LAUSR

To the Editor, We read with interest the article by Mediano and colleagues, “Microbial Diversity in Milk of Women With Mastitis: Potential Role of Coagulase-Negative Staphylococci, Viridans Group Streptococci, and Corynebacteria” (Mediano et al., 2017). The authors conducted an extensive microbial characterization of milk samples from women described as suffering from mastitis to determine the etiology of this condition. Although the findings were of interest, the study has a number of failings. The inclusion criteria for mastitis were either “both local (breast redness, pain, and engorgement) and systemic symptoms (fever or flu-like symptoms) or only local symptoms (pain, engorgement, reduced milk secretion)” (p. 310). We believe that local symptoms only do not indicate mastitis. In most instances, breast symptoms are relieved with more frequent feeding or expressing, massaging, using hot/cold compresses, etc. Consensus guidelines define mastitis as when a woman has systemic symptoms in addition to local inflammatory breast symptoms (Amir & The Academy of Breastfeeding Medicine Protocol Committee, 2014). Local breast symptoms only are diagnosed as engorgement or a blocked duct (Berens, 2015; Witt, Bolman, Kredit, & Vanic, 2016). Therefore, we believe that milk samples collected from women with local breast symptoms only may not indicate mastitis. Furthermore, milk samples were analyzed from both breasts, even if only one was affected. Since breasts produce milk independently and women commonly experience mastitis in one breast only (Jahanfar, Ng, & Teng, 2013), we believe that analysis of milk from unaffected breasts was inappropriate. However, our greatest concern is the lack of milk samples from women without mastitis to serve as a comparison group. Whereas staphylococci, streptococci, and corynebacteria were the most prevalent bacterial groups reported, how can we be sure that these groups are not also present in comparable ratios in milk samples from women without mastitis? The authors concluded that the identified bacteria may be important etiologic agents of mastitis. Although this may be true, we believe that the absence of a control group precludes any conclusion. Human milk is not a sterile substance, and several hundred bacterial species have been identified in human milk (Cabrera-Rubio et al., 2012); therefore, the presence of bacterial species may not necessarily be indicative of infection. It is concerning that the authors argue that “a proper diagnosis of mastitis requires a microbiological analysis of milk” (p. 316). However, in practice, the lactating mother with mastitis is an unwell woman with a clear clinical diagnosis based on history (febrile symptoms and breast pain) and examination (redness and hardness of part of the breast) (Amir & Lumley, 2006; Iatrakis et al., 2013), and it is impractical to wait for microbiological confirmation. This contrasts with veterinary medicine, where greater reliance on investigations may be standard practice. Another point that we would like to make relates to the comments about the isolation of Candida spp. The authors are confused when they say “human mastitis cases have been frequently misdiagnosed as mammary candidiasis” (p. 316). Although we agree that some women with breast pain (e.g., cases of nipple vasospasm) are misdiagnosed as having a Candida infection, this condition presents as radiating breast pain in a normal looking breast (Amir et al., 2013) and never presents as redness, lump, or fever; therefore, the differential diagnosis of mastitis is never Candida infection of the breast. In conclusion, we hope that readers understand that the absence of a control group is a fatal flaw in this article. We look forward to seeing appropriately conducted studies in which standard demographic characteristics are collected and samples of milk are clearly defined, that is, milk from the affected breast of women with mastitis according to the standard definition, in addition to milk from women without mastitis for comparison.