NU648 Unit 3 Adverse Drug Event Analysis Example

Probable, Not Certain: An Adverse Drug Event Analysis of Falling Platelets During Linezolid Treatment for Enterococcal Bacteremia

[Student Name]

Purdue University Global

NU648: Specialized Pharmacology for the AGACNP

Unit 3 Adverse Drug Event Analysis

[Instructor Name]

[Date]

The patient is composite; counts, doses, levels and days are bracketed, and the causality score is shown item by item.

The Event

A [68]-year-old man with stage 3b chronic kidney disease, eGFR [35], was admitted with fever and found to have vancomycin-resistant Enterococcus faecium in [two of two] blood culture sets, with a presumed urinary source. Linezolid [600 mg] every 12 hours was started on day 0. Blood cultures cleared by day [3]. Venous thromboembolism prophylaxis was mechanical only, because of a gastrointestinal bleed [four months] earlier; he received no heparin in any form, and line flushes used saline.

Platelet counts, in thousands per microliter:

1. Day 0: [228].

2. Day 4: [214].

3. Day 7: [176].

4. Day 10: [109].

5. Day 13: [61].

On day 13, a peripheral smear showed no platelet clumping and no schistocytes. Linezolid was stopped and daptomycin started. Platelets recovered to [98] on day 16 and [142] on day 20.

Why Linezolid Is Suspected

Linezolid is known to cause reversible, duration-related myelosuppression, with thrombocytopenia the most common form. In pooled clinical trial data, the risk rose with treatment longer than about two weeks and reversed after the drug was stopped (Gerson et al., 2002). Two features of this patient increased his risk. His course was nearing two weeks. And his kidney function was reduced: although standard labeling does not adjust linezolid for kidney disease, patients with renal impairment accumulate the drug and its metabolites, and higher exposure is linked to thrombocytopenia, which has led to proposals for dose reduction in this group (Crass et al., 2019). A trough level drawn on day [11] was [9.4] mg/L, above the range commonly associated with toxicity.

The platelets fell steadily while he improved clinically, which is the opposite of what an infection-driven fall would usually look like.

Alternative Causes

Sepsis-related consumption: bacteremia can lower platelets through consumption and marrow suppression. Against it: his cultures cleared by day 3, fever resolved by day [4], and the steepest fall came after day 7, when he was clinically better. It cannot be excluded entirely, because a slow recovery from sepsis-related marrow suppression is possible.

Heparin-induced thrombocytopenia: excluded by the absence of any heparin exposure.

Pseudothrombocytopenia: excluded by a smear without clumping.

Other drugs: his other medications were [amlodipine], [atorvastatin] and [pantoprazole]. Proton pump inhibitors have been reported rarely in association with thrombocytopenia, but pantoprazole had been taken for years without change, so it is an unlikely cause.

Dilution and bleeding: no transfusion of large volumes and no bleeding.

Thrombotic microangiopathy: excluded by the absence of schistocytes and a normal hemoglobin trend.

Causality Score

The Naranjo algorithm assigns points to ten questions and groups totals into doubtful, possible, probable and definite (Naranjo et al., 1981).

1. Published reports of this reaction exist and are conclusive: yes, +1.

2. The fall followed the start of linezolid in time: yes, +2.

3. Counts improved after linezolid was withdrawn: yes, +1.

4. Linezolid was restarted to see whether the fall returned: not done, 0.

5. Another cause could explain the fall on its own: yes, sepsis cannot be ruled out, minus 1.

6. A placebo was given: not done, 0.

7. A blood level in the toxic range was documented: yes, trough [9.4] mg/L, +1.

8. The effect changed with a dose change: no dose change was made, 0.

9. A similar reaction occurred on earlier exposure: no known prior exposure, 0.

10. Objective evidence confirmed the event: yes, serial counts and a smear, +1.

Total: 5, which falls in the probable range of 5 to 8.

Why the Verdict Stops at Probable

A definite verdict needs a total of 9 or more. Two items keep this case below that line. A positive rechallenge, giving linezolid again and watching the platelets fall a second time, would add two points, and ruling out sepsis would turn a minus 1 into a plus 1. Only both together would reach 9. Neither is available or appropriate. Restarting linezolid to prove causation would expose him to a known harm for no clinical benefit, and the infection, though an unlikely explanation for this pattern, cannot be removed from the record after the fact. The analysis stops at probable because the missing points are ones no one should try to earn. It is not doubt about the likely cause; it is honesty about the limits of proof in a single patient.

Probable is still a strong finding for practice. It is enough to list the reaction in his allergy and intolerance record, to report it, and to change how the next prescriber approaches linezolid in this man. What it does not support is a claim that the drug alone explains every point on the curve, and the report makes no such claim.

Risk Factors in Context

Three features of this course matched what the literature links to linezolid thrombocytopenia. Duration was the first. His platelets were stable for the first week and fell mainly in the second, the pattern described in the pooled trial experience, where changes tended to appear after about two weeks of therapy and reversed after the drug was stopped (Gerson et al., 2002). Kidney function was the second. With an eGFR of [35], he was in the group in which exposure rises and in which lower doses have been proposed (Crass et al., 2019). The trough was the third, and it links the first two: a high level in a patient with reduced clearance is the mechanism the dosing literature describes.

A lower baseline count was not a factor here, since he started at [228,000]. That matters for the argument. A patient who began near [120,000] would have reached a dangerous level sooner, and the same drug course would have produced a bleeding risk rather than a laboratory finding. The report says so, because the next patient on this unit may start lower.

What Follows

The event was reported through the hospital's adverse drug event system, to FDA MedWatch as a serious event, and to the pharmacy and therapeutics committee, with a note on renal function and duration as risk factors. The record now lists linezolid-associated thrombocytopenia as a probable adverse reaction, so that future prescribers will consider an alternative or monitor counts twice weekly from the start. Two practice points emerge. Platelet counts should be checked at least weekly on linezolid, more often after the first week or in renal impairment. And when the course will exceed two weeks, an alternative agent should be considered at the outset. For this patient, daptomycin completed the course, and his platelets were rechecked at the follow-up visit rather than assumed to have recovered fully.

References

Crass, R. L., Cojutti, P. G., Pai, M. P., & Pea, F. (2019). Reappraisal of linezolid dosing in renal impairment to improve safety. Antimicrobial Agents and Chemotherapy, 63(8), Article e00605-19. https://doi.org/10.1128/AAC.00605-19

Gerson, S. L., Kaplan, S. L., Bruss, J. B., Le, V., Arellano, F. M., Hafkin, B., & Kuter, D. J. (2002). Hematologic effects of linezolid: Summary of clinical experience. Antimicrobial Agents and Chemotherapy, 46(8), 2723-2726. https://doi.org/10.1128/AAC.46.8.2723-2726.2002

Naranjo, C. A., Busto, U., Sellers, E. M., Sandor, P., Ruiz, I., Roberts, E. A., Janecek, E., Domecq, C., & Greenblatt, D. J. (1981). A method for estimating the probability of adverse drug reactions. Clinical Pharmacology and Therapeutics, 30(2), 239-245. https://doi.org/10.1038/clpt.1981.154