- Email: [email protected]
Pediatric acute osteomyelitis in the postvaccine, methicillin-resistant Staphylococcus aureus era
American Journal of Emergency Medicine xxx (2015) xxx–xxx
Contents lists available at ScienceDirect
American Journal of Emergency Medicine journal homepage: www.elsevier.com/locate/ajem
Original Contribution
Pediatric acute osteomyelitis in the postvaccine, methicillin-resistant Staphylococcus aureus era☆ Kristin Ratnayake, MD a, Andrew J. Davis, MD b, Lance Brown, MD, MPH b, Timothy P. Young, MD b,⁎ a b
Division of Pediatric Emergency Medicine, Department of Pediatrics, Rady Children's Hospital, San Diego, CA Division of Pediatric Emergency Medicine, Department of Emergency Medicine, Loma Linda University Medical Center and Children's Hospital, Loma Linda, CA
a r t i c l e
i n f o
Article history: Received 26 February 2015 Received in revised form 6 July 2015 Accepted 6 July 2015 Available online xxxx
a b s t r a c t Objective: We sought to describe the causative organisms, bones involved, and complications in cases of pediatric osteomyelitis in the postvaccine age and in the era of increasing infection with community-associated methicillin-resistant Staphylococcus aureus (MRSA). Methods: We reviewed the medical records of children 12 years and younger presenting to our pediatric emergency department between January 1, 2003, and December 31, 2012, with the diagnosis of osteomyelitis. We reviewed operative cultures, blood cultures, and imaging studies. We identified causative organisms, bone(s) involved, time to therapeutic antibiotic treatment, and local and hematogenous complications. Results: The most common organism identified was methicillin-sensitive S aureus (26/55), followed by MRSA (21/55). Seventy-three bone areas were affected in 67 subjects. The most common bone area was the femur (24/73). Forty-six subjects had 75 local complications. The most common organism in cases with local complications was MRSA (49%). Three subjects had hematogenous complications of deep venous thrombosis, septic pulmonary embolus, and endophthalmitis. Subjects with complications had shorter time to therapeutic antibiotic treatment. When an operative culture was done after therapeutic antibiotics were given, an organism was identified from the operative culture in 84% of cases. Conclusion: Treatment of pediatric osteomyelitis should include antibiotic coverage for MRSA. Most cases of pediatric osteomyelitis occur in the long bones. Hematogenous complications may include deep venous thrombosis and may be related to treatment with a central venous catheter. Operative culture yield when antibiotics have already been given is high, and antibiotic treatment should not be delayed until operative cultures are obtained. © 2015 Elsevier Inc. All rights reserved.
1. Introduction Pediatric hematogenous osteomyelitis typically has a prolonged treatment course involving multiple specialties. Emergency physicians may treat children with osteomyelitis at various stages in the illness, including at diagnosis, at recurrence of disease, and when complications of the disease or its treatment arise. Historically, the long bones have been the major bones affected in pediatric osteomyelitis, and Staphylococcus aureus has been the most common causative organism, responsible for 66% to 70% of cases [1,2]. Streptococcus pyogenes, Streptococcus pneumoniae, Haemophilus influenzae, and Pseudomonas aeruginosa have also been reported as causative agents [2–6]. Recently, increasing numbers of children with invasive infection caused by community-associated methicillin-resistant S aureus (MRSA) have been reported [7]. In the United States, rates of osteomyelitis caused by MRSA increased in children's hospitals during the first
☆ The authors have no financial or other conflicts of interest related to the submission. ⁎ Corresponding author at: Department of Emergency Medicine, Loma Linda University Medical Center, 11234 Anderson St, A-108, Loma Linda, CA 92354. E-mail address: [email protected] (T.P. Young).
decade of this century, whereas those caused by methicillin-sensitive S aureus (MSSA) remained constant [8]. Children with osteomyelitis caused by MRSA experience more severe illness than do their MSSA counterparts, with greater need for procedures and longer hospitalizations [4,9]. Complications recently attributed to MRSA osteomyelitis include deep venous thrombosis (DVT), as well as problems at sites distant to the primary infection such as endocarditis and septic pulmonary emboli [5,10–12]. It is possible that changes in bacteriology have affected the distribution of site of infection in osteomyelitis. Prior to the era of MRSA, Craigen and colleagues [3] in the United Kingdom found that long bone infection (femur and tibia-fibula) was most common, but saw a decrease in long bone infection rate mirroring a decrease in S aureus infection. Infection rates at other sites remained constant. More recently, SaavedraLozano and colleagues [6] found increasing rates of infection with MRSA in Texas and that infection in the foot was more common than either the femur or tibia-fibula. Because of the rapidly changing nature of pediatric osteomyelitis, a more current description of its epidemiology and complications would benefit clinicians who diagnose and treat children with the disease. We previously reported a high rate of MRSA in cases of septic arthritis at our institution [13] and became interested in examining
http://dx.doi.org/10.1016/j.ajem.2015.07.011 0735-6757/© 2015 Elsevier Inc. All rights reserved.
Please cite this article as: Ratnayake K, et al, Pediatric acute osteomyelitis in the postvaccine, methicillin-resistant Staphylococcus aureus era, Am J Emerg Med (2015), http://dx.doi.org/10.1016/j.ajem.2015.07.011
2
K. Ratnayake et al. / American Journal of Emergency Medicine xxx (2015) xxx–xxx
cases of osteomyelitis. At our institution, it is sometimes recommended that antibiotics be withheld until operative cultures are obtained in an effort to increase yield. Although adult studies suggest that preoperative antibiotics do not affect operative culture yield, there are less available data in the pediatric setting [14–16]. Our objective was to examine cases of acute culture- and/or imaging-proven osteomyelitis involving healthy prepubescent children seen in our pediatric emergency department. Our primary aim was to determine the rate of MRSA infection in cases of pediatric osteomyelitis. Our secondary aims were to determine organism susceptibilities for MRSA osteomyelitis, to describe site of involvement and rate of complications of osteomyelitis, to determine whether complications were associated with antibiotic discordance or time to antibiotic administration, and to evaluate the yield of operative cultures when antibiotics had already been administered.
was given in cases where there was no positive culture obtained. We also recorded the day of illness that an operative culture was obtained. We excluded children with repeat visits, chronic osteomyelitis, a history of previous surgery on the affected bone, and chronic illness. We defined chronic illness as the presence of a ventriculoperitoneal shunt, the presence of an indwelling central venous catheter, and conditions such as sickle cell disease, spina bifida, extreme prematurity, complex congenital cardiac lesions, and cancer. Descriptive statistics were calculated using Microsoft Excel 2010 (Microsoft Corporation, Redmond, WA). Where 95% confidence intervals (CIs) are reported, the modified Wald (Agresti-Coull) method was used and was calculated using STATA IC 12 (Statacorp LP, College Station, TX). Subjects with complications were compared with those without complications with respect to both time to therapeutic antibiotic treatment and antibiotic discordance using a Mann-Whitney U test and a χ 2 test, respectively. Both were calculated in STATA, with an α level set at .05.
2. Materials and methods
3. Results
We performed a retrospective medical record review of children 12 years and younger presenting to the pediatric emergency department at our university-based, tertiary care institution between January 1, 2003, and December 31, 2012, with the diagnosis of osteomyelitis. We chose this age group to reflect the epidemiology and pathogenesis of a largely prepubertal sample of children. Gonococcal osteomyelitis has been reported in adolescents, and we aimed to avoid this entity [17]. Subjects were identified by the Ninth Revision of the International Classification of Diseases (ICD-9) diagnostic codes for osteomyelitis (730.00-730.99). We used a standardized data collection form. Data were collected by a trained data abstractor. The data extracted from the medical records were the patient's age, sex, bone(s) involved, presence of comorbidities, previous diagnosis of bone infection, antibiotics given, results of bone aspirate culture and blood culture, and results of advanced imaging studies. A positive evaluation for osteomyelitis was defined by the presence of positive operative or blood culture for a known pathogen or advanced imaging studies indicating the presence of osteomyelitis. A known pathogen was considered to be any organism other than diptheroids. We included coagulase-negative staphylococcus (CONS) as a pathogen, as until 2008 our laboratory did not speciate CONS. Coagulase-negative staphylococcus has also been reported as a rare cause of osteomyelitis in healthy children [18]. We defined the presence of osteomyelitis on advanced imaging studies as a positive radiologic result for magnetic resonance imaging or nuclear medicine bone scan of the affected area. We defined the following as possible sites of infection: the hand, radius/ulna, humerus, scapula, foot, tibia/fibula, patella, femur, pelvis, spine, and cranium. We reviewed all cases of osteomyelitis for local and hematogenous complications. We defined local complications as abscess, myositis, or fasciitis seen on imaging studies or described in operative reports. We defined hematogenous complications as DVT, septic pulmonary emboli, or other infections at secondary locations attributed to the primary infection of osteomyelitis. Medical records were reviewed from initial presentation to the time of data collection for complications. We included DVTs related to catheters placed for long-term intravenous antibiotics as well as those without mechanical inciting factors. We reviewed cases with a pathogen-positive culture for antibiotic discordance. We defined antibiotic discordance as culture positive for one of the following: an organism explicitly resistant to the antibiotic(s) given prior to the culture result, an organism resistant to an antibiotic of the same class (eg, nafcillin and oxacillin), or an organism resistant to the antibiotic(s) administered according to the Sanford Guide in the year that the subject was treated. Our laboratory does not report sensitivities for all organisms and does not test all antibiotics. We recorded the day of illness that therapeutic antibiotics were given. We defined this as the day of illness at which an antibiotic was given that the organism was sensitive to in cases where an organism was identified, or the day of illness at which the first antibiotic
We identified 194 patient encounters with an ICD-9 code for osteomyelitis during the 10-year period. All medical records were complete. A total of 102 encounters met our exclusion criteria. This resulted in a group of 92 subjects who were evaluated for osteomyelitis. Twentyfive subjects had a negative evaluation for osteomyelitis. Therefore, our main study group consisted of the remaining 67 subjects (Fig. 1). After our exclusion criteria were applied to identify cases of interest for this study, the use of ICD-9 codes had a 73% (67/92) positive predictive value for osteomyelitis as we defined it. All included subjects had either a positive imaging study or a positive operative culture. Forty subjects (60%) were male. The median age was 5 years (interquartile range [IQR], 2-10 years; range, 1 month to 12 years; Fig. 2). The most common pathogenic organism was MSSA, followed by MRSA (Table 1). The most common organism recovered by blood culture was MRSA (12/23; 52%), followed by MSSA (9/23; 39%). All isolates of MRSA were sensitive to vancomycin and trimethoprim-sulfamethoxazole. Two were resistant to clindamycin. The most common site was the femur (Fig. 3). Forty-six subjects (69%) had a total of 75 local complications (Table 2). Three subjects had hematogenous complications (Table 3). Two of the subjects with hematogenous complications had infection with MRSA. All subjects with hematogenous complications also had local complications. For cases with complications in which the organism was identified, the most common organism was MRSA (19/39; 49%), followed by MSSA (16/39,; 41%). Subjects with complications had a median time to therapeutic antibiotics of 6 days (IQR, 4-9 days), whereas those without complications had a median time to therapeutic antibiotics of 8 days (IQR, 6-22 days; P = .03). For subjects without complications in which the organism was identified, the most common organism was MSSA (8/14; 57%), followed by MRSA (3/14; 21%). Subjects with and without complications did not differ regarding antibiotic discordance (P = .50). When an operative culture was done after therapeutic antibiotics were given, an organism was still identified from the operative culture in 42 (84%) of 50 cases (95% CI, 71%-92%). An organism was identified from operative cultures when therapeutic antibiotics had not been given prior to operative cultures in 4 (80%) of 5 cases (95% CI, 36%-98%). 4. Discussion We found a 38% rate of infection with MRSA in cases of pediatric osteomyelitis at our institution, nearly equal to that of MSSA. Pediatric osteomyelitis most commonly affected the long bones. A high proportion of our subjects had complications (69%). Subjects with complications had a shorter time to therapeutic antibiotic administration. Operative culture yield was 84% when therapeutic cultures had already been given.
Please cite this article as: Ratnayake K, et al, Pediatric acute osteomyelitis in the postvaccine, methicillin-resistant Staphylococcus aureus era, Am J Emerg Med (2015), http://dx.doi.org/10.1016/j.ajem.2015.07.011
K. Ratnayake et al. / American Journal of Emergency Medicine xxx (2015) xxx–xxx
3
ICD-9 code for osteomyelitis (n = 194)
Previous diagnosis of osteomyelitis (n = 34)
Mastoiditis (n = 3)
Previous surgery or open fracture to bone (n = 6)
Chronic illness (n = 59)*
Negative evaluation for osteomyelitis (n = 25)
Positive evaluation for osteomyelitis (n = 67)
Positive operative culture, blood culture and imaging (n = 16) Positive operative culture and blood culture (n = 1) Positive operative culture and imaging (n = 24) Positive blood culture and imaging (n = 6) Positive operative culture only (n = 6) Positive imaging only (n = 14) Fig. 1. Subject flow diagram. *We defined chronic illness as the presence of a ventriculoperitoneal shunt, the presence of an indwelling central venous catheter, or conditions such as sickle cell disease, spina bifida, extreme prematurity, complex congenital cardiac lesions and cancer.
7 6 4
Organism
No. (%) of isolatesa (n = 55)
95% CI
3
5
Table 1 Pathogenic organism isolated
MSSA Community-associated MRSA β-Hemolytic group A Streptococcus CONS Streptococcus viridans Enterococcus faecium Mycobacterium chelonae S pneumoniae
26 (47%) 21 (38%) 3 (5%) 1 (2%) 1 (2%) 1 (2%) 1 (2%) 1 (2%)
35%-60% 26%-51% 1%-15% 0%-11% 0%-11% 0%-11% 0%-11% 0%-11%
0
1
2
Number of Subjects
8
9
10
We observed a higher proportion of MRSA infection in our study compared with a similar study ranging from 1999 to 2003 [6] and another comprising a time range from 2000 to 2004 [4]. This is consistent with multiple reports of a continuing uptrend in the incidence of MRSA infection in children [7,8]. S pneumoniae was identified as the cause of one case of osteomyelitis in our sample. This is consistent with the low but continued presence of S pneumoniae in cases of pediatric
0
1
2
3
4
5
6
7
8
9
10
Age (years) Fig. 2. Age distribution for children with osteomyelitis.
11
12
a There were 55 isolates in 53 subjects. One subject had MSSA and CONS, and another subject had MSSA and S viridans.
Please cite this article as: Ratnayake K, et al, Pediatric acute osteomyelitis in the postvaccine, methicillin-resistant Staphylococcus aureus era, Am J Emerg Med (2015), http://dx.doi.org/10.1016/j.ajem.2015.07.011
4
K. Ratnayake et al. / American Journal of Emergency Medicine xxx (2015) xxx–xxx
Scapula 4/73 (6%; 95% CI 2-14%)
Cranium 2/73 (3%; 95% CI 0-10%)
Humerus 4/73 (6%; 95% CI 2-14%) Spine 1/73 (1%; 95% CI 0-8%) Radius/Ulna 3/73 (4%; 95% CI 1-12%) Pelvis 6/73 (8%; 95% CI 4-17%)
Patella 1/73 (1%; 95% CI 0-8%)
Femur 24/73 (33%; 95% CI 23-43%)
Tibia-Fibula 20/73 (27%; 95% CI 18-39%)
Foot 8/73 (11%; 95% CI 5-20%)
Fig. 3. Distribution of affected body area in osteomyelitis. Four subjects had infection in 2 bone areas and 1 subject had infection in 3 bone areas, for a total of 73 bone areas in 67 subjects.
osteomyelitis that has been seen since the implementation of the heptavalent pneumococcal conjugate vaccine in the United States in 2001 [4,6]. This incidence may be expected to fall further, as overall rates of pneumococcal infection have declined in the United States since the introduction of the 13-valent pneumococcal vaccine in 2010 [19]. Two of 20 infections with MRSA were resistant to clindamycin in our sample. None were resistant to either vancomycin or trimethoprimsulfamethoxazole. A recent nationwide antibiogram using pooled data from pediatric hospitals in the United States reported a resistance to clindamycin in cases of staphylococcal infection of 21% [20]. In that report, staphylococcal isolates were at least 97% sensitive to trimethoprimsulfamethoxazole, vancomycin, and linezolid. Although clindamycin does provide coverage for streptococcus that trimethoprim-sulfamethoxazole lacks, its inferior coverage of MRSA limits its use for the empiric treatment of pediatric osteomyelitis. When the cost of linezolid is considered, vancomycin may be the most reasonable of these agents for the empiric treatment of pediatric osteomyelitis in previously healthy children. Infection in the femur was most common in our sample, followed by the tibia-fibula and the foot. This contrasts with the findings of Saavedra-Lozano and colleagues [6], who found infection in the foot more common than either the femur or tibia. Our 10-year study period began the same year that theirs concluded. We observed a higher combined rate of infection with MSSA and MRSA (85%) compared with their study (68%). A predilection of S aureus for the long bones could play a role in this discrepancy. Two case series from an institution in Glasgow encompassing a time range from 1970 to 1997, largely prior to the emergence of MRSA, saw a decreasing rate of pediatric osteomyelitis caused by S aureus tied to a correspondent decrease in infection of long bones [1,3]. It is possible that emergence of MRSA has caused a shift back to infection in the long bones. We observed a high rate of complications of osteomyelitis (69%) in our population. This is higher than rates reported in previous studies
Table 2 Local complications of osteomyelitis Complication
No. (%) (n = 75)
95% CI
Subperiosteal/Intraosseous abscess Myositis Soft tissue abscess Fasciitis
30 (40%) 26 (35%) 14 (19%) 5 (7%)
30%-51% 25%-46% 11%-29% 3%-15%
that defined complications similarly [4,6]. When compared with infection with MSSA, musculoskeletal infections with MRSA are known to be more severe and invasive [9,21]. Our high rate of local complications may reflect the high rate of MRSA infection in our sample. A higher proportion of complications were caused by MRSA compared with MSSA, although more of our cases of osteomyelitis were caused by MSSA. We found hematogenous complications in 3 patients, 2 of which had protracted illness courses related to infection with MRSA. Bouchoucha and colleagues [11] found a rate of DVT of 10% in cases of osteomyelitis in Tunisia. As in our study, all cases of DVT occurred in infection with staphylococcus, and more occurred in cases of MRSA. Crary and colleagues [5] reported a case series of 35 children in Texas with osteomyelitis, 10 of whom had a DVT. These authors included only children with osteomyelitis occurring in proximity to large veins (femur, tibia/fibula, pelvis, etc), which makes comparison with our study problematic. The rate of DVT observed in our study is likely a more accurate reflection of the overall rate of DVT in pediatric osteomyelitis. We observed one case of septic pulmonary embolus, a severe complication that has been described in cases of disseminated MRSA infection and DVT [4,12,22]. We also observed one case of endogenous endophthalmitis related to MRSA sepsis. This is an exceedingly rare condition in children that often has devastating effects on vision. Our subject's vitreal cultures were negative but were taken after the initiation of antibiotic therapy. A large review of cases of endogenous endopthalmitis reported that vitreal cultures were only positive in 56% of cases [23], as did a series of cases of pediatric endophthalmitis [24]. Endogenous endopthalmitis has been reported in the setting of a diabetic adult with foot osteomyelitis [25] as well as in the setting of neonatal MRSA sepsis [26]. The yield of operative cultures after therapeutic antibiotic administration was high in our study. This is likely a reflection of the time required to eradicate bacteria from bone. This finding suggests that antibiotics should not be withheld when operative cultures are planned, and that pursuing operative cultures when antibiotics have already been given can still play a valuable role in refining antibiotic therapy. We were surprised to find that time to therapeutic antibiotic treatment was shorter in subjects with complications. By nature of our retrospective study design, this represents an association that is not necessarily a causal relationship. This observation may be related to differences in bacteriology between the 2 groups. The rate of MRSA, which causes more severe infection [4,9,21], was higher in the complications group (49% vs 21%). Infections may have been clinically apparent, and hence received antibiotics, sooner in the complications group due to severity of illness. Time to antibiotic treatment in the noncomplication group may have been longer due to a more indolent course. 5. Limitations This study was a retrospective medical record review and relied on physician documentation to collect data, particularly related to the subject's presentation and medical history. It is possible that patients may have developed complications that were not treated at our institution, and thus, the number of complications may have been underestimated. We used ICD-9 codes for identification of cases for this study. It is possible that we missed additional cases of pediatric osteomyelitis at our institution that would have been of interest. The use of ICD-9 codes for identification of pediatric osteomyelitis has been used in multiple similar previous studies [4–6,8,13], but has not been validated to our knowledge. However, we did not rely solely on ICD-9 codes to define cases of pediatric osteomyelitis. We applied exclusion criteria to arrive at a sample of subjects who met a strict definition of osteomyelitis, as we were largely interested in identifying a sample of true positive cases of osteomyelitis. All 67 subjects in our study had a positive imaging test result or operative culture. We feel that it is unlikely that the number of cases in our institution who met our definition but were not assigned one of the related ICD-9 diagnostic codes was high.
Please cite this article as: Ratnayake K, et al, Pediatric acute osteomyelitis in the postvaccine, methicillin-resistant Staphylococcus aureus era, Am J Emerg Med (2015), http://dx.doi.org/10.1016/j.ajem.2015.07.011
K. Ratnayake et al. / American Journal of Emergency Medicine xxx (2015) xxx–xxx
5
Table 3 Hematogenous complications of osteomyelitis by case Age
Bone area
Organism identified
Hematogenous complication(s)
Clinical course
12 y
Right femur (with subperiosteal abscess)
MRSA
MRSA sepsis; decreased unilateral vision during hospitalization; necrosis and infiltrate of optic disk; vitrectomy done and intravitreous antibiotics given; vitreous culture negative
6y
Right humerus (with subperiosteal abscess) and right tibia Right humerus (with subperiosteal abscess)
MRSA
1. DVT (without central venous catheter) 2. Endophthalmitis 1. MRSA septic pulmonary nodule 2. Catheter-related DVT 1. Catheter-related DVT
7 wk
MSSA
MRSA sepsis; left cephalic vein thrombus around PICC; cavitary lesion on CT chest; respiratory culture positive for MRSA Left cephalic vein thrombosis around PICC line; hypercoagulability workup negative
CT, computed tomography; PICC, peripherally inserted central catheter.
6. Conclusions Our data suggest that treatment of pediatric osteomyelitis should include antibiotic coverage for MRSA. Clinicians should expect to find most cases of pediatric osteomyelitis in the long bones. Local complications are common, especially accompanying MRSA infection. Hematogenous complications may include DVT and may be related to treatment with a central venous catheter. Operative culture yield when antibiotics have already been given is high, and antibiotic treatment should not be delayed until operative cultures are obtained. Acknowledgments The authors wish to thank Alex Thomas for his assistance with data collection. References [1] Blyth MJ, Kincaid R, Craigen MA, Bennet GC. The changing epidemiology of acute and subacute haematogenous osteomyelitis in children. J Bone Joint Surg Br 2001;83: 99–102. [2] Karwowska A, Davies HD, Jadavji T. Epidemiology and outcome of osteomyelitis in the era of sequential intravenous-oral therapy. 1998;17:1021–6. [3] Craigen MA, Watters J, Hackett JS. The changing epidemiology of osteomyelitis in children. J Bone Joint Surg Br 1992;74:541–5. [4] Arnold SR, Elias D, Buckingham SC, Thomas ED, Novais E, Arkader A, et al. Changing patterns of acute hematogenous osteomyelitis and septic arthritis: emergence of community-associated methicillin-resistant Staphylococcus aureus. J Pediatr Orthop 2006;26:703–8. http://dx.doi.org/10.1097/01.bpo.0000242431.91489.b4. [5] Crary SE, Buchanan GR, Drake CE, Journeycake JM. Venous thrombosis and thromboembolism in children with osteomyelitis. J Pediatr 2006;149:537–41. http://dx.doi. org/10.1016/j.jpeds.2006.06.067. [6] Saavedra-Lozano J, Mejías A, Ahmad N, Peromingo E, Ardura MI, Guillen S, et al. Changing trends in acute osteomyelitis in children: impact of methicillin-resistant Staphylococcus aureus infections. J Pediatr Orthop 2008;28:569–75. http://dx.doi. org/10.1097/BPO.0b013e31817bb816. [7] Iwamoto M, Mu Y, Lynfield R, Bulens SN, Nadle J, Aragon D, et al. Trends in invasive methicillin-resistant Staphylococcus aureus infections. Pediatrics 2013;132:e817–24. http://dx.doi.org/10.1542/peds.2013-1112. [8] Gerber JS, Coffin SE, Smathers SA, Zaoutis TE. Trends in the incidence of methicillinresistant Staphylococcus aureus infection in children's hospitals in the United States. Clin Infect Dis 2009;49:65–71. http://dx.doi.org/10.1086/599348. [9] Hawkshead JJ, Patel NB, Steele RW, Heinrich SD. Comparative severity of pediatric osteomyelitis attributable to methicillin-resistant versus methicillin-sensitive Staphylococcus aureus. J Pediatr Orthop 2009;29:85–90. http://dx.doi.org/10.1097/ BPO.0b013e3181901c3a.
[10] Gonzalez BE. Venous thrombosis associated with staphylococcal osteomyelitis in children. Pediatrics 2006;117:1673–9. http://dx.doi.org/10.1542/peds.2005-2009. [11] Bouchoucha S, Benghachame F, Trifa M, Saied W, Douira W, Nessib MN, et al. Deep venous thrombosis associated with acute hematogenous osteomyelitis in children. Orthop Traumatol Surg Res 2010;96:890–3. http://dx.doi.org/10.1016/j.otsr.2010. 05.006. [12] Hollmig ST. Deep venous thrombosis associated with osteomyelitis in children. J Bone Joint Surg Am 2007;89:1517. http://dx.doi.org/10.2106/JBJS.F.01102. [13] Young TP, Maas L, Thorp AW, Brown L. Etiology of septic arthritis in children: an update for the new millennium. Am J Emerg Med 2011;29:899–902. http://dx.doi.org/ 10.1016/j.ajem.2010.04.008. [14] Burnett RSJ, Aggarwal A, Givens SA, McClure JT, Morgan PM, Barrack RL. Prophylactic antibiotics do not affect cultures in the treatment of an infected TKA: a prospective trial. Clin Orthop Relat Res 2009;468:127–34. http://dx.doi.org/10.1007/s11999009-1014-4. [15] Marschall J, Bhavan KP, Olsen MA, Fraser VJ, Wright NM, Warren DK. The impact of prebiopsy antibiotics on pathogen recovery in hematogenous vertebral osteomyelitis. Clin Infect Dis 2011;52:867–72. http://dx.doi.org/10.1093/cid/cir062. [16] Zhorne DJ, Altobelli ME, Cruz AT. Impact of antibiotic pretreatment on bone biopsy yield for children with acute hematogenous osteomyelitis. Hosp Pediatr 2015;5: 337–41. http://dx.doi.org/10.1542/hpeds.2014-0114. [17] Black JR, Cohen MS. Gonococcal osteomyelitis: a case report and review of the literature. Sex Transm Dis 1984;11:96–9. [18] Paley D, Moseley CF, Armstrong P, Prober CG. Primary osteomyelitis caused by coagulase-negative staphylococci. J Pediatr Orthop 1986;6:622–6. [19] Moore MR, Link-Gelles R, Schaffner W, Lynfield R, Lexau C, Bennett NM, et al. Effect of use of 13-valent pneumococcal conjugate vaccine in children on invasive pneumococcal disease in children and adults in the USA: analysis of multisite, population-based surveillance. Lancet Infect Dis 2015. http://dx.doi.org/10.1016/ S1473-3099(14)71081-3. [20] Tamma PD, Robinson GL, Gerber JS, Newland JG, DeLisle CM, Zaoutis TE, et al. Pediatric antimicrobial susceptibility trends across the United States. Infect Control Hosp Epidemiol 2013;34:1244–51. http://dx.doi.org/10.1086/673974. [21] Martinez-Aguilar G, Avalos-Mishaan A, Hulten K, Hammerman W, Mason Jr EO, Kaplan SL. Community-acquired, methicillin-resistant and methicillin-susceptible Staphylococcus aureus musculoskeletal infections in children. Pediatr Infect Dis J 2004;23:701–6. http://dx.doi.org/10.1097/01.inf.0000133044.79130.2a. [22] Nourse C, Starr M, Munckhof W. Community-acquired methicillin-resistant Staphylococcus aureus causes severe disseminated infection and deep venous thrombosis in children: literature review and recommendations for management. J Paediatr Child Health 2007;43:656–61. http://dx.doi.org/10.1111/j.1440-1754.2007.01153.x. [23] Jackson TL, Eykyn SJ, Graham EM, Stanford MR. Endogenous bacterial endophthalmitis: a 17-year prospective series and review of 267 reported cases. Surv Ophthalmol 2003; 48:403–23. http://dx.doi.org/10.1016/S0039-6257(03)00054-7. [24] Thordsen JE, Harris L, Hubbard GB. Pediatric endophthalmitis. A 10-year consecutive series. Retina 2008;28:S3–7. http://dx.doi.org/10.1097/IAE.0b013e318159ec7f. [25] Mavrakanas TA, de Haller R, Philippe J. Endogenous endophthalmitis in a patient with diabetes and foot osteomyelitis. Can J Diabetes 2015;39:18–20. http://dx.doi. org/10.1016/j.jcjd.2014.05.011. [26] Basu S, Kumar A, Kapoor K, Bagri NK, Chandra A. Neonatal endogenous endophthalmitis: a report of six cases. Pediatrics 2013;131:e1292–7. http://dx.doi.org/10.1542/peds. 2011-3391.
Please cite this article as: Ratnayake K, et al, Pediatric acute osteomyelitis in the postvaccine, methicillin-resistant Staphylococcus aureus era, Am J Emerg Med (2015), http://dx.doi.org/10.1016/j.ajem.2015.07.011
Contents lists available at ScienceDirect
American Journal of Emergency Medicine journal homepage: www.elsevier.com/locate/ajem
Original Contribution
Pediatric acute osteomyelitis in the postvaccine, methicillin-resistant Staphylococcus aureus era☆ Kristin Ratnayake, MD a, Andrew J. Davis, MD b, Lance Brown, MD, MPH b, Timothy P. Young, MD b,⁎ a b
Division of Pediatric Emergency Medicine, Department of Pediatrics, Rady Children's Hospital, San Diego, CA Division of Pediatric Emergency Medicine, Department of Emergency Medicine, Loma Linda University Medical Center and Children's Hospital, Loma Linda, CA
a r t i c l e
i n f o
Article history: Received 26 February 2015 Received in revised form 6 July 2015 Accepted 6 July 2015 Available online xxxx
a b s t r a c t Objective: We sought to describe the causative organisms, bones involved, and complications in cases of pediatric osteomyelitis in the postvaccine age and in the era of increasing infection with community-associated methicillin-resistant Staphylococcus aureus (MRSA). Methods: We reviewed the medical records of children 12 years and younger presenting to our pediatric emergency department between January 1, 2003, and December 31, 2012, with the diagnosis of osteomyelitis. We reviewed operative cultures, blood cultures, and imaging studies. We identified causative organisms, bone(s) involved, time to therapeutic antibiotic treatment, and local and hematogenous complications. Results: The most common organism identified was methicillin-sensitive S aureus (26/55), followed by MRSA (21/55). Seventy-three bone areas were affected in 67 subjects. The most common bone area was the femur (24/73). Forty-six subjects had 75 local complications. The most common organism in cases with local complications was MRSA (49%). Three subjects had hematogenous complications of deep venous thrombosis, septic pulmonary embolus, and endophthalmitis. Subjects with complications had shorter time to therapeutic antibiotic treatment. When an operative culture was done after therapeutic antibiotics were given, an organism was identified from the operative culture in 84% of cases. Conclusion: Treatment of pediatric osteomyelitis should include antibiotic coverage for MRSA. Most cases of pediatric osteomyelitis occur in the long bones. Hematogenous complications may include deep venous thrombosis and may be related to treatment with a central venous catheter. Operative culture yield when antibiotics have already been given is high, and antibiotic treatment should not be delayed until operative cultures are obtained. © 2015 Elsevier Inc. All rights reserved.
1. Introduction Pediatric hematogenous osteomyelitis typically has a prolonged treatment course involving multiple specialties. Emergency physicians may treat children with osteomyelitis at various stages in the illness, including at diagnosis, at recurrence of disease, and when complications of the disease or its treatment arise. Historically, the long bones have been the major bones affected in pediatric osteomyelitis, and Staphylococcus aureus has been the most common causative organism, responsible for 66% to 70% of cases [1,2]. Streptococcus pyogenes, Streptococcus pneumoniae, Haemophilus influenzae, and Pseudomonas aeruginosa have also been reported as causative agents [2–6]. Recently, increasing numbers of children with invasive infection caused by community-associated methicillin-resistant S aureus (MRSA) have been reported [7]. In the United States, rates of osteomyelitis caused by MRSA increased in children's hospitals during the first
☆ The authors have no financial or other conflicts of interest related to the submission. ⁎ Corresponding author at: Department of Emergency Medicine, Loma Linda University Medical Center, 11234 Anderson St, A-108, Loma Linda, CA 92354. E-mail address: [email protected] (T.P. Young).
decade of this century, whereas those caused by methicillin-sensitive S aureus (MSSA) remained constant [8]. Children with osteomyelitis caused by MRSA experience more severe illness than do their MSSA counterparts, with greater need for procedures and longer hospitalizations [4,9]. Complications recently attributed to MRSA osteomyelitis include deep venous thrombosis (DVT), as well as problems at sites distant to the primary infection such as endocarditis and septic pulmonary emboli [5,10–12]. It is possible that changes in bacteriology have affected the distribution of site of infection in osteomyelitis. Prior to the era of MRSA, Craigen and colleagues [3] in the United Kingdom found that long bone infection (femur and tibia-fibula) was most common, but saw a decrease in long bone infection rate mirroring a decrease in S aureus infection. Infection rates at other sites remained constant. More recently, SaavedraLozano and colleagues [6] found increasing rates of infection with MRSA in Texas and that infection in the foot was more common than either the femur or tibia-fibula. Because of the rapidly changing nature of pediatric osteomyelitis, a more current description of its epidemiology and complications would benefit clinicians who diagnose and treat children with the disease. We previously reported a high rate of MRSA in cases of septic arthritis at our institution [13] and became interested in examining
http://dx.doi.org/10.1016/j.ajem.2015.07.011 0735-6757/© 2015 Elsevier Inc. All rights reserved.
Please cite this article as: Ratnayake K, et al, Pediatric acute osteomyelitis in the postvaccine, methicillin-resistant Staphylococcus aureus era, Am J Emerg Med (2015), http://dx.doi.org/10.1016/j.ajem.2015.07.011
2
K. Ratnayake et al. / American Journal of Emergency Medicine xxx (2015) xxx–xxx
cases of osteomyelitis. At our institution, it is sometimes recommended that antibiotics be withheld until operative cultures are obtained in an effort to increase yield. Although adult studies suggest that preoperative antibiotics do not affect operative culture yield, there are less available data in the pediatric setting [14–16]. Our objective was to examine cases of acute culture- and/or imaging-proven osteomyelitis involving healthy prepubescent children seen in our pediatric emergency department. Our primary aim was to determine the rate of MRSA infection in cases of pediatric osteomyelitis. Our secondary aims were to determine organism susceptibilities for MRSA osteomyelitis, to describe site of involvement and rate of complications of osteomyelitis, to determine whether complications were associated with antibiotic discordance or time to antibiotic administration, and to evaluate the yield of operative cultures when antibiotics had already been administered.
was given in cases where there was no positive culture obtained. We also recorded the day of illness that an operative culture was obtained. We excluded children with repeat visits, chronic osteomyelitis, a history of previous surgery on the affected bone, and chronic illness. We defined chronic illness as the presence of a ventriculoperitoneal shunt, the presence of an indwelling central venous catheter, and conditions such as sickle cell disease, spina bifida, extreme prematurity, complex congenital cardiac lesions, and cancer. Descriptive statistics were calculated using Microsoft Excel 2010 (Microsoft Corporation, Redmond, WA). Where 95% confidence intervals (CIs) are reported, the modified Wald (Agresti-Coull) method was used and was calculated using STATA IC 12 (Statacorp LP, College Station, TX). Subjects with complications were compared with those without complications with respect to both time to therapeutic antibiotic treatment and antibiotic discordance using a Mann-Whitney U test and a χ 2 test, respectively. Both were calculated in STATA, with an α level set at .05.
2. Materials and methods
3. Results
We performed a retrospective medical record review of children 12 years and younger presenting to the pediatric emergency department at our university-based, tertiary care institution between January 1, 2003, and December 31, 2012, with the diagnosis of osteomyelitis. We chose this age group to reflect the epidemiology and pathogenesis of a largely prepubertal sample of children. Gonococcal osteomyelitis has been reported in adolescents, and we aimed to avoid this entity [17]. Subjects were identified by the Ninth Revision of the International Classification of Diseases (ICD-9) diagnostic codes for osteomyelitis (730.00-730.99). We used a standardized data collection form. Data were collected by a trained data abstractor. The data extracted from the medical records were the patient's age, sex, bone(s) involved, presence of comorbidities, previous diagnosis of bone infection, antibiotics given, results of bone aspirate culture and blood culture, and results of advanced imaging studies. A positive evaluation for osteomyelitis was defined by the presence of positive operative or blood culture for a known pathogen or advanced imaging studies indicating the presence of osteomyelitis. A known pathogen was considered to be any organism other than diptheroids. We included coagulase-negative staphylococcus (CONS) as a pathogen, as until 2008 our laboratory did not speciate CONS. Coagulase-negative staphylococcus has also been reported as a rare cause of osteomyelitis in healthy children [18]. We defined the presence of osteomyelitis on advanced imaging studies as a positive radiologic result for magnetic resonance imaging or nuclear medicine bone scan of the affected area. We defined the following as possible sites of infection: the hand, radius/ulna, humerus, scapula, foot, tibia/fibula, patella, femur, pelvis, spine, and cranium. We reviewed all cases of osteomyelitis for local and hematogenous complications. We defined local complications as abscess, myositis, or fasciitis seen on imaging studies or described in operative reports. We defined hematogenous complications as DVT, septic pulmonary emboli, or other infections at secondary locations attributed to the primary infection of osteomyelitis. Medical records were reviewed from initial presentation to the time of data collection for complications. We included DVTs related to catheters placed for long-term intravenous antibiotics as well as those without mechanical inciting factors. We reviewed cases with a pathogen-positive culture for antibiotic discordance. We defined antibiotic discordance as culture positive for one of the following: an organism explicitly resistant to the antibiotic(s) given prior to the culture result, an organism resistant to an antibiotic of the same class (eg, nafcillin and oxacillin), or an organism resistant to the antibiotic(s) administered according to the Sanford Guide in the year that the subject was treated. Our laboratory does not report sensitivities for all organisms and does not test all antibiotics. We recorded the day of illness that therapeutic antibiotics were given. We defined this as the day of illness at which an antibiotic was given that the organism was sensitive to in cases where an organism was identified, or the day of illness at which the first antibiotic
We identified 194 patient encounters with an ICD-9 code for osteomyelitis during the 10-year period. All medical records were complete. A total of 102 encounters met our exclusion criteria. This resulted in a group of 92 subjects who were evaluated for osteomyelitis. Twentyfive subjects had a negative evaluation for osteomyelitis. Therefore, our main study group consisted of the remaining 67 subjects (Fig. 1). After our exclusion criteria were applied to identify cases of interest for this study, the use of ICD-9 codes had a 73% (67/92) positive predictive value for osteomyelitis as we defined it. All included subjects had either a positive imaging study or a positive operative culture. Forty subjects (60%) were male. The median age was 5 years (interquartile range [IQR], 2-10 years; range, 1 month to 12 years; Fig. 2). The most common pathogenic organism was MSSA, followed by MRSA (Table 1). The most common organism recovered by blood culture was MRSA (12/23; 52%), followed by MSSA (9/23; 39%). All isolates of MRSA were sensitive to vancomycin and trimethoprim-sulfamethoxazole. Two were resistant to clindamycin. The most common site was the femur (Fig. 3). Forty-six subjects (69%) had a total of 75 local complications (Table 2). Three subjects had hematogenous complications (Table 3). Two of the subjects with hematogenous complications had infection with MRSA. All subjects with hematogenous complications also had local complications. For cases with complications in which the organism was identified, the most common organism was MRSA (19/39; 49%), followed by MSSA (16/39,; 41%). Subjects with complications had a median time to therapeutic antibiotics of 6 days (IQR, 4-9 days), whereas those without complications had a median time to therapeutic antibiotics of 8 days (IQR, 6-22 days; P = .03). For subjects without complications in which the organism was identified, the most common organism was MSSA (8/14; 57%), followed by MRSA (3/14; 21%). Subjects with and without complications did not differ regarding antibiotic discordance (P = .50). When an operative culture was done after therapeutic antibiotics were given, an organism was still identified from the operative culture in 42 (84%) of 50 cases (95% CI, 71%-92%). An organism was identified from operative cultures when therapeutic antibiotics had not been given prior to operative cultures in 4 (80%) of 5 cases (95% CI, 36%-98%). 4. Discussion We found a 38% rate of infection with MRSA in cases of pediatric osteomyelitis at our institution, nearly equal to that of MSSA. Pediatric osteomyelitis most commonly affected the long bones. A high proportion of our subjects had complications (69%). Subjects with complications had a shorter time to therapeutic antibiotic administration. Operative culture yield was 84% when therapeutic cultures had already been given.
Please cite this article as: Ratnayake K, et al, Pediatric acute osteomyelitis in the postvaccine, methicillin-resistant Staphylococcus aureus era, Am J Emerg Med (2015), http://dx.doi.org/10.1016/j.ajem.2015.07.011
K. Ratnayake et al. / American Journal of Emergency Medicine xxx (2015) xxx–xxx
3
ICD-9 code for osteomyelitis (n = 194)
Previous diagnosis of osteomyelitis (n = 34)
Mastoiditis (n = 3)
Previous surgery or open fracture to bone (n = 6)
Chronic illness (n = 59)*
Negative evaluation for osteomyelitis (n = 25)
Positive evaluation for osteomyelitis (n = 67)
Positive operative culture, blood culture and imaging (n = 16) Positive operative culture and blood culture (n = 1) Positive operative culture and imaging (n = 24) Positive blood culture and imaging (n = 6) Positive operative culture only (n = 6) Positive imaging only (n = 14) Fig. 1. Subject flow diagram. *We defined chronic illness as the presence of a ventriculoperitoneal shunt, the presence of an indwelling central venous catheter, or conditions such as sickle cell disease, spina bifida, extreme prematurity, complex congenital cardiac lesions and cancer.
7 6 4
Organism
No. (%) of isolatesa (n = 55)
95% CI
3
5
Table 1 Pathogenic organism isolated
MSSA Community-associated MRSA β-Hemolytic group A Streptococcus CONS Streptococcus viridans Enterococcus faecium Mycobacterium chelonae S pneumoniae
26 (47%) 21 (38%) 3 (5%) 1 (2%) 1 (2%) 1 (2%) 1 (2%) 1 (2%)
35%-60% 26%-51% 1%-15% 0%-11% 0%-11% 0%-11% 0%-11% 0%-11%
0
1
2
Number of Subjects
8
9
10
We observed a higher proportion of MRSA infection in our study compared with a similar study ranging from 1999 to 2003 [6] and another comprising a time range from 2000 to 2004 [4]. This is consistent with multiple reports of a continuing uptrend in the incidence of MRSA infection in children [7,8]. S pneumoniae was identified as the cause of one case of osteomyelitis in our sample. This is consistent with the low but continued presence of S pneumoniae in cases of pediatric
0
1
2
3
4
5
6
7
8
9
10
Age (years) Fig. 2. Age distribution for children with osteomyelitis.
11
12
a There were 55 isolates in 53 subjects. One subject had MSSA and CONS, and another subject had MSSA and S viridans.
Please cite this article as: Ratnayake K, et al, Pediatric acute osteomyelitis in the postvaccine, methicillin-resistant Staphylococcus aureus era, Am J Emerg Med (2015), http://dx.doi.org/10.1016/j.ajem.2015.07.011
4
K. Ratnayake et al. / American Journal of Emergency Medicine xxx (2015) xxx–xxx
Scapula 4/73 (6%; 95% CI 2-14%)
Cranium 2/73 (3%; 95% CI 0-10%)
Humerus 4/73 (6%; 95% CI 2-14%) Spine 1/73 (1%; 95% CI 0-8%) Radius/Ulna 3/73 (4%; 95% CI 1-12%) Pelvis 6/73 (8%; 95% CI 4-17%)
Patella 1/73 (1%; 95% CI 0-8%)
Femur 24/73 (33%; 95% CI 23-43%)
Tibia-Fibula 20/73 (27%; 95% CI 18-39%)
Foot 8/73 (11%; 95% CI 5-20%)
Fig. 3. Distribution of affected body area in osteomyelitis. Four subjects had infection in 2 bone areas and 1 subject had infection in 3 bone areas, for a total of 73 bone areas in 67 subjects.
osteomyelitis that has been seen since the implementation of the heptavalent pneumococcal conjugate vaccine in the United States in 2001 [4,6]. This incidence may be expected to fall further, as overall rates of pneumococcal infection have declined in the United States since the introduction of the 13-valent pneumococcal vaccine in 2010 [19]. Two of 20 infections with MRSA were resistant to clindamycin in our sample. None were resistant to either vancomycin or trimethoprimsulfamethoxazole. A recent nationwide antibiogram using pooled data from pediatric hospitals in the United States reported a resistance to clindamycin in cases of staphylococcal infection of 21% [20]. In that report, staphylococcal isolates were at least 97% sensitive to trimethoprimsulfamethoxazole, vancomycin, and linezolid. Although clindamycin does provide coverage for streptococcus that trimethoprim-sulfamethoxazole lacks, its inferior coverage of MRSA limits its use for the empiric treatment of pediatric osteomyelitis. When the cost of linezolid is considered, vancomycin may be the most reasonable of these agents for the empiric treatment of pediatric osteomyelitis in previously healthy children. Infection in the femur was most common in our sample, followed by the tibia-fibula and the foot. This contrasts with the findings of Saavedra-Lozano and colleagues [6], who found infection in the foot more common than either the femur or tibia. Our 10-year study period began the same year that theirs concluded. We observed a higher combined rate of infection with MSSA and MRSA (85%) compared with their study (68%). A predilection of S aureus for the long bones could play a role in this discrepancy. Two case series from an institution in Glasgow encompassing a time range from 1970 to 1997, largely prior to the emergence of MRSA, saw a decreasing rate of pediatric osteomyelitis caused by S aureus tied to a correspondent decrease in infection of long bones [1,3]. It is possible that emergence of MRSA has caused a shift back to infection in the long bones. We observed a high rate of complications of osteomyelitis (69%) in our population. This is higher than rates reported in previous studies
Table 2 Local complications of osteomyelitis Complication
No. (%) (n = 75)
95% CI
Subperiosteal/Intraosseous abscess Myositis Soft tissue abscess Fasciitis
30 (40%) 26 (35%) 14 (19%) 5 (7%)
30%-51% 25%-46% 11%-29% 3%-15%
that defined complications similarly [4,6]. When compared with infection with MSSA, musculoskeletal infections with MRSA are known to be more severe and invasive [9,21]. Our high rate of local complications may reflect the high rate of MRSA infection in our sample. A higher proportion of complications were caused by MRSA compared with MSSA, although more of our cases of osteomyelitis were caused by MSSA. We found hematogenous complications in 3 patients, 2 of which had protracted illness courses related to infection with MRSA. Bouchoucha and colleagues [11] found a rate of DVT of 10% in cases of osteomyelitis in Tunisia. As in our study, all cases of DVT occurred in infection with staphylococcus, and more occurred in cases of MRSA. Crary and colleagues [5] reported a case series of 35 children in Texas with osteomyelitis, 10 of whom had a DVT. These authors included only children with osteomyelitis occurring in proximity to large veins (femur, tibia/fibula, pelvis, etc), which makes comparison with our study problematic. The rate of DVT observed in our study is likely a more accurate reflection of the overall rate of DVT in pediatric osteomyelitis. We observed one case of septic pulmonary embolus, a severe complication that has been described in cases of disseminated MRSA infection and DVT [4,12,22]. We also observed one case of endogenous endophthalmitis related to MRSA sepsis. This is an exceedingly rare condition in children that often has devastating effects on vision. Our subject's vitreal cultures were negative but were taken after the initiation of antibiotic therapy. A large review of cases of endogenous endopthalmitis reported that vitreal cultures were only positive in 56% of cases [23], as did a series of cases of pediatric endophthalmitis [24]. Endogenous endopthalmitis has been reported in the setting of a diabetic adult with foot osteomyelitis [25] as well as in the setting of neonatal MRSA sepsis [26]. The yield of operative cultures after therapeutic antibiotic administration was high in our study. This is likely a reflection of the time required to eradicate bacteria from bone. This finding suggests that antibiotics should not be withheld when operative cultures are planned, and that pursuing operative cultures when antibiotics have already been given can still play a valuable role in refining antibiotic therapy. We were surprised to find that time to therapeutic antibiotic treatment was shorter in subjects with complications. By nature of our retrospective study design, this represents an association that is not necessarily a causal relationship. This observation may be related to differences in bacteriology between the 2 groups. The rate of MRSA, which causes more severe infection [4,9,21], was higher in the complications group (49% vs 21%). Infections may have been clinically apparent, and hence received antibiotics, sooner in the complications group due to severity of illness. Time to antibiotic treatment in the noncomplication group may have been longer due to a more indolent course. 5. Limitations This study was a retrospective medical record review and relied on physician documentation to collect data, particularly related to the subject's presentation and medical history. It is possible that patients may have developed complications that were not treated at our institution, and thus, the number of complications may have been underestimated. We used ICD-9 codes for identification of cases for this study. It is possible that we missed additional cases of pediatric osteomyelitis at our institution that would have been of interest. The use of ICD-9 codes for identification of pediatric osteomyelitis has been used in multiple similar previous studies [4–6,8,13], but has not been validated to our knowledge. However, we did not rely solely on ICD-9 codes to define cases of pediatric osteomyelitis. We applied exclusion criteria to arrive at a sample of subjects who met a strict definition of osteomyelitis, as we were largely interested in identifying a sample of true positive cases of osteomyelitis. All 67 subjects in our study had a positive imaging test result or operative culture. We feel that it is unlikely that the number of cases in our institution who met our definition but were not assigned one of the related ICD-9 diagnostic codes was high.
Please cite this article as: Ratnayake K, et al, Pediatric acute osteomyelitis in the postvaccine, methicillin-resistant Staphylococcus aureus era, Am J Emerg Med (2015), http://dx.doi.org/10.1016/j.ajem.2015.07.011
K. Ratnayake et al. / American Journal of Emergency Medicine xxx (2015) xxx–xxx
5
Table 3 Hematogenous complications of osteomyelitis by case Age
Bone area
Organism identified
Hematogenous complication(s)
Clinical course
12 y
Right femur (with subperiosteal abscess)
MRSA
MRSA sepsis; decreased unilateral vision during hospitalization; necrosis and infiltrate of optic disk; vitrectomy done and intravitreous antibiotics given; vitreous culture negative
6y
Right humerus (with subperiosteal abscess) and right tibia Right humerus (with subperiosteal abscess)
MRSA
1. DVT (without central venous catheter) 2. Endophthalmitis 1. MRSA septic pulmonary nodule 2. Catheter-related DVT 1. Catheter-related DVT
7 wk
MSSA
MRSA sepsis; left cephalic vein thrombus around PICC; cavitary lesion on CT chest; respiratory culture positive for MRSA Left cephalic vein thrombosis around PICC line; hypercoagulability workup negative
CT, computed tomography; PICC, peripherally inserted central catheter.
6. Conclusions Our data suggest that treatment of pediatric osteomyelitis should include antibiotic coverage for MRSA. Clinicians should expect to find most cases of pediatric osteomyelitis in the long bones. Local complications are common, especially accompanying MRSA infection. Hematogenous complications may include DVT and may be related to treatment with a central venous catheter. Operative culture yield when antibiotics have already been given is high, and antibiotic treatment should not be delayed until operative cultures are obtained. Acknowledgments The authors wish to thank Alex Thomas for his assistance with data collection. References [1] Blyth MJ, Kincaid R, Craigen MA, Bennet GC. The changing epidemiology of acute and subacute haematogenous osteomyelitis in children. J Bone Joint Surg Br 2001;83: 99–102. [2] Karwowska A, Davies HD, Jadavji T. Epidemiology and outcome of osteomyelitis in the era of sequential intravenous-oral therapy. 1998;17:1021–6. [3] Craigen MA, Watters J, Hackett JS. The changing epidemiology of osteomyelitis in children. J Bone Joint Surg Br 1992;74:541–5. [4] Arnold SR, Elias D, Buckingham SC, Thomas ED, Novais E, Arkader A, et al. Changing patterns of acute hematogenous osteomyelitis and septic arthritis: emergence of community-associated methicillin-resistant Staphylococcus aureus. J Pediatr Orthop 2006;26:703–8. http://dx.doi.org/10.1097/01.bpo.0000242431.91489.b4. [5] Crary SE, Buchanan GR, Drake CE, Journeycake JM. Venous thrombosis and thromboembolism in children with osteomyelitis. J Pediatr 2006;149:537–41. http://dx.doi. org/10.1016/j.jpeds.2006.06.067. [6] Saavedra-Lozano J, Mejías A, Ahmad N, Peromingo E, Ardura MI, Guillen S, et al. Changing trends in acute osteomyelitis in children: impact of methicillin-resistant Staphylococcus aureus infections. J Pediatr Orthop 2008;28:569–75. http://dx.doi. org/10.1097/BPO.0b013e31817bb816. [7] Iwamoto M, Mu Y, Lynfield R, Bulens SN, Nadle J, Aragon D, et al. Trends in invasive methicillin-resistant Staphylococcus aureus infections. Pediatrics 2013;132:e817–24. http://dx.doi.org/10.1542/peds.2013-1112. [8] Gerber JS, Coffin SE, Smathers SA, Zaoutis TE. Trends in the incidence of methicillinresistant Staphylococcus aureus infection in children's hospitals in the United States. Clin Infect Dis 2009;49:65–71. http://dx.doi.org/10.1086/599348. [9] Hawkshead JJ, Patel NB, Steele RW, Heinrich SD. Comparative severity of pediatric osteomyelitis attributable to methicillin-resistant versus methicillin-sensitive Staphylococcus aureus. J Pediatr Orthop 2009;29:85–90. http://dx.doi.org/10.1097/ BPO.0b013e3181901c3a.
[10] Gonzalez BE. Venous thrombosis associated with staphylococcal osteomyelitis in children. Pediatrics 2006;117:1673–9. http://dx.doi.org/10.1542/peds.2005-2009. [11] Bouchoucha S, Benghachame F, Trifa M, Saied W, Douira W, Nessib MN, et al. Deep venous thrombosis associated with acute hematogenous osteomyelitis in children. Orthop Traumatol Surg Res 2010;96:890–3. http://dx.doi.org/10.1016/j.otsr.2010. 05.006. [12] Hollmig ST. Deep venous thrombosis associated with osteomyelitis in children. J Bone Joint Surg Am 2007;89:1517. http://dx.doi.org/10.2106/JBJS.F.01102. [13] Young TP, Maas L, Thorp AW, Brown L. Etiology of septic arthritis in children: an update for the new millennium. Am J Emerg Med 2011;29:899–902. http://dx.doi.org/ 10.1016/j.ajem.2010.04.008. [14] Burnett RSJ, Aggarwal A, Givens SA, McClure JT, Morgan PM, Barrack RL. Prophylactic antibiotics do not affect cultures in the treatment of an infected TKA: a prospective trial. Clin Orthop Relat Res 2009;468:127–34. http://dx.doi.org/10.1007/s11999009-1014-4. [15] Marschall J, Bhavan KP, Olsen MA, Fraser VJ, Wright NM, Warren DK. The impact of prebiopsy antibiotics on pathogen recovery in hematogenous vertebral osteomyelitis. Clin Infect Dis 2011;52:867–72. http://dx.doi.org/10.1093/cid/cir062. [16] Zhorne DJ, Altobelli ME, Cruz AT. Impact of antibiotic pretreatment on bone biopsy yield for children with acute hematogenous osteomyelitis. Hosp Pediatr 2015;5: 337–41. http://dx.doi.org/10.1542/hpeds.2014-0114. [17] Black JR, Cohen MS. Gonococcal osteomyelitis: a case report and review of the literature. Sex Transm Dis 1984;11:96–9. [18] Paley D, Moseley CF, Armstrong P, Prober CG. Primary osteomyelitis caused by coagulase-negative staphylococci. J Pediatr Orthop 1986;6:622–6. [19] Moore MR, Link-Gelles R, Schaffner W, Lynfield R, Lexau C, Bennett NM, et al. Effect of use of 13-valent pneumococcal conjugate vaccine in children on invasive pneumococcal disease in children and adults in the USA: analysis of multisite, population-based surveillance. Lancet Infect Dis 2015. http://dx.doi.org/10.1016/ S1473-3099(14)71081-3. [20] Tamma PD, Robinson GL, Gerber JS, Newland JG, DeLisle CM, Zaoutis TE, et al. Pediatric antimicrobial susceptibility trends across the United States. Infect Control Hosp Epidemiol 2013;34:1244–51. http://dx.doi.org/10.1086/673974. [21] Martinez-Aguilar G, Avalos-Mishaan A, Hulten K, Hammerman W, Mason Jr EO, Kaplan SL. Community-acquired, methicillin-resistant and methicillin-susceptible Staphylococcus aureus musculoskeletal infections in children. Pediatr Infect Dis J 2004;23:701–6. http://dx.doi.org/10.1097/01.inf.0000133044.79130.2a. [22] Nourse C, Starr M, Munckhof W. Community-acquired methicillin-resistant Staphylococcus aureus causes severe disseminated infection and deep venous thrombosis in children: literature review and recommendations for management. J Paediatr Child Health 2007;43:656–61. http://dx.doi.org/10.1111/j.1440-1754.2007.01153.x. [23] Jackson TL, Eykyn SJ, Graham EM, Stanford MR. Endogenous bacterial endophthalmitis: a 17-year prospective series and review of 267 reported cases. Surv Ophthalmol 2003; 48:403–23. http://dx.doi.org/10.1016/S0039-6257(03)00054-7. [24] Thordsen JE, Harris L, Hubbard GB. Pediatric endophthalmitis. A 10-year consecutive series. Retina 2008;28:S3–7. http://dx.doi.org/10.1097/IAE.0b013e318159ec7f. [25] Mavrakanas TA, de Haller R, Philippe J. Endogenous endophthalmitis in a patient with diabetes and foot osteomyelitis. Can J Diabetes 2015;39:18–20. http://dx.doi. org/10.1016/j.jcjd.2014.05.011. [26] Basu S, Kumar A, Kapoor K, Bagri NK, Chandra A. Neonatal endogenous endophthalmitis: a report of six cases. Pediatrics 2013;131:e1292–7. http://dx.doi.org/10.1542/peds. 2011-3391.
Please cite this article as: Ratnayake K, et al, Pediatric acute osteomyelitis in the postvaccine, methicillin-resistant Staphylococcus aureus era, Am J Emerg Med (2015), http://dx.doi.org/10.1016/j.ajem.2015.07.011