Study Design and Setting
A retrospective study was performed analyzing pediatric HO patients treated in our hospital from January 2013 to December 2016. The main inclusion criteria were as follows: 1) pediatric patients aged below 18 years old. 2) patients with HO underwent surgical bone decompression. 3) patients persisted to the follow-up and had been followed for at least 24 months. The main exclusion criteria included: 1) patients suffered from direct spreading osteomyelitis, such as post-traumatic osteomyelitis, or infection after internal fixation. 2) patients received a soft tissue surgery, 3) follow-up losted patients. The collectiong clinical data include name, sex, age, diagnosis, serum inflammatory indexs (White blood cell count, Erythrocyte Sedimentation Rate, C-reactive protein), bacteriology, imaging data, site of osteomyelitis, operation process, discharge date, surgery date, telephone number and duration of follow-up .
Description of Study Population
Forty-one patients met the criterion and were included in our study. (Figure 1) Twenty-one patients (13 boys, 8 girls), a mean age of 10.14 years old (range, 3-18 years old), received decompression surgery with antibiotic-impregnated CS were allocated to CS group. Twenty children (17 boys, 3 girls), a mean age of 11.85 years old (range, 2-18 years old), received decompression surgery alone were allocated to control group. (Table 1) All patients had no draining sinuses. All patients had no comorbidities such as sickle cell anemia, deep vein thrombosis, septic pulmonary embolism, pneumonia,empyema and endocarditis.
Table 1. Demographics of Hemotogenous Osteomyelitis patients.
|
CS Group †
|
Control Group
|
p value
|
Sex
|
|
boy
|
13
|
17
|
0.159
|
girl
|
8
|
3
|
Age (years old)
|
10.14 (3-18)
|
11.85 (2-18)
|
0.226
|
Course of disease (days)
|
53 (8-127)
|
59 (10-110)
|
0.92
|
Patients with fever
|
18
|
12
|
0.085
|
Days of fever
|
7 (3-14)
|
8 (4-12)
|
0.448
|
Bacteriology
|
|
MRSA ‡
|
5
|
4
|
1.00
|
Non MRSA
|
16
|
16
|
Staphylococcus aureus
|
10
|
10
|
|
Enterobacter cloacae
|
1
|
0
|
Negative
|
5
|
6
|
Previous operation history £
|
|
Yes
|
8
|
4
|
0.306
|
NO
|
13
|
16
|
White blood cell count(*10^9/L)
|
9.47 (5.99-21.78)
|
10.05(5.46-25.46)
|
0.682
|
Erythrocyte Sedimentation Rate(mm/L)
|
37.86 (3-102)
|
34.60 (5-85)
|
0.695
|
C-Reactive
protein (mg/L)
|
28.21 (0-87.18)
|
26.70 (0.2-107.8)
|
0.909
|
† CS, Calcium Sulfate
‡ MRSA, Methicillin-Resistant Staph. Aureus
£ Previous operation history: a biopsy surgery
Description of surgery
The deep soft-tissue fluid collections or abscesses and subperiosteal were removed. Decompression was performed by creating an appropriately placed cortical window in long bone osteomyelitis, followed by curettage of intraosseous abscess. Eggshell-like decompresion technology[17] was used in calcaneus osteomyelitis, followed by curettage of abscess.
Then difference between CS group and control group: The former one involves antibiotic-impregnated CS application to obliterate the dead space and release local antibiotic. 1g vancomycin (or 160mg gentamicin or both) and 10cc calcium sulfate were mixed thoroughly using the solvent provided by the manufacturer until a smooth paste is formed (approximately 30 s). Calcium sulfate (beads or block-shaped or both) is then placed or filled into the defect as well as supero-inferior medullary cavity. (Figure 2) The paste is allowed to cure undisturbed for at least 15 min after mixing. The volume of calcium sulfate varied according to the size of the bone defect. No drainage tube was used. The later one without this antibiotic material, an irrigation tube was placed in the lower medullary cavity and a drainage tube was placed in the upper medullary cavity. (Figure 3) If the pus is small enough, we place a drainage tube only. In both CS group or control group, a cast or external fixation were applied if bone is unstable after decompression.
Aftercare
All patients were routinely given intravenous antibiotics post operation. Cefmetazole, 100mg/kg/day, was chosen in most patients. Clindamycin, 30mg/kg/day, was chosen in a few patients who were allergic to cefmetazole. Vancomycin, 40 mg/kg/day, was chosen when the pathogenic bacterium showed resistance to cefmetazole. Postoperative antibiotic including cefmetazole (34 cases), clindamycin (7 cases), vancomycin (9 cases), some patients received combinations of medications. In CS group, intravenous antibiotics were administered for a mean of 10.9 days followed by oral antibiotics for a mean of 15.2 days. In control group, intravenous antibiotics were administered for a mean of 12.3 days followed by oral antibiotics for a mean of 16.4 days. In CS group, cleaning the wound regularly by anerdian skin disinfectant. In control group, normal saline solution was used to irrigate by continuous flow every day. No adjuvant dilute Betadine (povidone-iodine) wash or antibiotic powder in the wound was used. The irrigation tube was removed when the fluid looks clear and the drainage tube was removed when the daily output is less than 5ml. The patients were released when their symptoms improved and the drain removed.
Outcome Measures
The patients are regually reviewed in the outpatient department. Defined outcome parameters included infection recurrence, hospital stay, wound healing duration, serum inflammatory markers, pathological fracture, leakage of the incision and follow-up time. Infection recurrence was defined as a worsening clinical symptoms, an continuously increasing serum inflammatory markers, formation of sinus tract and(or) a continuing to be eroded bone tissue by X-ray. Hospital stay is calculated based on the number of days following surgery till discharge. Wound healing duration was defined as the postoperation till incision healing with no exudation and suture removed. Pathological fracture was defined fractures without apparent violence.
Statistical Analysis
Statistical analysis was performed using the SPSS (Version 13) software package (SPSS Inc, Chicago, IL). Descriptive statistics were conducted for all variables. Continuous variables were expressed as the mean standard deviation or the minimum and the maximum depending on data distribution. Student t-test was used for continuous variables. Nonparametric test was used for variance inhomogeneity. A chi square test was used for categorical variables between the two groups. Fisher exact test was used in cases in which one or more of the expected variables was less than five. A value of p<0.05 was considered to be statistically significant.