To the Editor:
Peripheral neuropathy (PN) is a frequent complication of HIV infection, but its actual mechanisms are largely unknown.1 Kallianpur et al2 showed that patients with HFE (hemochromatosis) gene mutations, the C282Y heterozygosis mutation in particular, had a decreased risk of PN. It has been hypothesized that peripheral nervous system damage is facilitated by decreased iron storage related to HIV infection and that HFE mutations compensate for this phenomenon. Although this observation has potential implications for patient risk assessment and treatment, it has been obtained in a limited population under specific highly active antiretroviral therapy (HAART), including stavudine (d4T) and didanosine (ddI). Moreover, diagnosis of PN was based on patient self-reporting of symptoms without any confirmation by electromyography. In this study, we aimed at assessing the possible protective effect of HFE mutations, taking into account potential codeterminants of PN, whose diagnosis was confirmed by electromyography.
A case-control study was performed. All patients with a diagnosis of PN confirmed by electromyography (ie, cases) among HIV-positive patients attending the Clinic of Infectious and Tropical Diseases of the University of Brescia were studied. Patients who did not have signs or symptoms of PN (ie, controls) were selected from the same clinic and matched with cases for the following factors: age (±5 years), gender, reported history of alcohol abuse (defined as alcohol consumption exceeding 60 g of ethanol per day), hepatitis C virus (HCV) antibody (Ab) reactivity, hepatitis B surface antigen (HBsAg)-positive serostatus, and concomitant diabetes. The possible influence of other factors that may also cause PN was explored: risk factors for HIV acquisition; nadir CD4+ T-cell count; occurrence of previous major opportunistic infections (eg, 1993 Centers for Disease Control and Prevention [CDC] clinical class C); exposure to dideoxynucleoside analogues (ddX), such as d4T, ddI, and zalcitabine; and HIV RNA level. To reduce confounding, we excluded patients with other strong risk factors for PN (eg, exposure to drugs toxic for the nerves, cancer, autoimmune diseases).
Analyses of exons 2 and 4 of the HFE gene, mainly aimed at detecting the H63D and C282Y mutations, were performed by denaturing high-performance liquid chromatography (HPLC) as previously described.3
The data were analyzed by a conditional logistic regression model to account for matched case-control data and to identify variables associated with outcome (ie, PN). Variables reaching P ≤ 0.2 on univariate analysis were put into the multivariable model, including HFE gene mutations. Results are expressed as odds ratios (ORs) with their 95% confidence intervals (CIs). Analyses were performed with STATA (STATA Statistical Software, release 9.0; Stata Corporation, College Station, TX).
We studied 57 HIV-positive patients with from PN and 57 HIV-positive controls. Patient characteristics used for matching were as follows: 95% Italian nationality, 82% male gender, mean age of 45 (SD = 5) years, 61% HCV-Ab-positive serostatus, 19% alcohol abusers, and 5% diabetics. Most patients (35 [61%] of 57) in the case group had 3 or more possible risk factors for PN (among alcohol abuse, diabetes, HCV-Ab reactivity, CD4 nadir <100 cells/mm3, CDC clinical class C, HIV RNA >5 log10 copies/mL at the time of electromyography, and ddX exposure). The prevalence of HFE gene mutations did not seem significantly different between the cases and the controls. Sixty-three percent of patients in the case group had no HFE mutations, 25% had H63D heterozygosis, 5% had H63D homozygosis, and 7% had C282Y heterozygosis. Among the controls, 73% had no HFE mutations, 25% had H63D heterozygosis, and 2% had H63D homozygosis.
Table 1 shows results of the conditional logistic regression analysis. In the univariate model, the following variables seemed to be significantly associated with the risk of PN: intravenous drug use (IVDU) as a risk factor for HIV acquisition, nadir CD4+ T-cell count, 1993 CDC clinical class C, and HIV RNA level. Moreover, exposure to d4T was associated with a risk of PN, but the association did not reach statistical significance. Conversely, HFE mutations were not associated with protection from PN, whereas a trend toward increased risk was found for the presence of H63D homozygosis (OR = 3.04, 95% CI: 0.31 to 29.63; P = 0.338). In the multivariable analysis, IVDU and HIV RNA level emerged as the only 2 factors significantly associated with the risk of PN.
Our study did not show any apparent protective role of HFE mutations (including C282Y) for PN in HIV-positive patients on HAART. The prevalence of HFE gene mutations in the group of HIV-positive patients with a confirmed diagnosis of PN was even higher than that found in the Italian general population in the same region (15% in the general population vs. 30% in our HIV-positive population for H63D homozygosis and 5% vs. 7% for C282Y heterozygosis).4 Although a low prevalence of C282Y heterozygosis was found in our population, an apparent paradoxic effect was demonstrated, because all 4 patients with C282Y heterozygosis presented with PN. Moreover, diverse HFE mutation patterns that could also lead to iron loading were found at a significant prevalence in our population, but none of them was associated with protection from PN. For instance, Kallianpur et al2 did not find any statistical association between H63D heterozygosis and protection from PN. In our study, H63D homozygosis was not associated with PN, although its effect on iron loading could be superior to that produced by the heterozygosis pattern. Therefore, the protective mechanism of iron loading hypothesized by others2 needs to be re-evaluated.
The only 2 factors independently associated with the risk of PN were IVDU and a high HIV RNA level. Our data are consistent with a particularly high prevalence of PN in patients who acquired HIV through IVDU.5 IVDU was a risk factor independently of HCV or alcohol abuse; thus, a possible neuropathogenic effect of recreational drugs by themselves could be postulated. Moreover, in univariate analysis, advanced viroimmunologic and clinical conditions as well as exposure to d4T were associated with an increased risk of PN. Our findings go in the same direction as those of others,6 suggesting that HIV by itself, host factors, and the effect of drugs (eg, d4T) are codeterminants in influencing the risk of PN, perhaps more importantly than HFE gene mutations. More studies are needed to support whether iron deficiency is implicated in the etiology of PN, and thus the need of iron supplementation to treat or prevent PN.
The authors acknowledge the dedicated effort of all their colleagues who participated in the recruitment of patients: Valeria Tirelli, Elena Sosta, Giuliana Cologni, Francesca Gatti, Paola Nasta, Alessandro Matti, Roberto Allegri, Alberto Bergamasco, and Salvatore Casari. This study was presented in part at the 14th Conference on Retroviruses and Opportunistic Infections [abstract E-128]. This work was supported in part by funding from the University of Brescia, School of Medicine, assigned to Dr. Quiros-Roldan (“ex-60%,” year 2006).
Silvia Costarelli, MD*
Carlo Torti, MD*
Luisa Benerini Gatta, BS†
Carmine Tinelli, MD‡
Giuseppe Lapadula, MD*
Eugenia Quiros-Roldan, MD, PhD*
Ilaria Izzo, MD*
Filippo Castelnuovo, MD§
Giorgio Biasiotto, BS†
Paolo Arosio, BS†
Giampiero Carosi, MD*
*Institute of Institute of Infectious and Tropical Diseases University of Brescia Brescia, Italy
†Institute of Biochemistry University of Brescia Brescia, Italy
‡Service of Biostatistics IRCCS Policlinico S. Matteo Pavia, Italy
§Spedali Civili di Brescia Brescia, Italy
1. Lewis W, Copeland WC, Day BJ. Mitochondrial DNA depletion, oxidative stress, and mutation: mechanism of dysfunction from nucleoside reverse transcriptase inhibitors. Lab Invest
2. Kallianpur AR, Hulgan T, Canter JA, et al. Hemochromatosis (HFE) gene mutations and peripheral neuropathy during antiretroviral therapy. AIDS
3. Biasiotto G, Belloli S, Ruggeri G, et al. Identification of new mutations of the HFE, hepcidin, and transferrin receptor 2 genes by denaturing HPLC analysis of individuals with biochemical indications of iron overload. Clin Chem
4. Salvioni A, Mariani R, Oberkanikis C, et al. Prevalence of C282Y and E168X HFE mutations in Italian population of Northern European ancestry. Haematologica
5. Berger AR, Schaumburg HH, Gourevitch MN, et al. Prevalence of peripheral neuropathy in injection drug users. Neurology
6. Lichtenstein KA, Armon C, Baron A, et al, for the HIV Outpatient Study Investigators. Modification of the incidence of drug-associated symmetrical peripheral neuropathy by host and disease factors in the HIV Outpatient Study cohort. Clin Infect Dis