Skeletal fluorosis after prolonged voriconazole therapy

PRACTICAL TIP FOR PAEDIATRICIANS

A 6-year-old girl underwent allogeneic hematopoietic stem cell transplantation for a relapsing acute myelogenous leukemia. She developed severe chronic graft-versus-host disease which required immunosuppressive therapies, including a prolonged course of high-dose steroids. She also presented a thrombotic microangiopathy complicated by renal failure requiring dialysis.

Eleven months after transplantation, she was diagnosed with pulmonary and cerebral aspergillosis, prompting initiation of intravenous voriconazole. Unusually high doses were needed (11 mg/kg/dose every 8 hours) in order to achieve therapeutic plasmatic trough concentrations above 1 to 2 mg/L. After 2 months, she was transitioned to oral voriconazole at equivalent dosing.

Six months after being on oral voriconazole, she complained of fatigue and diffuse musculoskeletal pain, initially attributed to her steroid treatment. Radiographs (Figure 1) of her femurs showed diffuse osteopenia and pronounced bilateral periosteal reactions with exostosis (arrows).

Figure 1.

Radiographs revealed osteopenia and unusual periosteal reaction with exostosis (arrows) of the femurs.

The diagnosis of osseous fluorosis induced by prolonged voriconazole therapy was suspected and confirmed by a very high fluoride plasma concentration of 27.0 µmol/L (0.25 to 1.0 µmol/L). Her alkaline phosphatase level was also increased at 627 U/L (134 to 518 U/L), with a known normal baseline level preceding therapy. Voriconazole was discontinued and oral posaconazole was substituted in its place. Five months after voriconazole discontinuation, her musculoskeletal symptoms resolved and her fluoride plasma concentration decreased to 1.30 µg/mL.

DISCUSSION

Voriconazole is the first-line therapy for invasive aspergillosis. Voriconazole and its metabolite (NO-voriconazole) contain three fluoride atoms, and long-term treatment can lead to elevated plasma fluoride levels, increasing the risk of skeletal fluorosis by enhancing osteoblast proliferation (1). Given that fluoride clearance is renal, fluoride exposure increases with renal failure, potentially resulting in fluorosis, as illustrated by this case. Moreover, fluoride accumulates more rapidly in the bones of children compared to those of adults related to their higher metabolic activity and greater vascularity, thereby increasing the risk of skeletal fluorosis (2). All these factors, along with the use of an unusually high dosage of voriconazole to achieve therapeutic plasmatic concentrations, contributed to the occurrence of fluorosis in this patient. She was receiving an equivalent of 93.6 mg of fluoride daily (400 mg of voriconazole contains 65mg of fluoride), which is 15-folds above the threshold associated with an increased risk of bone effect according to WHO (3,4). Interestingly, daily and cumulative voriconazole doses, but not trough levels, have been shown to correlate with the risk of periostitis (5).

We recommend that children with prolonged voriconazole therapy, especially with high doses, and/or those with impaired renal function, should have monitoring of fluoride levels. The diagnosis of skeletal fluorosis could be delayed or mistaken with other more common painful conditions, particularly among oncology patients, who may have various reasons to complain of musculoskeletal pain (1,2). Alkaline phosphatase monitoring may also be useful, although cases of periostitis with normal values have been reported (6). Prompt cessation of treatment leads to resolution of symptoms, but this may take days to months (1,7). In these cases, posaconazole, an azole containing less fluoride, has been suggested as an alternative treatment.

Funding information: There are no funders to report for this submission.

Potential Conflicts of Interest: HB reports consulting fees from Novartis Oncology and consulting fees and a travel grant from Jazz Pharmaceuticals, outside the submitted work. There are no other disclosures. All authors have submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest. Conflicts that the editors consider relevant to the content of the manuscript have been disclosed.