MRSA has become the most prevalent cause of AHO in pediatric patients[1]. Vancomycin therapy has been well established in children patients and demonstrates positive clinical outcomes. However, there are few options for children with renal impairment or intolerance to vancomycin. Linezolid could be an alternative to vancomycin for MRSA AHO[7], but it has been only approved for the use in pneumonia, bacteremia, skin infections, and periprosthetic joint infections caused by antibiotic-resistant Gram-positive cocci[11, 22-24]. To date, no specific AHO study in children has been conducted[8, 10]. There is no literature comparing the clinical outcomes and adverse effects of linezolid and vancomycin for the treatment of specific MRSA AHO
As far as we know, this is the first research comparing the efficacy and safety of linezolid and vancomycin in the treatment of MRSA AHO. In aspects of clinical cure rates, microbiological eradiation, and hospital length of stay, linezolid and vancomycin have similar efficacy. No significantly differences were observed in most of the adverse events, apart from anemia.
It is difficult to compare our results with previous literature on MRSA AHO for the following reasons. First, the data provided by some previous studies cannot be compared, because the prescribed dose was unavailable or some other antibiotics were used with linezolid or vancomycin[4, 5]. Second, previous studies on linezolid were limited by sparse AHO clinical outcome data coupled with concern over adverse effects with long term usage[9, 10, 22, 24]. To date no specific study of linezolid in the treatment of AHO has been conducted[8, 11, 24]. Thirdly, in other published articles with the treatment of linezolid or vancomycin in children, clinical outcomes were not compared between different antibiotics[11, 15, 16, 19].
Although no significant differences were observed in hemoglobin value at admission and EOT, the incidence rate of anemia was higher in the linezolid group (5/6, 83.3%) than in the vancomycin group (3/11, 27.3%, P=0.049), and the minimum value of hemoglobin in the ward also differed between treatment groups (P=0.009). These findings seem to be inconsistent with previous studies that showed the incidence of anemia of 7.1%-15.3% with the treatment of linezolid in children[8, 10]. However, the treatment for MRSA AHO requires weeks of therapy[1] [30 (IQR 22-37) days in our study], and previous adverse effect data in children were limited to those seen when pediatric patients received linezolid for short duration (11.3±5.0 days)[9, 15]. What’s more, the total incidence rate of anemia previously was low, but 2 patients with anemia received linezolid for 32 and 36 days respectively, which was longer than the others’ [20 (IQR 15-28) days][8]. According to a previous study in adults, pretreatment hemoglobin value of < 10.5 g/L was an independent risk factor in patients who received more than 4 weeks of linezolid treatment[25]. However, in 3 patients with anemia (60%) in our study, hemoglobin values were higher than 10.5 g/L at admission. The prolonged use of linezolid was associated with an increased frequency of anemia, and children seemed to be more sensitive to the transient bone marrow suppression effect of linezolid. Additionally, in children with AHO, factors including surgical blood loss, lesion range, and nutritional status may affect hemoglobin level, hence further prospective randomized controlled trials are needed to certify this observation.
The most common sites of infection were femur (52.9%) and tibia (29.4%), which were consistent with previous studies [26]. Inflammatory markers, including leukocyte, ESR, and CRP were compared at admission and EOT between treatment groups. Both CRP and ESR were elevated at the time of admission (17/17 >4mg/mL, 17/17 >20 mm/first h, respectively), and declined to normal within therapy (15/17 <4mg/mL, 9/17 <20 mm/first h, respectively). This observation was compatible with that in previous articles [26-29]. According to a previous prospective study, leukocyte was an inappropriate predictor of response because only 35% of children demonstrated leukocytosis (leukocyte >12 ×109/L) [30]. In our study, leukocytosis only occurred in 5/17 (29.4%) of children at admission, that is consistent with previous study [30].
The clinical chemistry criteria for DILI we used was applicable to patients of all ages[17], and no DILI was reported in this study. However, there is no independent diagnostic criteria for DILI in children[18], and the increase of ALP in serum is more common in children in the period of vigorous bone growth and development. Therefore, further studies are needed to explore potential DILI induced by linezolid or vancomycin. RMS is a rare but serious adverse reaction during treatment with intravenous antibiotics [20]. Although not observed in patients in this study, it occurred in one of our patients diagnosed with subacute hematogenous osteomyelitis (data not shown), and it was supposedly associated with rapid infusion of the first dose of the drug [21].
Discontinuation rate was higher in the vancomycin group (2/11, 18.2%) than in the linezolid group (0/6, 0.0%). This difference was not significant and may be caused by the sample size. The two failed patients underwent severe adverse events (renal impairment, 1, neutropenia, 1). Renal impairment caused by vancomycin in children was reported in several studies[16, 19, 31-33], and neutropenia associated with linezolid happened in 5 (7.57%) children in a previous study[11]. However, few articles reported vancomycin related neutropenia in the treatment of AHO in children. Fortunately, the neutrophil level of this child returned to normal level (> 1.5×109/L) at EOT.
Interestingly, although no racial differences were observed in treatment outcomes and adverse events, ethnic minority children seemed to have a longer time between symptom onset and admission (P=0.102) and a lower rate previous antimicrobial use before admission (P=0.053) than Han children. It was suggested in previous studies that in cases of MRSA, patients should receive empirical antibiotic treatment at an early stage[26, 34]. This manifests the importance and greatness of the health poverty alleviation project proposed and carried out by our government, the aim of which is to provide basic needs to those still disadvantaged and in poor health, including improving their accessibility to, and affordability of, health services[35].
Therefore, with appropriate monitoring for signs of hematologic and toxicity, linezolid could be an alternative to vancomycin for the treatment of MRSA AHO in children. Additionally, its use allows patients to switch to the oral route more quickly, reducing the incidence of venous complications[8, 36, 37].
There are several limitations in this study. Although this study was non-randomized observational research, the authors followed strict inclusion criteria to decrease the potential bias. Despite this study used 8 years of data from our institution, only 17 MRSA AHO cases were included. Due to the small sample size, some differences in clinical outcomes and adverse events between treatment groups may be shadowed. However, to our knowledge, this study is the first to compare the clinical efficacy and safety of linezolid and vancomycin in the treatment of MRSA AHO in children.