Effect of Methylprednisolone or Zoledronic Acid on Resolution of Active Charcot Neuroarthropathy in Diabetes: A Randomized, Double-Blind, Placebo-Controlled Study

Charcot neuroarthropathy (CN) is a complex syndrome affecting people with diabetes and peripheral neuropathy but preserved vascularity. Without early intervention, active CN may have long-term consequences (1). Recent studies suggest the role of proinflammatory cytokines, particularly tumor necrosis factor-α (TNF-α) and interleukin-1β (IL-1β), in mediating inflammatory osteolysis by activating the receptor activator of nuclear factor-κB (RANKL-NF-κB) pathway, thus perpetuating bone loss in an insensate foot with active CN (2,3). The goal of treatment in active CN is therefore to curtail the ongoing inflammation and/or osteolysis. Antiresorptive agents to inhibit osteolysis have been used with modest success (4,5), but anti-inflammatory agents have not been evaluated for CN. We studied the efficacy of methylprednisolone (MP) or zoledronic acid (ZA) in comparison with placebo in active CN along with total contact cast (TCC) as the primary modality.

Included participants had a clinical diagnosis of active CN (erythematous and swollen) with a foot temperature difference exceeding 2°C by infrared dermal thermometry (FLIR Systems, Inc., Orlando, FL) compared with a similar site on the opposite foot, substantiated by X-ray and/or MRI. Bone turnover markers (BTMs) P1NP and CTX (Elecsys; Roche Diagnostics, Mannheim, Germany) were estimated by electrochemiluminescence immunoassay (ECLIA) (Roche Cobas 600 Analyzer; Roche Diagnostics) with coefficient of variation (CV) 1.5% to 4.7%. Inflammatory cytokines (TNF-α, IL-1β) were estimated by ELISA (human high sensitivity kits BMS223HS and BMS224HS; Thermo Fisher Scientific) with CV of 6% to 8%. Participants were immobilized with standardized fiberglass TCC and subsequently randomized to receive MP 1 g in 100 mL normal saline (group A), ZA 5 mg in 100 mL infusion (group B), or placebo 100 mL normal saline (group C) once a month for three consecutive months. All participants were followed fortnightly, with change of cast as required. An average of three temperature recordings 30 min after removal of cast was recorded. Remission was defined as a temperature difference <2°C on two successive follow-up visits. Bone mineral content, BTMs, and inflammatory cytokines were evaluated at baseline and 6 months. The study was approved by the institutional ethics committee, and informed consent was obtained from all participants (NCT03289338). A modified intention-to-treat analysis was performed.

Thirty-six participants were included in the study. Baseline characteristics of the cohort are summarized in Table 1. The mean ± SD time for clinical resolution in the whole cohort was 15.5 ± 4.2 weeks. Time to resolution was significantly higher in the MP group (19.4 ± 2.8 weeks) as compared with either the ZA (14.6 ± 4.4 weeks, P = 0.01) or placebo (13.5 ± 2.9, P = 0.01) group; however, there was no difference between ZA and placebo (P = 0.98). MP reduced TNF-α by 27% (P = 0.03) and IL-1β by 30% (P = 0.02) from baseline as compared with placebo. The reduction of cytokines was comparable between ZA and placebo. BTM analysis revealed an 18% increase in CTX and 5% reduction in P1NP with MP. CTX was reduced by 35% and 11% and P1NP was reduced by 24% and 19% with ZA and placebo, respectively. Adverse events included worsened glycemia with MP and flu-like reaction (n = 5) and acute kidney injury (n = 2) with ZA.

This is the first randomized, placebo-controlled study comparing the effects of MP and ZA with standard of care in patients with active CN of the foot. Pulse MP prolonged time to remission as compared with either ZA or placebo. Inflammatory markers declined in all groups, but bone resorption increased and bone formation decreased with the use of MP, translating into an overall loss of bone mass.

The role of inflammation in the etiopathogenesis of active Charcot foot is unequivocal (2). We observed that MP, a potent anti-inflammatory agent, delays time to remission compared with TCC alone for active CN, despite an effective reduction of the proinflammatory milieu, suggesting that ongoing osteolysis is mediated through activation of the cytokine-independent RANKL pathway. Moreover, MP aggravates hyperglycemia and negatively regulates both CGRP (a neuropeptide that antagonizes RANKL) and osteoprotegerin (a decoy receptor for RANKL), and these may contribute to increased RANKL-NF-κB activity independent of cytokines. Therefore, it is conceivable that although inflammation is pivotal to the pathogenesis of active CN, reducing inflammatory cytokines alone may not suffice for an earlier resolution of active CN.

We observed no added benefit of ZA for earlier remission of CN as compared with TCC alone, as previously documented (5). The strengths of this study include an intensive follow-up, a comparison of a potent anti-inflammatory agent (MP) with ZA and TCC, and a prospective analysis of BTMs and inflammatory markers. In conclusion, MP does not reduce time to remission in active CN of the foot despite reduction in inflammatory cytokines.

• Received August 17, 2019.
• Accepted September 8, 2019.

• © 2019 by the American Diabetes Association.