Etiology and Clinical Characteristics of Acute Suppurative Osteomyelitis in Children a Single Center Study in China

Background Acute suppurative osteomyelitis is a common disease in pediatric orthopedics, but there are few studies on the etiology and treatment of this disease in children. Objective To investigate the etiology and clinical characteristics of acute suppurative osteomyelitis in children. Methods A retrospective study was conducted of pediatric in-patients with acute suppurative osteomyelitis between January 2011 and December 2017. Subperiosteal or intramedullary specimens were collected from each patient for culture. Bacterial identication and antibiotic susceptibility test were performed with the Vitek system. Results A total of 104 patients were included and 64 (61.5%) were male. Sixty-six isolates were obtained from 65 (62.5%) patients. The most common pathogen was Staphylococcus aureus (51.0%, 53/104), followed by Escherichia coli (2.9%, 3/104), Pseudomonas aeruginosa (1.9%, 2/104), Streptococcus pneumoniae (1.9%, 2/104), Ochrobactrum anthropi (1.9%, 2/104), and other bacteria (3.0%, 3/104). Two pathogens were isolated from one patient. The etiology was unknown based on culture in 39 (37.5%) patients. The age of the patients infected with S. aureus was 6.67 (1.7, 9.1) years, which was signicantly higher than that of patients infected with non-S. aureus, whose age was 2.42 (0.7, 7.8) years (Z = 2.20, P=0.028). Results of the antibiotic susceptibility test in 53 S. aureus strains showed that the resistance rates to oxacillin, trimethoprim-sulfamethoxazole, clindamycin, and erythromycin were 43.4%, 11.3%, 67.9%, and 69.8%, respectively. All S. aureus strains were sensitive to linezolid and vancomycin. More patients with sequelae were found among those infected with


Background
Acute suppurative osteomyelitis is an in ammation of the periosteum, bone mass, and bone marrow caused by bacteria. It is common in orthopedic infections of children [1] . If it is not effectively treated at an early stage, severe sepsis, pyogenic meningitis, and infectious shock may occur secondarily [2][3][4] , and severe cases are life-threatening. Some patients may progress to chronic infections, bone growth abnormalities, and growth retardation [3,5] . Therefore, it is an infectious disease that seriously threatens child health. Early sensitive antibacterial treatment is the key to improve the prognosis of this disease, and the rational choice of antibiotics is based on the correct judgment of the pathogen. The clinical manifestations of osteomyelitis caused by different bacteria vary, and the preferred antibiotics for treatment are also different. Therefore, understanding the clinical characteristics, common pathogens, and drug resistance patterns of suppurative osteomyelitis in children is very important for the empirical medication by clinicians. This study retrospectively analyzed the clinical characteristics, pathogen types, and drug resistance patterns of children with acute suppurative osteomyelitis treated in our hospital from January 2011 to December 2017, to provide a reference for clinical diagnosis and treatment.
1 Study Subjects And Methods

Study design
This was a retrospective study. All case information of patients with a discharge diagnosis of acute suppurative osteomyelitis from January 1, 2011 to December 31, 2017 was searched in the electronic case system of our hospital. Inclusion criteria were as follows: (1) Children with local redness, swelling, heat, and pain, or restricted local movement; (2) A disease course of less than 2 weeks [3] ; (3) Local tenderness; (4) Imaging examinations suggesting in ammation of the periosteum, bone mass, and bone marrow; (5) Purulent specimens obtained by subperiosteal puncture or surgery. Exclusion criteria were as follows: failure of subperiosteal puncture or surgery to collect su cient specimens for bacterial culture.

Bacterial identi cation and antibiotic resistance test
Bacterial identi cation and antibiotic resistance test were performed with Vitek compact (Mérieux, France). Repeated isolation from the same patient was considered as one S. pyogenes clone.

Statistical analysis
Comparison of the incidence data between groups was performed using the χ2 test. Data assumed to be normally distributed were presented as mean ± standard deviation. Comparison of means between groups was performed using the t test. Data not assumed to be normally distributed were presented using the median with interquartile range (IQR) in parentheses. The Mann-Whitney U test was used to compare medians between groups. P < 0.05 was considered to indicate statistical signi cance.

Demographic information
A total of 104 patients with acute suppurative osteomyelitis and whose ages ranged from 6 days to 14 years (mean age was 5.5±4.2 years) were included in the study. Sixty-four (61.8%) patients were male, and 49.0% (51/104) were aged < 4 years. In 87 cases of osteomyelitis, the lesions were located in the long bones of the extremities (83.7%, 87/104). Among these cases, in 33 cases (31.7%), the lesions were located in the femur; in 23 cases (22.1%), the lesions were located in the tibia; in 14 cases (13.5%), the lesions were located in the humerus; and in 61 cases, the lesions were located in the metaphysis of long bones.

Isolation of pathogens
A total of 104 children underwent subperiosteal puncture or surgical specimen culture. The positive rate was 62.5% (65/104). A total of 66 bacterial strains were isolated. Among them, 53 were Staphylococcus aureus strains, with an isolation rate of 51.0% and a composition rate of 81.5% (53/65) in pathogenpositive individuals. Other bacteria included 3 strains of Escherichia coli, 2 strains of Pseudomonas aeruginosa, 2 strains of Streptococcus pneumoniae, 2 strains of Ochrobactrum anthropi, 1 strain of Streptococcus viridans, 1 strain of Streptococcus pyogenes, 1 strain of Salmonella dublin, and 1 strain of Enterobacter cloacae. S. aureus and O. anthropi were successively isolated from one patient. A total of 37.5% (39/104) of the specimens did not show pathogenic bacteria. There was no signi cant difference in the proportion of patients who used antibiotics before sample collection between the pathogen-culture positive patients (90.8%, 59/65) and pathogen-culture negative patients (92.3%, 36/39) (χ 2 = 0.07, P = 0.79). There was no signi cant difference in the proportion of patients who used antibiotics before sample collection between the S. aureus group (94.3%, 50/53 cases) and other patients (88.2%, 45/51) (χ 2 = 0.58, P = 0.45).

Staphylococcus aureus drug sensitivity results
All 53 strains of S. aureus produced β-lactamase, and they were resistant to penicillin G. Methicillinresistant S. aureus (MRSA) accounted for 43.4% (23/53) of all strains. The sensitivity of the 53 strains to commonly used antibiotics is shown in Table 1

Clinical characteristics, treatment methods, and outcomes
A total of 77 cases among these 104 children had fever, 19 cases were trauma causes, and 9 cases had infections in adjacent tissues. A total of 84.6% (88/104) of the lesions occurred in long bones, and 67.3%  The initial treatment of the 8 patients was β-lactams, which was switched to vancomycin or linezolid after bacterial culture con rmed MRSA infection; # 1 case was initially treated with vancomycin, and it was switched to imipenem after bacterial culture con rmed Ochrobactrum anthropi infection; ** All cases were infected by MSSA. Table 2 Clinical aspects of acute suppurative osteomyelitis caused by Staphylococcus aureus or non-Staphylococcus aureus Acute suppurative osteomyelitis can be caused by local trauma, infections of adjacent sites, or bloodborne infections; and blood-borne infections are the most common route of infection in children [6] . Only 18.3% of the children in this study had suspicious causes of trauma and 73% of the patients had hematogenous osteomyelitis, which is consistent with the literature report [6] . The lesions in most of the children were located in the metaphysis of long bones, and the most common locations were the femur, tibia, and humerus, which are also consistent with the results of previous studies [7] .
Pathogen isolation by culture is the key for targeted antibiotic therapy. Most of the bacteria isolated in this study were S. aureus, which is consistent with the results of other studies [3,6,[8][9] . The incidence of MRSA in our study was as high as 43.4%. While this value is consistent with the national resistance rate of 36.1% for Staphylococcus to methicillin in Chinese children [10] , it is signi cantly higher than that reported in other countries [5,9,[11][12] , which suggests that S. aureus is seriously drug resistant in Chinese children. It is generally believed that a long time is required for the antibiotics to reach bone tissues and to achieve effective bactericidal concentrations [13] ; therefore, short-term administration of antibiotics is thought to have little effect on bacterial culture results of bone tissue specimens [14] . Although 91.3% of the children in this study had been treated with antibiotics before collecting bone marrow cavity purulent samples for culture, 62.5% of the children were still found to have pathogenic bacteria. There was no difference in the antibiotic usage rate between the bacterial culture positive patients and negative patients, suggesting that the type of pathogen is also an important factor affecting the positive rate of bone marrow cavity purulent sample culture. At different age stages, the common pathogens that cause suppurative osteomyelitis differ. S. aureus is more common in children over 4-5 years, and Streptococcus or Kingella kingae is more common in children under 4 years [8,9] . The age of children with S. aureus infection in this study was signi cantly higher than that of children with non-S. aureus infection, which is consistent with previous reports [8] ; however, the proportion of children with Streptococcus infection was relatively low, which might be related to the use of sensitive antibiotics before sampling.
Oxacillin is an effective drug for the treatment of MSSA infection. Cephalosporin is safe and effective for the treatment of MSSA osteomyelitis in children. Combined treatment with vancomycin can signi cantly reduce the mortality of severe sepsis [15] . MRSA is resistant to cephalosporins; clindamycin is the preferred option if the patient is not critically ill and the MRSA isolates are sensitive to clindamycin. The drug has good tissue penetration and good effects against bacterial toxins and is recommended as an empirical treatment for bone and joint infections in regions where the drug resistance rate is lower than 10-15% [3,16] . The drug is also recommended in combination with other sensitive drugs to enhance the e cacy and improve the prognosis [17] . Chinese strains are highly resistant to clindamycin and macrolides [10] . The resistance rate of S. aureus to clindamycin was as high as 67.9% in this study; therefore, all children were not treated with clindamycin. MRSA has an even higher resistance rate to clindamycin; therefore, many patients in this study were treated with or in combination with vancomycin/linezolid. In view of the drug resistance features of pathogens in this region, vancomycin or linezolid should be selected for the initial empirical treatment of severe suppurative osteomyelitis in children over 4 years old [8,18−19] . Trimethoprim-sulfamethoxazole has been successfully used for the treatment of pediatric osteomyelitis due to MRSA in a small retrospective study by Messina et al [[20] . S. aureus in China is generally less resistant to sulfa drugs [10] . In this study, the resistance rate of S. aureus to Trimethoprim-sulfamethoxazole was only 11.3%, which can be considered as a combined treatment drug for MRSA infection or as a candidate drug for patients in the recovery stage and in patients with mild conditions. Its exact treatment effect needs to be determined by performing clinical controlled studies with a large sample size in the future. Children with S. aureus osteomyelitis had more severe in ammatory reactions and clinical symptoms and had higher CRP levels. They also had signi cantly more surgical treatments and higher incidences of sequelae than patients with osteomyelitis caused by other pathogens or unknown pathogens, which is consistent with previous reports [15,17] .
Streptococcus is generally believed to be the second most common cause of suppurative osteomyelitis in children [15] . In this study the proportion of streptococcal infections in the non-Staphylococcus aureus group was small, which may be related to the administration of β-lactams in most of the patients before sample collection. Cases with unknown pathogens accounted for a large proportion, and most of the infections were cured after treatments, which are postulated to be caused by β-lactam sensitive bacterial infections. Recently, K. kingae has become an emerging pathogen in the etiology of pediatric acute osteoarticular infection [21] . K. kingae was reported as the most frequent etiologic agent for acute osteoarticular infections in children between 6 months and four years [21][22] . However, this study did not isolate K. kingae, which may be related to the di culty with in vitro culture of the bacteria [21] and to the administration of β-lactams before sampling. A polymerase-chain-reaction (PCR) assay in the tissue sample may enhance the yield of K. kingae [21−2] . In the future, we will perform bacterial PCR detection or high-throughput sequencing in the bone marrow cavity or in subperiosteal purulent samples of acute suppurative osteomyelitis in children with negative culture results to improve the rate of pathogen diagnosis in children.
In summary, in this study, bacterial culture con rmed that the pathogens of suppurative osteomyelitis in children were mainly S. aureus, with a high proportion of MRSA. Most of them were multi-drug resistant bacteria, with signi cant in ammatory reaction of osteomyelitis and high incidence of sequelae. Empirical treatment of osteomyelitis in older children should include vancomycin or linezolid. Children with non-S. aureus infections are younger and are mostly infected with unknown pathogens; further study is needed to determine the etiology of this infection.