumab have been reported to be associated with immune-mediated DITMA [26,27]. Other immunosuppressive agents, including calcineurin inhibitors, appear to cause a toxicity-mediated DITMA. (See 'Immunosuppressive agents' below.)
Drugs (non-immune mechanism) — The three major categories of drugs that have been associated with non-immune DITMA are cancer therapies, including cytotoxic agents and vascular endothelial growth factor (VEGF) inhibitors; immunosuppressive agents, especially calcineurin inhibitors; and drugs of abuse. The hemophilia A therapy emicizumab has also been implicated.
Cancer therapies — DITMA following exposure to chemotherapeutic agents is postulated to result from dose-dependent toxicity, which may be either from a cumulative dose over weeks to months or may occur acutely from an unusually high individual dose. When DITMA occurs from a cumulative dose, the clinical course is characterized by the gradual development of kidney injury (either acute kidney injury or chronic kidney disease), sometimes even after the chemotherapy has been stopped. When it occurs because of an unusually high dose, the onset is sudden, similar to patients with immune DITMA. (See 'Clinical features of immune DITMA' below.)
●Gemcitabine – The many reports (over 100 patients have been reported [1,16]) of gemcitabine-induced DITMA suggest it is one of the more common chemotherapeutic agents responsible for DITMA; however, only 17 of these reports described definite or probable evidence for a causal association with TMA. The mechanism appears to be dose related in most patients. One series of 264 patients who received gemcitabine reported development of a TMA in three (1 percent) [28]. One patient had a relatively large cumulative dose, while the other two did not. There is also a report of immune-mediated DITMA from gemcitabine [23]. (See 'Drugs (immune mechanism)' above.)
●Mitomycin – Mitomycin has been the subject of many case reports (71 [1]) as a possible cause of DITMA. Only six reports have presented definite or probable evidence for mitomycin as the cause because mitomycin is almost always given in combination with other cancer therapies [29-37].
●Pentostatin – Two reports have described definite evidence supporting a causal association of pentostatin with DITMA [38,39].
●Vincristine – One report has described definite evidence supporting a causal association of vincristine with DITMA [40].
●Proteasome inhibitors – Several cases of DITMA have been described with proteasome inhibitors (bortezomib, carfilzomib, and ixazomib) [16,41,42]. Some of these had definite evidence for an association based on recurrence of TMA following re-exposure to the drug.
●Vascular endothelial growth factor (VEGF) inhibitors and tyrosine kinase inhibitors – TMA caused by VEGF inhibitors has been well described. Implicated agents include the antiangiogenic monoclonal antibody bevacizumab and the small molecule tyrosine kinase inhibitor sunitinib [11,43-46]. The small molecule BCR-ABL tyrosine kinase inhibitor ponatinib has also been implicated [47,48]. (See "Overview of angiogenesis inhibitors" and "Toxicity of molecularly targeted antiangiogenic agents: Cardiovascular effects".)
●Hematopoietic cell transplantation – Patients developing a TMA after hematopoietic cell transplant (HCT; either allogeneic or autologous) have been described; the reported incidence varies from 1 to 20 percent [49,50]. DITMA may be the etiology of TMA in these patients because they have typically received multiple therapies that could be responsible, including cytotoxic chemotherapy and/or immunosuppressive therapy. (See "Early complications of hematopoietic cell transplantation", section on 'Thrombotic microangiopathy' and "Kidney disease following hematopoietic cell transplantation", section on 'Thrombotic microangiopathy'.)
Immunosuppressive agents — Calcineurin inhibitors such as cyclosporine A (CSA) and tacrolimus cause dose-dependent endothelial dysfunction and increased platelet aggregation, possibly by inhibiting prostacyclins [51]. Organ involvement typically is restricted to the kidneys [52].
Sirolimus, an immunosuppressive agent that inhibits the mechanistic target of rapamycin (mTOR), can also cause DITMA, alone or in combination with CSA [53]. (See "Pharmacology of mammalian (mechanistic) target of rapamycin (mTOR) inhibitors", section on 'Hemolytic uremic syndrome/thrombotic microangiopathy'.)
In contrast, DITMA from the monoclonal antibody OKT3 and from the immunomodulatory agent adalimumab appears to be immune mediated. (See 'Immunomodulatory agents' above.)
Interferon alfa and interferon beta are type 1 interferons used to treat a variety of neoplastic and immunologic disorders including multiple sclerosis (MS). A causative role for interferon was suggested in a series of eight patients with MS who developed TMA during treatment with interferon beta, as there did not appear to be other causes such as other implicated drugs, severe ADAMTS13 deficiency, or evidence of complement dysregulation [12,54]. DITMA occurred exclusively in those who received an interferon dose above 50 mcg per week, and the effect appeared to be dose dependent. Subsequent studies in transgenic mice engineered to produce type 1 interferon in the brain demonstrated that interferon caused dose-dependent TMA [12].
Drugs of abuse — DITMA has been reported with certain drugs of abuse, including the following [13,55-57]:
●Inappropriate intravenous use of the extended-release opioid oxymorphone (Opana ER), which is intended for oral administration
●Inappropriate intravenous use of the formulation of oxycodone (OxyContin) that contains polyethylene oxide (PEO)
●Ingestion of ecstasy (MDMA, also known as Molly)
●Recreational use of cocaine
The mechanisms are not necessarily dose dependent. It is often not known whether the culprit is the known drug or another agent or substance that has been added to it. DITMA caused by intravenous use of Opana ER and hydrocodone has been documented to result from the high molecular-weight PEO (an inert ingredient intended to make the pills tamper-resistant) [13].
Emicizumab — Emicizumab is a monoclonal antibody that simultaneously binds activated factor IX (factor IXa) and factor X, essentially substituting for the role of factor VIII in hemostasis. It can be used for prophylaxis against bleeding in individuals with hemophilia A (factor VIII deficiency). (See "Hemophilia A and B: Routine management including prophylaxis", section on 'Emicizumab for hemophilia A'.)
Early experience identified a TMA in three individuals with hemophilia A who had factor VIII inhibitors and were treated with emicizumab along with high doses of activated prothrombin complex concentrate (aPCC; factor eight inhibitor bypassing agent [FEIBA]) [58]. This led to a Boxed Warning in the product information and advice to avoid combining high doses of emicizumab with an aPCC. Discontinuation of the aPCC led to resolution of the TMA in the three affected individuals.
Gene therapy (mechanism unclear) — A limited number of case reports has emerged in patients treated with certain gene therapy constructs.
●SMA – The gene therapy construct onasemnogene abeparvovec provides a normal copy of the SMN1 gene to patients with spinal muscular atrophy (SMA). The construct uses an adeno-associated virus (AAV) vector with tropism to the liver and other tissues [59]. (See "Spinal muscular atrophy", section on 'Onasemnogene abeparvovec'.)
Several cases of TMA have been reported, including TMA with fatal complications, in infants and young children treated with onasemnogene abeparvovec [60,61]. The time course is rapid (within one week to 10 days). The clinical presentations included severe thrombocytopenia, microangiopathic hemolysis with schistocytes on the blood film, transaminase elevations, and acute kidney injury (AKI) progressing to kidney failure. The role of complement abnormalities is unclear; one individual had a variant of uncertain significance (VUS) in the gene for complement factor I [60]. Some had concomitant infections.
●Other conditions – TMA may have occurred with other gene therapy constructs, but details are limited. As examples, thrombocytopenia with complement activation was seen in some individuals treated with an investigational gene therapy construct for degenerative muscular dystrophy (DMD), and one case of TMA was seen following treatment with an investigational gene therapy construct for Danon disease, although specific findings have not been published [62,63].
Possible mechanisms of TMA may include the gene therapy construct itself; inflammation induced by the gene therapy, underlying condition, or other interventions; complement dysregulation; other genetic variants; or other aspects of the patient's underlying disorder. Further research is needed. (See "Pathophysiology of TTP and other primary thrombotic microangiopathies (TMAs)", section on 'Complement-mediated TMA pathogenesis'.)
EPIDEMIOLOGY — DITMA is uncommon; in many patients, a drug association may be unrecognized. In a review of 487 patients in the Oklahoma Thrombotic Thrombocytopenic Purpura-Hemolytic Uremic Syndrome (TTP-HUS) Registry that included all patients with microangiopathic hemolytic anemia (MAHA) and thrombocytopenia referred for plasma exchange, DITMA with definite or probable evidence supporting a causal association with the suspected drug accounted for 23 cases (5 percent) [2].
DITMA is mostly seen in adults although, rarely, children may be affected. This distribution probably reflects the greater likelihood that adults are exposed to implicated drugs than children.
CLINICAL MANIFESTATIONS — The spectrum of disease severity in DITMA is large. However, all patients have microangiopathic hemolytic anemia (MAHA) and thrombocytopenia, and most have kidney injury. There is clinical overlap between DITMA and other primary TMAs, and clinical features that clearly distinguish DITMA from other primary TMAs are lacking.
Clinical features of immune DITMA — Patients with immune-mediated DITMA may have a history of daily exposure to the implicated drug for less than two to three weeks; a longer duration of daily drug exposure makes an immune-mediated DITMA much less likely. Alternatively, individuals with immune-mediated DITMA may have had intermittent exposure for years without any apparent illness or a history of nonspecific illness with previous exposures that was not attributed to the drug at the time they occurred [6,19].
The classic clinical features of immune-mediated DITMA are illustrated by the experience with quinine-induced DITMA [6,19]. The onset is sudden with severe systemic symptoms including chills, fever, abdominal pain, diarrhea, and/or nausea/vomiting, beginning within hours after quinine exposure. Patients can often recall the exact time and place when these symptoms began. Anuric acute kidney injury (AKI) occurs that is often misinterpreted as a manifestation of dehydration. Often, the onset of anuria is abrupt (within hours). Neurologic findings may range from mild confusion to coma. Acute gastrointestinal symptoms may manifest as nausea, vomiting, diarrhea, or abdominal pain.
In addition to the characteristic clinical features of TMA, patients with quinine-induced DITMA may also have additional organ involvement [64]. Patients may have chills, fever, hypotension, and/or disseminated intravascular coagulation (DIC) mimicking sepsis [65-67]. They may also have rhabdomyolysis or acute liver injury [65,68]. Many organ systems may be involved in a single patient (eg, fever, chills, hypotension, DIC, abnormal liver function, rhabdomyolysis, and cardiac ischemia) [69].
Clinical features of non-immune DITMA — DITMA caused by a dose-dependent mechanism, such as that due to cancer therapy or calcineurin inhibitors, may develop gradually over weeks to months with weakness, fatigue, symptoms of hypertension such as headache, and/or renal failure [70]. For chemotherapeutic agents, the findings may occur after several cycles of chemotherapy; some patients have a systemic malignancy, whereas others have a small tumor burden or no sign of disease [71-74]. There may be weeks to months of progressive kidney injury that may be attributed to other comorbidities. Neurologic findings may range from mild confusion to coma.
In contrast, DITMA associated with cocaine or intravenous use of extended-release oxymorphone (Opana ER) may present with sudden onset MAHA, thrombocytopenia, and AKI [10,13-15,55,75].
Laboratory/pathologic findings — As with other TMAs, DITMA is characterized in all patients by microangiopathic hemolysis, with schistocytes (fragmented red blood cells [RBCs] (picture 1)) on the peripheral blood smear, anemia, a negative direct antiglobulin test (DAT; Coombs test), and increased lactate dehydrogenase (LDH). Thrombocytopenia may be mild to severe. It is also possible for the patient to have a platelet count within the normal range if the count represents a significant decrease from the patient's baseline (eg, decline from 350,000/microL to 150,000/microL); however, thrombocytopenia is generally seen. Findings of organ damage, especially renal insufficiency with increased serum creatinine, proteinuria, and a bland urine sediment, are also common.
Some patients with quinine-induced DITMA may have severe neutropenia, a positive DAT, and/or DIC [3,4,65-67].
ADAMTS13 activity is normal or only mildly decreased (ie, activity is not <10 percent) in DITMA in contrast to thrombotic thrombocytopenic purpura (TTP), which is characterized by severe ADAMTS13 deficiency (activity <10 percent in almost all patients). ADAMTS13 activity <10 percent is strong evidence against the diagnosis of DITMA. Evidence of severe complement dysregulation is also absent, but nonspecific findings such as low C3 or C4 may be seen. Stool studies in DITMA are negative for Shiga toxin-producing Escherichia coli or other diarrheal organisms.
Drug-dependent antibodies may be identified in immune-mediated DITMA and establish the drug-induced etiology and the rationale for avoidance of the drug for life. However, the absence of drug-dependent antibodies cannot be used to eliminate the possibility of DITMA, because in vitro testing for drug-dependent antibodies may be negative if the antibodies are directed at a drug metabolite or if the drug is not soluble in the in vitro assay. Practical aspects of obtaining this testing are discussed below. (See 'Laboratory testing' below.)
Kidney biopsy is not required in the evaluation of DITMA, nor is it helpful in confirming the diagnosis or eliminating other causes of TMA. However, a kidney biopsy may appropriately be performed if the etiology of the AKI is not apparent (eg, possible acute tubular necrosis [ATN] in the setting of hypotension). Kidney biopsy may be especially useful in individuals with heavy (nephrotic range) proteinuria, who may have nephrotic syndrome due to another cause (eg, paraneoplastic membranous nephropathy or minimal change disease in a patient with cancer). If biopsy of an affected organ is performed, it is likely to show typical findings of TMA such as platelet-rich thrombi in small arterioles and capillaries, which are described in detail separately. (See "Pathophysiology of TTP and other primary thrombotic microangiopathies (TMAs)", section on 'Histopathology of TMA'.)
DIAGNOSTIC EVALUATION
When to suspect DITMA — DITMA may be suspected in any individual who presents with an unexplained decrease in platelet count and schistocytes on the blood smear, typically with worsening renal function. Most individuals have thrombocytopenia and hemolytic anemia and are receiving one of the implicated drugs. However, there are some exceptions (settings in which the diagnosis may be considered even in the absence of these findings) such as the following:
●As noted above, most patients have relatively severe thrombocytopenia, but even a normal platelet count that represents a large decrease from the patient's baseline may indicate evolving DITMA.
●Likewise, most patients have anemia, but evidence of microangiopathic changes without anemia may occur, especially if there is a brisk reticulocyte response.
●Increasing serum creatinine is typical, but some individuals may have a low baseline (especially if their nutritional status is tenuous, which may be the case in some individuals receiving chemotherapy drugs or immunosuppressive agents).
●Certain drugs are obvious, such as a chemotherapy drug or immunosuppressive agent; however, some individuals may not volunteer the use of an over-the-counter substance such as quinine, which is a common cause of DITMA. Direct questioning about quinine (table 2) may be required, as noted below. (See 'History and physical examination' below.)
History and physical examination — The patient history should include explicit questioning about exposures to drugs associated with DITMA and should focus on drugs that have been reported with definite or probable evidence for a causal association (table 1). It is also important to determine the duration of the exposure and whether use of the drug occurred on a daily basis or intermittently.
Especially important are over-the-counter remedies and illicit drug use, which are not considered medications and are generally not reported by patients unless specifically queried.
●We also ask specifically about leg cramps because a history of leg cramps raises our suspicion for quinine use. Sources of quinine include certain antimalarial tablets, various over-the-counter remedies for leg cramps, and a number of beverages that contain tonic water (table 2).
●Implicated recreational drugs include ecstasy (MDMA, also known as Molly), cocaine, and opioid medications used by an inappropriate route (eg, intravenous administration of extended-release oxymorphone [Opana ER] or oxycodone [OxyContin], which are intended to be taken orally).
The remainder of the history and physical examination is similar to other suspected primary TMAs. Importantly, a high level of suspicion for other potential systemic causes of microangiopathic hemolytic anemia (MAHA) and thrombocytopenia and other primary TMAs should be maintained. (See "Diagnostic approach to suspected TTP, HUS, or other thrombotic microangiopathy (TMA)", section on 'Initial evaluation (all patients)'.)
Laboratory testing — Laboratory testing in suspected DITMA includes the testing done for all patients with a suspected primary TMA. Some of the testing is used to find alternative causes of MAHA and thrombocytopenia. The following is appropriate, if not done already:
●Complete blood count (CBC) with platelet count and examination of a peripheral blood smear, to assess the degree of MAHA and thrombocytopenia
●Serum metabolic profile and creatinine, to assess the degree of kidney injury
●Lactate dehydrogenase (LDH), to assess the degree of hemolysis
●Coombs testing (also called direct antiglobulin testing [DAT]), to eliminate the possibility of autoimmune or drug-induced immune hemolysis (or diagnose it if positive)
●Coagulation testing, to assess for disseminated intravascular coagulation (DIC)
●ADAMTS13 activity level, to diagnose thrombotic thrombocytopenic purpura (TTP) if severely deficient
●Urinalysis, to evaluate for other causes of kidney injury (table 3) (See "Urinalysis in the diagnosis of kidney disease".)
Importantly, as noted above, a finding of severe ADAMTS13 deficiency (activity <10 percent) at any time during the evaluation for DITMA is strong evidence for the diagnosis of TTP, which has implications for management (urgent therapeutic plasma exchange [TPE] and other therapies). ADAMTS13 activity ≥10 percent is consistent with DITMA or any of the other non-TTP TMAs. (See "Diagnostic approach to suspected TTP, HUS, or other thrombotic microangiopathy (TMA)", section on 'Immediate management decisions'.)
Patients with acute kidney injury (AKI) that is progressive or requires dialysis and who are not receiving a drug implicated in DITMA may have a complement-mediated TMA, which requires anti-complement therapy. Details of the evaluation and management of complement-mediated TMA are presented separately. (See "Complement-mediated hemolytic uremic syndrome in children".)
Patients with suspected immune-mediated DITMA based on the history and clinical presentation may benefit from testing for drug-dependent antibodies, especially in cases where it is necessary to determine which of several potential drugs is responsible or if knowledge of the need to avoid a specific drug or substance for life would be valuable to the patient. This testing is available at the Versiti Blood Center of Wisconsin [76]. Typical turnaround time is approximately one week.
The finding of drug-dependent antibodies necessitates complete avoidance of the implicated drug or substance for life; however, the converse is not true (ie, lack of identification of drug-dependent antibodies cannot be used as evidence that a drug is safe). Thus, decisions regarding future drug avoidance must be made on a case-by-case basis. (See 'Future drug avoidance' below.)
Identification of drug-dependent antibodies associated with a drug not previously reported to cause DITMA is valuable both for management of the individual patient as well as making other clinicians aware of the potential association (eg, by description in a publication or reporting to the US Food and Drug Administration MedWatch program in the United States) [77].
Diagnosis — The diagnosis of DITMA is made clinically, based on the findings of MAHA and thrombocytopenia with the appropriate history of exposure to a drug previously documented to be associated with TMA (table 1). Our confidence in the diagnosis is increased when the drug has previously been reported to be associated with DITMA based on definite or probable evidence for a causal relationship (eg, presence of drug-dependent antibodies, no other drug exposures, appropriate temporal relationship) [1,2].
There is no specific diagnostic test for DITMA. A finding of drug-dependent antibodies is supportive if positive but not required. Absence of severe ADAMTS13 deficiency is also consistent with DITMA but not diagnostic, as this is seen with other primary, non-TTP TMAs. A kidney biopsy may show findings of a TMA if done, but is not required to make a presumptive diagnosis and cannot differentiate DITMA from other causes of TMA. (See 'Laboratory testing' above.)
DIFFERENTIAL DIAGNOSIS — The differential diagnosis of DITMA includes other causes of thrombotic microangiopathy (TMA) and other drug-induced causes of anemia, thrombocytopenia, and/or renal insufficiency.
●Other primary TMAs – Other primary TMAs include thrombotic thrombocytopenic purpura (TTP), complement-mediated TMA, Shiga toxin-mediated hemolytic uremic syndrome (ST-HUS), and TMA due to inherited defects in genes involved in regulating hemostasis or cobalamin metabolism. Like DITMA, these conditions can present with an acute, life-threatening illness associated with microangiopathic/fragmentation hemolysis, thrombocytopenia, and organ damage, especially kidney failure. Unlike DITMA, these other TMAs do not have a clear association with an implicated drug such as quinine, chemotherapeutic agents, calcineurin inhibitors, or drugs of abuse. Unlike DITMA, these other TMAs often reveal a specific diagnosis from appropriate laboratory testing, such as severe ADAMTS13 deficiency or a Shiga toxin-producing diarrheal organism. Other distinguishing factors are discussed in more detail separately. (See "Diagnostic approach to suspected TTP, HUS, or other thrombotic microangiopathy (TMA)".)
●Other drug-induced thrombocytopenias – Other drug-induced thrombocytopenias include drug-induced immune thrombocytopenia (DITP) and drug-induced bone marrow suppression of thrombopoiesis. Like DITMA, a temporal relationship with a drug may be elicited. Unlike DITMA, other forms of drug-induced thrombocytopenia are not associated with microangiopathic hemolysis with schistocytes on the peripheral blood smear or signs of organ damage from microvascular thrombosis. (See "Drug-induced immune thrombocytopenia".)
●Other drug-induced anemias – Other drug-induced anemias include anemia caused by a variety of immune and nonimmune mechanisms, including immune-mediated hemolysis and enzymatic defects such as glucose-6-phosphate dehydrogenase (G6PD) deficiency. Like DITMA, other drug-induced causes of hemolytic anemia are associated with a temporal relationship to drug exposure. Like DITMA, the Coombs test may be negative in some other drug-induced anemias. Unlike DITMA, patients with other drug-induced anemias typically do not have red blood cell fragmentation, thrombocytopenia, renal insufficiency, or other end-organ manifestations. Potential causes of drug-induced anemias are presented separately. (See "Drug-induced hemolytic anemia".)
●Other drug-induced kidney injury – Many drugs are potentially nephrotoxic, and the possibility of direct kidney injury rather than injury mediated by the vascular lesion of TMA must be considered. Like DITMA, drug-induced acute kidney injury (AKI) is associated with a temporal relation to drug exposure, rising serum creatinine, and in some cases anuria. Unlike DITMA, patients with drug-induced AKI do not have microangiopathic/fragmentation hemolysis or thrombocytopenia. Potential causes of drug-induced AKI and acute interstitial nephritis are presented separately. (See "Etiology and diagnosis of prerenal disease and acute tubular necrosis in acute kidney injury in adults", section on 'Nephrotoxins' and "Clinical manifestations and diagnosis of acute interstitial nephritis", section on 'Drugs'.)
MANAGEMENT
Initial interventions — Management of DITMA involves drug discontinuation and supportive care. Thus, when our confidence in the diagnosis of DITMA is high, we do not use therapeutic plasma exchange (TPE) or anti-complement therapy. However, TPE may be appropriate when there is uncertainty about the diagnosis of DITMA versus thrombotic thrombocytopenic purpura (TTP), and anti-complement therapy may be appropriate in patients with rapidly progressing kidney failure and uncertainty about the diagnosis of DITMA versus complement-mediated TMA (see 'Complement inhibition' below). Early involvement of the consulting specialist is advised because there is a need to balance the risks and benefits of using or not using these other therapies, both of which have their own associated risks.
Our avoidance of TPE when there is a high level of confidence in the diagnosis of DITMA is supported by a lack of high-quality evidence for a benefit of TPE in DITMA and is consistent with recommendations from the American Society for Apheresis (ASFA), which publishes an evidence-based categorization of the usefulness of TPE on review of available evidence and considers quinine and gemcitabine to be category IV (TPE ineffective or harmful) [78]. The ASFA guideline considers TPE to be effective for ticlopidine-associated TMA; however, we consider these patients to have had acquired, autoimmune TTP rather than DITMA. (See "Therapeutic apheresis (plasma exchange or cytapheresis): Indications and technology", section on 'ASFA therapeutic categories'.)
For patients in whom TPE or anti-complement therapy has already been initiated, we base our decisions regarding continuation or discontinuation of therapy on our level of confidence that the diagnosis is DITMA and the patient's clinical course. As with other TMAs, these decisions can be re-evaluated when additional clinical information becomes available, including the response to therapy, identification of alternative causes of the patient's symptoms, and results of other diagnostic testing. However, it is important to note that improvement during TPE or anti-complement therapy may be coincidental and does not necessarily imply a treatment response. (See "Immune TTP: Initial treatment", section on 'Continuation and completion of therapy'.)
Immunosuppressive therapy also is not a component of DITMA management, although it may be appropriate if the diagnosis is unclear (eg, glucocorticoids for patients in whom the diagnosis of TTP is suspected).
Various levels of supportive care may be needed in the management of patients as they recover from DITMA. It is appropriate to transfuse platelets for severe thrombocytopenia associated with clinically important bleeding (see "Platelet transfusion: Indications, ordering, and associated risks", section on 'Actively bleeding patient' and "Platelet transfusion: Indications, ordering, and associated risks", section on 'TTP or HIT'). Indications for dialysis and avoidance of nephrotoxic drugs are similar as with other patients with renal failure. (See "Kidney replacement therapy (dialysis) in acute kidney injury in adults: Indications, timing, and dialysis dose".)
Dose reductions of nephrotoxic chemotherapeutic agents may be required for patients with residual kidney disease and reduced glomerular filtration rate. (See "Chemotherapy nephrotoxicity and dose modification in patients with kidney impairment: Conventional cytotoxic agents" and "Chemotherapy nephrotoxicity and dose modification in patients with kidney impairment: Molecularly targeted agents and immunotherapies".)
Investigational therapies for DITMA that does not resolve with drug discontinuation
Complement inhibition — There have been reports of patients with non-immune DITMA attributed to gemcitabine, mitomycin, interferon, doxorubicin, and other drugs; and case reports have described patients with acute kidney injury attributed to gemcitabine who improved after treatment targeting complement activation [79,80]. However, these case reports do not provide confidence that anti-complement therapy is appropriate. Often the patients have received multiple chemotherapeutic agents, and the selection of gemcitabine as the possible etiology is arbitrary (eg, based on previous case reports).
Although these preliminary observations do not provide confidence that anti-complement therapy is appropriate, we believe it is reasonable to consider the use of eculizumab in persistent non-immune DITMA that does not improve with supportive care and withdrawal of the offending agent. Similarly, we believe it is reasonable to consider eculizumab in patients with acute, severe immune-mediated DITMA who are at risk for the development of chronic kidney disease [6,81].
N-acetylcysteine — N-acetylcysteine (NAC) is a sulfhydryl-containing compound with potential roles in oxygen scavenging, endothelial cell relaxation, glutathione recycling, and reducing protein multimerization mediated by disulfide bonds. NAC is clinically available and is the standard treatment for acetaminophen overdose.
NAC is safe, inexpensive, and may be effective for some cases of DITMA. We would have a low threshold for trying NAC in a patient with DITMA that does not resolve with drug discontinuation alone.
Supporting evidence includes:
●NAC was shown to reduce the concentration of ultra-large von Willebrand factor (VWF) multimers in vitro and in an animal model of TTP, suggesting that it may be an effective treatment in individuals with acquired TTP by a similar mechanism (ie, it might reduce accumulation of ultra-large VWF multimers in the microvasculature) [82,83]. (See "Pathophysiology of TTP and other primary thrombotic microangiopathies (TMAs)", section on 'TTP pathogenesis'.)
●NAC was administered to a patient with ponatinib-associated DITMA affecting the coronary microvasculature [48]. The patient had dramatic resolution of abnormal electrocardiogram (ECG) findings following treatment with NAC (150 mg/kg loading dose followed by two doses of 50 mg/kg given 12 hours apart).
Future drug avoidance — Once a patient recovers from DITMA, the question of possible use of the implicated medication in the future may arise. The importance of future drug avoidance depends on whether the mechanism of DITMA is immune mediated or toxicity mediated.
●Immune-mediated – For immune-mediated DITMA, the implicated agent must be completely avoided for life. Subsequent exposures can result in severe symptoms that may be fatal, even with lower doses or levels of exposure. Strict avoidance of the implicated drug should be noted clearly in the medical record. Patients who have recovered from quinine-induced TMA must be warned that even the low concentrations of quinine in beverages such as tonic water can cause a recurrent, severe episode.
A patient should never be re-exposed to a drug potentially implicated in immune-mediated DITMA solely for the purpose of establishing the role of the drug, as this could be life threatening.
However, a patient with immune-mediated DITMA may be able to receive an alternative agent within the same drug class, since drug-dependent antibodies may be specific for the individual drug. As an example, antibodies that react with quinine may not react with quinidine, which is structurally similar. (See 'Immune-mediated mechanism' above.)
●Non-immune – For non-immune DITMA that appears to be dose related, re-exposure to the implicated drug may be possible in selected cases. The decision must balance the potential risks of recurrent DITMA, which may be negligible with lower doses of the medication, versus the potential benefits of drug administration. As examples, kidney injury with features of TMA may be seen with high serum levels of a calcineurin inhibitor but may not recur with lower doses; in contrast, it may be appropriate to avoid an implicated chemotherapy agent in the palliative setting when other alternatives are available. It may be possible to re-administer emicizumab if appropriate, as long as it is not coadministered with an activated prothrombin complex concentrate (aPCC).
Discussions of alternative options for immunosuppression following hematopoietic and solid organ transplantation are presented in detail separately. (See "Thrombotic microangiopathy after kidney transplantation" and "Kidney function and non-kidney solid organ transplantation" and "Cyclosporine and tacrolimus nephrotoxicity" and "Kidney disease following hematopoietic cell transplantation", section on 'Thrombotic microangiopathy'.)
PROGNOSIS/EXPECTED RECOVERY — Recovery from immune-mediated DITMA may be slow because kidney damage may resolve very gradually, even after ongoing kidney damage has ceased to occur. In immune-mediated DITMA, thrombocytopenia should begin to recover within several days, but recovery of kidney function may be very slow and incomplete. Chronic kidney disease, as defined by a creatinine clearance <60 mL/min, is common in patients with quinine-induced TMA [84]. Recovery from non-immune DITMA is also expected to be slow and may be incomplete.
SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Thrombotic microangiopathies (TTP, HUS, and related disorders)".)
SUMMARY AND RECOMMENDATIONS
●Classification – We divide drug-induced thrombotic microangiopathy (DITMA) into immune-mediated and nonimmune, which has implications for diagnosis and management. (See 'Terminology' above.)
●Mechanism – Immune-mediated DITMA can occur after exposure to small amounts of a substance via an antibody-dependent mechanism (figure 1). Nonimmune DITMA has multiple mechanisms and is often dose dependent. (See 'Pathophysiology' above.)
●Implicated drugs – Drugs most commonly associated with DITMA are listed in the table (table 1). (See 'Drugs associated with DITMA' above.)
•Immune – The most common cause is quinine (table 2). (See 'Drugs (immune mechanism)' above.)
•Nonimmune – Common causes include type 1 interferon, some cancer therapies (gemcitabine, bevacizumab, sunitinib, ponatinib, proteasome inhibitors), calcineurin inhibitors (cyclosporine, tacrolimus), and drugs of abuse (cocaine, intravenous use of oxycodone or extended-release oxymorphone). The hemophilia A drug emicizumab has been implicated when given in combination with an activated prothrombin complex concentrate (aPCC). (See 'Drugs (non-immune mechanism)' above.)
High-quality evidence implicating clopidogrel and ticlopidine is lacking. (See 'Overview of drugs and criteria' above.)
●Presentation
•Immune – Immune-mediated DITMA causes sudden onset of chills, fever, abdominal pain, diarrhea, nausea/vomiting, and/or hypotension, within hours after exposure to a drug (taken occasionally over years or new drug over two to three weeks). (See 'Clinical features of immune DITMA' above.)
•Nonimmune – Nonimmune DITMA causes gradual weakness, fatigue, headache, and/or kidney failure over weeks to months. DITMA from drugs of abuse may cause sudden-onset systemic symptoms and kidney injury. (See 'Clinical features of non-immune DITMA' above.)
●Laboratory – Microangiopathic hemolytic anemia (MAHA) and thrombocytopenia are universal; some degree of kidney injury is common regardless of the mechanism. Quinine-induced DITMA may also cause multiple abnormalities not characteristically associated with TMAs, including disseminated intravascular coagulation (DIC), rhabdomyolysis, and/or abnormal liver function. ADAMTS13 activity is normal or only mildly decreased. Drug-dependent antibodies may be identified in immune-mediated DITMA. (See 'Laboratory/pathologic findings' above.)
●Diagnosis – The diagnosis is made clinically and includes the exposure history. Laboratory testing includes blood smear review, hemolysis testing, kidney and liver function, and ADAMTS13 activity. (See 'Diagnostic evaluation' above.)
●Differential – The differential diagnosis includes other primary TMAs such as thrombotic thrombocytopenic purpura (TTP) and other causes of acute kidney injury. (See 'Differential diagnosis' above and "Early complications of hematopoietic cell transplantation", section on 'Thrombotic microangiopathy' and "Diagnostic approach to suspected TTP, HUS, or other thrombotic microangiopathy (TMA)" and "Kidney disease following hematopoietic cell transplantation", section on 'Thrombotic microangiopathy'.)
●Treatment – Management involves drug discontinuation and supportive care. Therapeutic plasma exchange (TPE) may be appropriate when there is uncertainty about possible TTP. Anti-complement therapy may be appropriate in patients with rapidly progressing kidney failure and possible complement-mediated TMA. Early involvement of the consulting specialist is advised. (See 'Initial interventions' above and 'Investigational therapies for DITMA that does not resolve with drug discontinuation' above.)
●Prevention – For immune-mediated DITMA, the implicated drug must be avoided for life. For nonimmune DITMA, re-exposure may be possible in selected cases. (See 'Future drug avoidance' above.)
●Recovery – Recovery from immune-mediated DITMA may be incomplete because kidney damage may be severe. Thrombocytopenia should begin to resolve within days. (See 'Prognosis/expected recovery' above.)
