Complete recovery of thrombotic microangiopathy with severe acute kidney injury due to intravenous use of modified release oxycodone

ABSTRACT

Since the introduction of the tamper-resistant oral formulation of modified-release oxycodone (OxyContin®), cases of thrombotic microangiopathy (TMA) have been reported following intravenous injection of the drug. Previous reports have suggested that withdrawal of the offending agent alone may be sufficient to reverse the condition; however, the associated acute kidney injury (AKI) in these cases was only mild to moderate in severity. We report the first case of OxyContin®-associated TMA presenting with anuric AKI requiring dialysis, followed by complete renal recovery after cessation of intravenous OxyContin® use. This case demonstrates that withdrawal of the causative agent may be sufficient to achieve resolution of TMA even in severe presentations. Accordingly, supportive management without the need for invasive and costly interventions may be appropriate in selected cases. Awareness of the association between intravenous OxyContin® use and TMA is essential for timely diagnosis, as identification often depends on direct inquiry regarding drug exposure.

CASE DESCRIPTION

A 35-year-old Aboriginal Australian woman with no significant past medical history or regular prescribed medications, presented with a 6-day history of malaise, myalgia, vomiting, and reduced urine output. She denied fever, diarrhoea, or other infective symptoms. At the onset of her illness, she reported a single episode of seizure-like activity, with no subsequent focal neurological symptoms. A comprehensive review of systems was otherwise unremarkable. She admitted to intravenous illicit drug use but was unable to identify the substance injected.

On presentation, she was alert and oriented, afebrile, and haemodynamically stable, with a blood pressure of 118/75 mmHg, pulse rate of 90 beats/min, respiratory rate of 18 breaths/min, and oxygen saturation of 99% on room air. Clinical assessment suggested hypovolaemia. There were no visible injection-site marks or peripheral stigmata of infective endocarditis. The remainder of the physical examination, including a detailed neurological assessment, was unremarkable. Her laboratory exam results are shown in Table 1.

Table 1:

The patient’s initial laboratory results.

Laboratory result	Value	Reference range
Haemoglobin	57 g/l (nadir)	115–160 g/l
Platelets	73 × 103/l (nadir)	140–400 × 103/l
Creatinine	1410 µmol/l	45–90 µmol/l
Urea	71.1 mmol/l	2.1–7.1 mmol/l
Bicarbonate	18 mmol/l	22–32 mmol/l
Lactate dehydrogenase (LDH)	1280 U/l	120–250 U/l
Sodium	130 mmol/l	135–145 mmol/l
Potassium	6.2 mmol/l	3.5–5.2 mmol/l
Phosphate	2.44 mmol/l	0.75–1.50 mmol/l
Total bilirubin	28 µmol/l	<20 µmol/l
Haptoglobin	0.15 g/l	0.36–1.95 g/l

Her white cell count and differential were within normal limits. Peripheral blood film demonstrated moderate numbers of schistocytes and microspherocytes, consistent with microangiopathic haemolysis. Fibrinogen was elevated at 6.0 g/l (reference range 1.7–4.5 g/l), while the remainder of the coagulation profile was normal. ADAMTS13 (a disintegrin and metalloproteinase with a thrombospondin type 1 motif, member 13) activity was pending during the initial days of admission.

Additional investigations revealed no evidence of an autoimmune, infectious, paraproteinaemic, or complement-mediated process (Table 2). Urine drug screening was positive for opiates but negative for methadone, buprenorphine, cocaine, amphetamines, cannabinoids, and benzodiazepines.

Table 2:

Summary of additional investigations.

Investigation category	Results
Autoimmune	Anti-nuclear Antibodies (ANA), Extractable Nuclear Antigen (ENA), Antineutrophil Cytoplasmic Antibody (ANCA), Anti-glomerular basement membrane (anti-GBM) antibodies, Anti-double stranded DNA (Anti-dsDNA) antibodies, lupus anticoagulant, anticardiolipin antibodies, anti-β2 glycoprotein antibodies all negative
Complement studies	C3 and C4 within normal limits
Infectious screening	Hepatitis B, hepatitis C, Human immunodeficiency virus (HIV), and Epstein–Barr virus (EBV) Immunoglobulin M (IgM) negative
Paraprotein screen	Serum protein electrophoresis and serum free light-chain ratio unremarkable
Other	Cryoglobulins and creatine kinase normal
Urine drug screen	Positive for opiates; negative for methadone, buprenorphine, cocaine, amphetamines, cannabinoids, and benzodiazepines. (The hospital panel did not include specific testing for oxycodone)

The random urine protein-to-creatinine ratio was 110 g/mol (reference range <15 g/mol). Urine microscopy demonstrated erythrocyte, leucocyte, and epithelial cell counts of 10 × 10⁶/l, 20 × 10⁶/l, and 10 × 10⁶/l, respectively, with no bacterial growth on culture. Non-contrast computed tomography (CT) scan imaging of the abdomen, pelvis, and brain revealed no significant abnormalities.

The patient was admitted with severe acute kidney injury (AKI) secondary to thrombotic microangiopathy (TMA) of uncertain aetiology. Thrombotic thrombocytopenic purpura (TTP) was considered given the reported seizure-like activity and an intermediate PLASMIC score of 5 points [1]. Within 24 h of presentation, she became anuric. She received two units of fresh frozen plasma and was commenced on therapeutic plasma exchange (TPE) and haemodialysis.

On subsequent review and direct questioning, she recalled crushing and intravenously injecting modified-release oxycodone (OxyContin®) with her partner shortly before the onset of symptoms. She reported a long history of this practice, typically injecting up to 120 mg at a time. Considering this information, intravenous OxyContin® use was considered the most likely trigger for her TMA. Therapeutic plasma exchange was discontinued after a single session. Subsequent ADAMTS13 activity returned normal at 94% (reference range 40%–130%).

She required a single 8-h session of continuous venovenous haemodiafiltration (CVVHDF) in the intensive care unit, followed by one session of intermittent haemodialysis after transfer to the ward. She had no further access to OxyContin® during her admission. Her urine output gradually improved, accompanied by a progressive decline in serum creatinine. Although kidney biopsy was considered, it was deferred because of the clear trajectory of renal recovery. Anti-complement therapy was also considered but deemed unnecessary.

Both her biochemical and haematological parameters improved steadily. At discharge, after a 12-day hospital stay, her serum creatinine had fallen to 209 μmol/l and lactate dehydrogenase (LDH) to 258 U/l. Her haemoglobin concentration was rising, and the platelet count had normalized. Although she did not attend renal outpatient follow-up, she re-presented to the emergency department 6 months later with back pain. At that time, both serum creatinine and LDH had returned to normal, measuring 54 μmol/l and 198 U/l, respectively, confirming complete renal and haematological recovery. Figure 1 illustrates the recovery trends in serum creatinine, LDH, haemoglobin, and platelet count over time.

Figure 1:

Trend in serum creatinine, LDH, haemoglobin, and platelet count over time. ULN (upper limit of normal), LLN (lower limit of normal).

METHODS

A PubMed search was conducted using the following search strategy: (‘thrombotic microangiopathy’[Title/Abstract] OR TMA[Title/Abstract] OR ‘drug-induced thrombotic microangiopathy’[Title/Abstract] OR drug-induced TMA (DITMA)[Title/Abstract] OR ‘haemolytic uremic syndrome’[Title/Abstract] OR ‘haemolytic uraemic syndrome’[Title/Abstract] OR HUS[Title/Abstract] OR TTP[Title/Abstract]) AND (oxycodone OR OxyContin OR ‘modified-release oxycodone’ OR ‘extended-release oxycodone’ OR opioid OR oxymorphone) AND (intravenous OR injection OR misuse OR abuse OR ‘intravenous drug use’). The search identified 40 articles, including three case reports of TMA associated with intravenous modified-release oxycodone use, systematic reviews of drug-induced TMA, reports of TMA associated with intravenous extended-release oxymorphone (Opana ER) use, studies investigating polyethylene oxide-induced haemolysis, and other articles that were not directly relevant to this case.

DISCUSSION

Our patient developed TMA following intravenous injection of OxyContin®, resulting in severe haematological and renal impairment that necessitated haemodialysis. Complete renal recovery was achieved following withdrawal of the offending agent.

A few cases of TMA associated with intravenous use of OxyContin® have previously been reported. In the first published case, a strong temporal relationship between drug exposure and the development of TMA was observed; however, a definitive causal association could not be established in the absence of drug rechallenge [2]. The patient had normal kidney function and was managed conservatively, with complete resolution of haematological abnormalities within 3 months. Another reported case posed a diagnostic challenge because the patient initially failed to disclose intravenous drug use. Although kidney function remained normal, therapeutic plasma exchange was commenced before TTP was biochemically excluded [3]. Haematological improvement was observed by the time of hospital discharge. However, the patient re-presented with recurrent TMA following re-exposure to the drug in the community, accompanied by mild renal impairment that resolved completely after cessation of the offending agent [3].

A similar pattern of recurrent TMA and modest renal impairment following re-exposure was seen in another case. Notably, TMA recurred despite ongoing treatment with the anti-complement agent eculizumab and improved only after identification and withdrawal of the trigger [4]. Collectively, these reports provide strong evidence supporting a causal association between intravenous administration of tamper-resistant modified-release oxycodone (OxyContin®) and TMA (Table 3).

Table 3:

Case reports on TMA associated with IV oxycodone use.

Citation	Setting	Diagnostic methods	Treatment	Outcome
Tate and Mollee	Australia; 56-year-old man with daily IV tamper-resistant extended-release oxycodone misuse.	Clinical diagnosis of TMA based on microangiopathic haemolytic anaemia (MAHA), thrombocytopenia, and exclusion of alternative causes including severe ADAMTS13 deficiency. Serum creatinine level was normal at 66 μmol/l.	Withdrawal of oxycodone and supportive care. No plasma exchange.	Complete haematological recovery after cessation of IV oxycodone.
Nataatmadja and Divi	Australia; 29-year-old woman with recurrent IV sustained-release oxycodone use.	Clinical TMA with MAHA, thrombocytopenia, and schistocytes; normal ADAMTS13 activity; relapse temporally associated with recurrent oxycodone injection. Peak serum creatinine was 118 mmol/l. Gene sequencing negative for known complement-mediated TMA -associated mutations.	Plasma exchange during the initial episode. Supportive care and cessation of IV oxycodone on representation.	Recovery after initial treatment; recurrence with renewed IV oxycodone use, followed by recovery after cessation.
Robson et al.	Australia; 35-year-old man with IV tamper-resistant extended-release oxycodone misuse.	Clinical TMA with MAHA, thrombocytopenia, and schistocytes, normal ADAMTS13 activity, and evidence of TMA on kidney biopsy. The serum creatinine was 242 μmol/l on representation.	Plasma exchange initially, Oral Prednisolone (weaned over 3 weeks) followed by eculizumab (ceased after 6 months). Substance abuse counselling.	Relapse occurred despite eculizumab when IV oxycodone use resumed. No evidence of recurrence on follow-up.

The tamper-resistant properties of OxyContin® are conferred by polyethylene oxide (PEO), a high-molecular-weight polymer incorporated into the tablet formulation [5]. PEO is also used in the modified-release oxymorphone formulation Opana® ER, which has similarly been linked to multiple cases of TMA [5, 6]. Further evidence supporting a pathogenic role for PEO comes from animal studies, in which intravenous administration of PEO induced reversible haemolytic anaemia and thrombocytopenia in rats in a dose-dependent manner [7]. By contrast, PEO is considered non-toxic when administered orally and has not been associated with TMA, as it is not absorbed through the gastrointestinal tract [8].

TMA may arise from a range of conditions, including TTP, Shiga toxin-mediated haemolytic uraemic syndrome, complement-mediated TMA (CM-TMA), metabolism-mediated TMA, coagulation-mediated TMA, and DITMA [9]. The consequences can be catastrophic, particularly in patients with TTP, in whom delayed diagnosis and treatment are associated with high mortality [10]. Accurate identification of the underlying aetiology is therefore essential, as management strategies differ substantially between TMA subtypes [11]. DITMA represents 10%–13% of all TMA cases and could result in severe complications and death without early recognition and withdrawal of the causative agent [12]. Current recommendations for DITMA include withdrawal of the offending agent and supportive care [13]. However, TPE and rituximab have been reported to be beneficial in selected cases of DITMA, particularly when an immune-mediated mechanism is suspected [12].

In our case, intravenous OxyContin® use was not initially apparent, and TTP could not be confidently excluded during the acute presentation while definitive investigations were pending. Given the potentially life-threatening nature of untreated TTP, initiation of therapy while awaiting confirmatory results was considered appropriate [13]. Furthermore, the patient’s PLASMIC score indicated an intermediate probability of TTP. Subsequent identification of intravenous OxyContin® use, together with the evolving clinical picture, provided additional support for early cessation of TPE.

This case highlights the importance of obtaining a detailed drug history, including explicit enquiry regarding use of medications known to be associated with TMA [13]. In our experience, awareness of the association between intravenous OxyContin® use and TMA remains limited. This may reflect the relative novelty of the tamper-resistant formulation and the fact that TMA occurs only in the setting of intravenous misuse rather than therapeutic oral administration [2]. Despite presenting with severe anuric acute kidney injury requiring haemodialysis, our patient experienced complete recovery of kidney function following cessation of the offending agent. Although fresh frozen plasma and a single session of TPE were administered, the temporal relationship between drug withdrawal and clinical improvement suggests that removal of the trigger was the principal determinant of recovery. This observation adds to the growing body of evidence that discontinuation of the offending agent, together with supportive care, may be sufficient for recovery in OxyContin®-associated TMA, even in patients with severe renal impairment.

Two principal mechanisms have been proposed for DITMA: immune-mediated and dose-dependent toxic injury [13, 14]. Immune-mediated DITMA typically occurs within 21 days of commencing daily drug exposure or within hours of intermittent re-exposure, whereas toxic DITMA results from exposure to a drug or substance with direct dose-related toxicity [14]. The pathogenesis of OxyContin®-associated TMA is more likely to involve the latter mechanism. Hunt et al. demonstrated dose-dependent haemolysis and thrombocytopenia following intravenous administration of PEO in animal models [15]. This effect was attributed to disruption of the cell-free plasma layer adjacent to the microvascular wall, resulting in increased shear stress and enhanced cleavage of von Willebrand factor. The observed dose-dependent relationship may partly explain why only a subset of individuals who inject OxyContin® develop TMA.

Although intravenous misuse of oral oxycodone preparations has declined since the introduction of tamper-resistant OxyContin®, the formulation remains susceptible to manipulation and carries a risk of severe TMA when injected intravenously [16, 17]. The development of safer tamper-resistant formulations that do not rely on PEO may therefore reduce the risk of this serious complication. The absence of a kidney biopsy is a limitation of this case, as it precludes histopathological confirmation of TMA and the exclusion of alternative diagnoses. However, the biochemical findings are consistent with a diagnosis of TMA. Additionally, although the urine toxicology screen was positive for opiates, it did not specifically assay for oxycodone. Standard urine opiate immunoassays have limited sensitivity for oxycodone [18]; therefore, a positive opiate screen alone cannot confirm oxycodone exposure. However, the patient consistently reported ingesting oxycodone, and the positive urine opiate screen supports recent opioid exposure. In the absence of any history or evidence of alternative opioid use, oxycodone remains the most likely opioid ingested.

In conclusion, our case demonstrates that OxyContin®-associated TMA can present with severe acute kidney injury requiring dialysis, and yet still be associated with excellent renal recovery following prompt identification and removal of the offending agent. Greater awareness of this association and a high index of suspicion are essential to facilitate timely diagnosis, avoid unnecessary prolonged invasive therapies, and optimize patient outcomes.

CONFLICT OF INTEREST STATEMENT

None to declare.

FUNDING

None declared.

PATIENT CONSENT

Written informed consent was given by the patient for their anonymized information to be written up and published.