We read with interest the papers by TIBERI et al. [1, 2] describing the effectiveness of meropenem/ clavulanate in treating multidrug-resistant (MDR) and extensively drug-resistant (XDR) tuberculosis (TB) patients. In these analyses, 96 patients were treated with meropenem/clavulanate for a median of 85 (interquartile range (IQR) 49-156) days with six adverse events, and 84 patients were treated with imipenem/clavulanate for a median (IQR) of 187 (60-428) days and three adverse events, none renally related. We report a patient with XDR-TB who developed renal Fanconi syndrome apparently due to meropenem/amoxicillin-clavulanate. A 25-year-old South Asian female with intermittently treated pulmonary XDR-TB since 2006, presented to the National Institutes of Health (NIH) in 2015 for treatment under an institutional review board-approved clinical protocol. She was started on daily linezolid, bedaquiline, clofazimine and meropenem 1.5 g i.v. with amoxicillin-clavulanate 500/125 mg orally, both every 8 h. Within 1 month, she developed mild hypophosphataemia (1.8-2.0 mg•dL −1). Her renal function and urinalysis remained normal at 3 months. Her sputum cultures converted to negative after 98 days of treatment. At 4 months, she developed a mild normal anion gap metabolic acidosis (bicarbonate 18-19 mmol•L −1). After 6-7 months, she developed fatigue, moderate proteinuria (600 mg/24 h), aminoaciduria, glycosuria (2-3+ with normal serum glucose), worsening hypophosphataemia with a high urine fractional excretion of phosphorus (24%) consistent with inappropriate urinary phosphate wasting, and hypouricaemia (uric acid 0.9 mg•dL −1), in addition to the normal anion gap metabolic acidosis, all consistent with generalised proximal tubular dysfunction and renal Fanconi syndrome. The aetiology was presumed secondary to medications as the patient did not have other diseases associated with acquired renal Fanconi syndrome. Renal Fanconi syndrome had not been reported with any of her active drugs so, initially, all medications were continued. Renal function remained stable but proteinuria gradually increased to 1200 mg/24 h. Urine electrophoresis was consistent with tubular protein losses. Beta-2-microglobulin levels were also elevated, consistent with proximal tubular dysfunction. Although bedaquiline was considered the most likely culprit due to its limited long-term clinical experience, its 5.5 month half-life made evaluating the effect of stopping this drug challenging. As proteinuria continued to worsen (figure 1) and the patient developed intermittent paresthesias, linezolid was discontinued first. After several days without improvement, linezolid was restarted and bedaquiline discontinued. 1 month later, all TB medications were held due to worsening fatigue and increased urinary protein excretion (2000 mg/24 h; eGFR >100 mL per min per 1.73 m 2). Over the next week, the patient's fatigue, metabolic acidosis, and hypophosphataemia improved and proteinuria declined to 575 mg/24 h, despite measurable serum bedaquiline and clofazimi...
