Thrombotic microangiopathy (TMA) is a pathologic term used to describe small vessel
injury, manifesting clinically as microangiopathic anemia, thrombocytopenia, and target
organ damage including kidney injury. The classical TMA syndromes are ADAMTS13 deficiency–associated
thrombotic thrombocytopenia (TTP) and the Shiga toxin–mediated hemolytic uremic syndrome.
Other primary TMA syndromes include drug-induced TMA (DITMA), complement-mediated
TMA, and rare hereditary disorders of hemostasis and vitamin B12 metabolism. In addition
to the primary syndromes, various systemic conditions can manifest with TMA. These
include malignant hypertension, malignancies, pregnancy-associated conditions such
as pre-eclampsia and HELLP (hemolysis, elevated liver enzymes, low platelet count)
syndrome, infections, autoimmune diseases, and stem cell or solid organ transplants.
Patients may present with symptoms related to anemia, thrombocytopenia, or evidence
of end-organ involvement including neurological symptoms, skin rash, and renal failure.
A peripheral blood smear is an essential first step to establish evidence of microangiopathic
anemia. Once microangiopathic anemia and thrombocytopenia have been confirmed, a primary
systemic condition should be excluded. If no systemic condition is identified, a diagnosis
of primary TMA syndrome should be sought. DITMA are acquired from specific drug exposures
and result from dose-dependent drug toxicity or immune-mediated mechanisms. Quinine
is the most common cause of immune-mediated DITMA.
Here we report a case of TMA associated with sulfamethoxazole-trimethoprim (SMX-TMP)
in a previously healthy man.
A 62-year-old man was transferred to our hospital for acute renal failure in the setting
of new onset anemia, thrombocytopenia, and hyperkalemia. His past medical history
included prostate cancer treated with radical prostatectomy 3 years ago. He had been
in his usual state of health until 2 weeks before admission when he presented to a
community hospital with new rash on his left ankle. Initial evaluation revealed hemoglobin
was 12.8 g/dl, hematocrit was 40.3, white blood cell count was 25,500/mm3, platelet
count was 200,000/mm3, and creatinine was 1.0. Electrolytes were within normal limits,
and his C-reactive protein was elevated to 38.5. He was treated with cephalexin for
suspected cellulitis, but then switched to i.v. ampicillin/sulbactam for poor response
and persistent leukocytosis (77% neutrophils). Eventually, a skin biopsy of the left
ankle plaque was performed, and it revealed findings of epidermal necrosis, dermal
inflammation, and absence of vasculitis or neoplastic changes. This raised concern
for Sweet's syndrome. He was discharged on prednisone 80 mg daily and SMX-TMP for
Pneumocystis carinii prophylaxis. Additional workup at this time included a computed
tomography scan of his chest, abdomen, and pelvis, which did not reveal any evidence
of malignancy. A transthoracic echocardiogram revealed a moderate pericardial effusion
without any hemodynamic compromise. This was attributed to a possible viral etiology
and was to be followed up as an outpatient.
One week after discharge, he returned to the community hospital for malaise, chills,
nausea, vomiting, diarrhea, and decreased amount of “brown-colored” urine. Initial
evaluation revealed anemia, thrombocytopenia, acute renal failure, and hyperkalemia.
Laboratory values are detailed in Table 1. Given concern for TTP, he was transferred
to our hospital for potential plasmapheresis.
Laboratory evaluation 1 week before current hospital admission, at the time of admission,
and at hospital day 2 and day 15
1 wk prior
No monoclonal Ig
Free K/L ratio
0.5% (negative on repeat)
B2GP1 Ab (IgA/M/G)
Cardiolipin Ab (IgG/M)
Anaplasma phagocytophilum Ab
Hepatitis C Ab
HSV 1 Ab
HSV 2 Ab
ALT, alanine transaminase; ANA, anti-nuclear antibody; ANCA, antineutrophil cytoplasmic
antibody; AST, aspartate transaminase; BUN, blood urea nitrogen; CMV, cytomegalovirus;
CRP, C-reactive protein; EBV, Epstein-Barr virus; ESR, erythrocyte sedimentation rate;
HCV, hepatitis C virus; HSV, herpes simplex virus; INR, international normalized ratio;
K/L, kappa:lambda ratio; LDH, lactate dehydrogenase; MMA, methylmalonic acid; PT,
prothrombin time; PTT, partial thromboplastin time; SPEP, serum protein electrophoresis;
WBC, white blood cell.
On arrival, he was afebrile. His vital signs were blood pressure 80/54 mm Hg, heart
rate 110/min, respiratory rate 20/min, and oxygen saturation of 99% on room air. On
examination, he was anasarcic, alert, oriented, and conversant. Pupils were equally
round and reactive to light bilaterally; extraocular movements were intact. Neck was
supple without any lymphadenopathy, thyromegaly, or carotid bruit. Heart sounds S1,
S2 were regular without any murmurs or rubs. Lungs were clear to auscultation bilaterally.
Abdominal examination revealed minimal bowel sounds with a soft and diffusely tender
abdomen. No organomegaly was noted. Skin examination was notable for small 2–3 cm
necrotic ulcers on both feet, worse on the medial side of the left ankle. There was
no purulent or bloody drainage for these ulcers. Neurologically, his motor strength
was 5/5 in all 4 extremities, sensations to light touch and pain were intact, facial
expressions were symmetrical with normal neck movements, and shoulder shrugs. Speech
was fluent. Asterixis was noted. Initial labs are given in Table 1. Electrocardiogram
showed widened QRS. An urgent peripheral smear revealed schistocytes (Figure 1). SMX-TMP
and prednisone were stopped. Emergent hemodialysis was first performed, followed by
plasmapheresis. A broad infectious and autoimmune workup were obtained (Table 1).
ADAMTS13 activity was low at 31% but did not meet criteria for continuing plasmapheresis.
His platelet counts recovered completely by hospital day 2, although he continued
to have worsening leukocytosis, anemia, and dialysis-dependent anuric renal failure
(initially continuous veno-venous hemodiafiltration due to hypotension). Methylprednisolone
40 mg every 24 hours was restarted on day 2. On day 4, he underwent coronary angiography,
right heart catheterization, and endomyocardial biopsy for newly reduced ejection
fraction 30%–35% (compared with 55%–60% 2 weeks ago) and elevated troponin. On the
same day, he also underwent a bone marrow biopsy, for persistent leukocytosis, which
did not show any neoplastic process. The endomyocardial biopsy revealed mild cardiomyocyte
hypertrophy but otherwise no evidence of myocarditis or infiltration.
Peripheral blood smear. The yellow arrow is pointing at a normal-appearing red blood
cell and blue arrows point at schistocytes.
Hospital course was further complicated by intermittent need for intensive care unit
and continuous veno-venous hemodialysis for renal replacement due to persistent hypotension.
His hemodynamics gradually improved, and a repeat echocardiogram on day 8 revealed
improvement in left ventricular ejection fraction to 41% and reduction in the pericardial
effusion. Renal biopsy was performed on day 15 for prognostic purposes because the
patient remained anuric and dialysis dependent. The renal biopsy showed a diffuse
TMA pattern involving glomeruli and arterioles with associated extensive tubular and
focally interstitial necrosis (Figure 2). The patient was discharged to an acute rehabilitation
facility and continues to be dialysis-dependent 6 months after discharge. In the interim,
he has also been started on mycophenolate mofetil as a steroid-sparing agent for an
undifferentiated systemic inflammatory syndrome in lieu of persistently elevated C-reactive
Renal biopsy. (a) Arterioles contain fibrin thrombi (arrows) and intramural fragmented
red blood cells or schistocytes (arrow heads). Tubular injury is seen diffusely. Masson
trichrome stain, original magnification ×200. (b) Glomeruli contain fibrin thrombi
(arrowheads) and show mesangiolysis (green arrow). The glomerular basement membrane
shows diffuse ischemic changes. An arteriole contains an intraluminal fibrin thrombus
(red arrow). Jones silver stain, original magnification ×200. (c) Arterioles and glomeruli
contain abundant fibrin thrombi (red arrows). Direct immunofluorescence antifibrin,
original magnification ×200. (d) Glomerular basement membranes show marked subendothelial
lucency and flocculent material (asterisks), consistent with endothelial injury/thrombotic
microangiopathy pattern. Electron microscopy, original magnification ×10,000.
TMA is characterized by microangiopathic hemolytic anemia and thrombocytopenia. The
former is diagnosed by the presence of schistocytes on a peripheral blood smear and
markers of hemolysis including elevated lactate dehydrogenase, elevated indirect bilirubin,
and haptoglobin typically less than 10. The pathologic mechanisms include endothelial
cell injury, leading to formation of platelet microthrombi in the microvasculature,
which shear red blood corpuscles as they traverse narrowed capillaries. Glomerular
capillaries are particularly susceptible to involvement by TMAs. Once TMA is suspected,
it is important to rule out TTP, which results from an ADAMTS13 deficiency or acquired
inhibitory antibodies. In TTP, ADAMTS13 activity is severely reduced (<10%). Plasma
exchange should be initiated pending results to replenish ADAMTS13 and remove inhibitory
antibodies in addition to the large von-Willebrand factor multimers. Hemolytic uremic
syndrome is another primary TMA syndrome that is caused by Shiga toxin–producing organisms
such as Escherichia coli O157:H7, O104:H4, O111 and Shigella dysenteriae. Other primary
TMA syndromes include complement-mediated TMA, DITMA, and rare hereditary disorders
of coagulation (thrombomodulin, plasminogen, diacylglycerol kinase epsilon) and cobalamin
metabolism (methylmalonic aciduria and homocystinuria type C gene).
The fundamental role of the complement system in various TMA syndromes is now recognized,
and testing for complement pathway components and presence of inhibitors is available.
Treatment for some disorders with anticomplement agents such as eculizumab has led
to further progress.
DITMA is rare. As per the Oklahoma TTP/hemolytic uremic syndrome registry, of all
487 patients with suspected TTP/hemolytic uremic syndrome referred for plasma exchange,
only 23 (5%) had definite or probable evidence for a causal association with a candidate
Almost 90% of these cases were associated with quinine (Table 2). Mechanisms of DITMA
include immune mediated, which involves drug-dependent antibodies targeting host cells,
or direct dose-dependent drug toxicity (Table 3). The criteria to determine causality
of the drug-induced thrombocytopenia were described by George et al. in 1998.
A definite causal relationship requires that 4 criteria be met: (i) therapy with the
candidate drug preceded thrombocytopenia and recovery of platelet count was complete
and sustained after the drug was discontinued, (ii) the candidate drug was the only
drug used before thrombocytopenia or other drugs were continued or reintroduced after
discontinuation of the offending drug with a sustained normal platelet count, (iii)
other causes of thrombocytopenia were excluded, and (iv) re-exposure to the candidate
drug led to recurrent thrombocytopenia. Although this definition was evaluated for
drug-induced thrombocytopenia, it should hold true for immune-mediated DITMA because
pathogenesis involves drug-dependent antibodies attacking not just platelets but also
endothelial cells and possibly organ tissues directly. Although testing for specific
drug-dependent antibodies has been performed for research purposes, it does not change
clinical management. Management involves discontinuation of the offending drug and
supportive treatment. Renal recovery is expected but a case series reported chronic
renal failure in 57% patients with quinine-associated DITMA.
TMA is characterized by microangiopathic hemolytic anemia and thrombocytopenia.
Glomerular capillaries are often involved in TMAs.
DITMA is rare, and 90% is associated with quinine.
Mechanisms of DITMA include immune-mediated, which involves drug-dependent antibodies
targeting host cells, or direct dose-dependent drug toxicity.
Management involves discontinuation of the offending drug and supportive treatment.
Renal recovery is expected, but up to 57% patients can progress to chronic kidney
DITMA, drug-induced thrombotic microangiopathy; TMA, thrombotic microangiopathy.
Mechanisms of drug-induced thrombotic microangiopathy
Direct drug toxicity
Exposure to offending drug → formation of drug-dependent antibodies to platelets,
neutrophils, complement factors, endothelial cells → microvascular injury → thrombosis
and platelet consumption
Exposure to offending drug → direct endothelial cell injury → thrombosis and platelet
Acute onset anemia, thrombocytopenia, thrombosis, and acute kidney injury usually
triggered by first contact with the drug
Acute kidney injury and systemic features on initial or prolonged exposure to the
Quinine, oxaliplatin, quetiapine, and gemcitabine
Cyclosporine, tacrolimus, sirolimus, interferons, bevacizumab, gemcitabine, and mitomycin
In our patient, SMX-TMP met criteria 1 through 3, suggesting “probable” causality.
This drug did not need to be reintroduced in our patient, and therefore criterion
4 was not assessed. The mechanism of drug injury in our case appears to be immune
mediated. This is supported by the hypocomplementemia, rapid development of symptoms
(within days of exposure), progressive anemia, thrombocytopenia, anuric renal failure
within days of exposure, and the rapid recovery of platelets after discontinuation
of SMX-TMP. Alternatively, dose-mediated drug toxicity usually presents with subacute
renal failure in the setting of prolonged exposure to a candidate drug.
To our knowledge, only 3 case reports of SMX-TMP–associated TTP have been published
S1 Lichtin et al.’s
case series focused on plasmapheresis and did not include specific details about the
patient. Martin et al.’s
patient had normal ADAMTS13 activity, no renal dysfunction, and no tissue diagnosis.
Bapani et al.’sS1 report demonstrated SMX-TMP–associated TTP with ADAMTS13 activity
less than 5% but only met the first criteria for causality as described above. None
of these cases had tissue evidence of TMA, and none resulted in any long-term renal
disease. Our case is the only reported case of SMX-TMP–induced TMA with histopathologic
diagnosis and resultant end-stage renal disease. The exact reason for the lack of
renal recovery and persistently elevated inflammatory markers remains elusive. Some
authors suggest the role of the complement system in DITMA, with the drug exposure
serving as a “second-hit” over an underlying genetic defect.
Further studies are warranted to better understand the exact mechanisms of DITMA and
elucidate risk factors so that at-risk individuals can be identified before prescribing
potentially offending medications.
All the authors declared no competing interests.