Institutional members access full text with Ovid®

Share this article on:

01241398-201012000-0002101241398_2010_30_883_courtney_osteomyelitis_8miscellaneous< 55_0_7_3 >Journal of Pediatric Orthopaedics© 2010 Lippincott Williams & Wilkins, Inc.Volume 30(8)December 2010p 883–887Clinical Indications for Repeat MRI in Children With Acute Hematogenous Osteomyelitis[Infection]Courtney, Paul Maxwell BS, BA; Flynn, John M. MD; Jaramillo, Diego MD; Horn, Bernard David MD; Calabro, Kristen BS; Spiegel, David A. MDChildren's Hospital of Philadelphia and the University of Pennsylvania School of Medicine, Philadelphia, PANone of the authors received financial support for this study.Reprints: David A. Spiegel, MD, Division of Orthopaedic Surgery, Childrens Hospital of Philadelphia, 2nd Floor Wood Building, 34th Street and Civic Center Blvd. Philadelphia, PA 19104. e-mail: At our center and at others, some children with acute hematogenous osteomyelitis (AHO) are evaluated with multiple magnetic resonance imagings (MRIs) during their treatment. Do these serial MRI studies have a role in the management of AHO? We examine several clinical indications for ordering a repeat MRI and whether the imaging study resulted in a change in management.Methods: A total of 59 children (60 cases) with AHO were imaged with more than 1 MRI. We retrospectively reviewed the MRI studies and hospital records to investigate whether the results of the MRIs prompted a change in clinical management (surgical exploration or drainage, biopsy, change in the course of antibiotics). We investigated several clinical indicators including C-reactive protein (CRP) levels, time since index study, anatomic location of infection, and blood cultures. Differences in the proportion of patients with specified clinical characteristics, whose repeat MRI resulted in a change in management, were assessed by the use of χ2 analysis.Results: The median age of our patient population was 8.4 years; a total of 104 repeat MRI studies were undertaken on 59 children. Eleven (10.6%) of these studies prompted a change in patient treatment. Statistically significant indications for repeat MRI in changing clinical management included failure to improve clinically in 10 studies (21%, P<0.001), persistently elevated or increasing CRP levels in 11 MRI studies (52%, P<0.001), and the repeat study occurring within 14 days of the diagnostic MRI in 8 studies (29%, P=0.003).Conclusions: The results of our study showed that additional MRI studies provide information that affected patient management in only a limited number of cases. Although repeat MRI does not have a role in routine surveillance in children with AHO undergoing treatment, it can be a useful adjunct to clinical evaluation in patients who do not respond to therapy or who have a persistently elevated CRP level.Level of Evidence: Level IV, Therapeutic Study.The diagnosis of acute hematogenous osteomyelitis (AHO) is based on clinical findings, supplemented by laboratory investigations and imaging studies. Although early diagnosis and prompt treatment lead to resolution in the vast majority of cases, a subset of patients do not respond to standard treatment and may develop complications such as chronic osteomyelitis. Magnetic resonance imaging (MRI) plays an important role in the diagnosis of AHO, as pathologic changes on plain radiographs are not evident in the early stages of disease.1 Typical findings in the bone marrow include hypointensity on T1-weighted images, hyperintensity on T2-weighted images, and abnormal signal enhancement after gadolinium administration.2 MRI is both highly sensitive and specific, with reported values ranging from 88% to 100% and 75% to 100%, respectively.3,4 In addition to directing biopsy and assessing viability and blood flow to the infected bone and tissues, MRI is also useful in defining the degree of soft tissue involvement, identifying both bone and soft tissue abscesses, and showing coexisting articular pathology such as effusion or synovitis.5,6Over the past several years, we have observed that a number of AHO patients at our institution were evaluated with additional MRI studies during the course of their treatment. Although the most common indication was an inadequate clinical response during treatment, many studies were obtained during routine follow-up. Once the diagnosis has been confirmed, the value of additional MRI studies in the management of AHO remains unclear, and has yet to be addressed in the literature. In this retrospective case series, we investigated whether repeat MRI showed any new findings and which clinical indications for these additional studies prompted a change in patient management.METHODSThe inclusion criteria for this study were patients between birth and 18 years of age who were diagnosed with AHO and had been imaged with more than 1 MRI scan between January 1, 2000 and January 1, 2008. International Review Board approval was obtained. Using the International Classification of Disease, Ninth Revision code, we identified a total of 567 patients who were diagnosed with AHO at the Children's Hospital of Philadelphia during this period. One hundred thirty-two patients were evaluated with 1 MRI study, whereas 68 had more than 1 MRI during their course of treatment. From this smaller group, we identified 59 patients (60 cases) who met the inclusion criteria. Patients were excluded from the study if they were more than 18 years of age at the time of the index MRI (3 patients), had other pathology complicating the imaging studies such as a decubitus ulcer (2 patients), or more than 1 year elapsed after the index MRI (4 patients). One patient was treated for AHO in 2 separate locations at 2 different times. We retrospectively reviewed the imaging studies and medical records for these 60 cases of AHO in 59 patients.The MRI reports were reviewed with respect to the following parameters: presence and type of signal change noted in the bone marrow, presence or absence of an intraosseous or subperiosteal abscess, joint effusion (or disk space involvement in spine studies), and soft tissue involvement (myositis, cellulitis, and fasciitis). For each subsequent MRI, we recorded whether there were any new findings and tracked changes in the aforementioned parameters. The hospital chart and outpatient orthopaedic notes of each patient were also reviewed to collect the indications for obtaining additional MRIs, organism cultured, surgical treatment, and response to treatment. All patients were followed by the orthopaedic clinic until resolution of their infection. We also followed C-reactive protein (CRP) levels throughout the course of treatment for each patient. At the time of each subsequent MRI, the CRP level of the patient was recorded as trending upward, downward, at an elevated plateau (no change), or normal (<1.0 mg/L). The findings of the radiologist were then correlated with the clinical data for each patient to determine whether the treatment course was altered based on the repeat MRI results.We determined that a change in clinical management occurred if after the repeat MRI study, the patient underwent a surgical procedure (biopsy, surgical exploration, or drainage) or a change in the course of antibiotics. Correlations between MRI findings and CRP levels, clinical indication for ordering the study, and time since the index MRI were investigated. Differences in the proportion of patients with the specified clinical characteristics whose repeat MRI resulted in a change in management were assessed by the use of χ2 analysis.RESULTSThe study population included 59 patients (43 boys, 16 girls) ranging in age from 2 months to 18 years (median 8.4 y). A total of 164 MRI examinations (range 2 to 7 studies) were ordered; 30 patients (50%) were imaged with 2 MRI examinations, 20 (33%) had 3 MRI examinations, and 8 (13%) patients had 4 MRI examinations. One patient was studied with 5 MRI examinations whereas another patient had 7 MRI examinations. Repeat MRI studies were ordered by pediatricians (58 studies), infectious disease specialists (15), physical medicine and rehabilitation specialists (3), and orthopaedic surgeons (28).The diagnostic MRI showed a hyperintense T2-weighted signal, a corresponding hypointense T1-weighted signal, and marrow enhancement after gadolinium administration, in all 60 cases. Additional findings on the index MRI are summarized in Table 1. New findings were observed in 17 (16%) repeat MRI studies, and included joint effusion (8), bone signal abnormality (5), abscess (4), and avascular necrosis (2). Eight of those 17 studies, prompted a change in treatment plan. Table 2 shows the changes in imaging findings from the index study. Eleven (10.6%) of the 104 additional MRIs were followed by a change in clinical management. Ten of these patients had symptoms that persisted or were worsening. In 7 of these 11 cases, either a new abscess (2) or a worsening abscess (5) prompted surgical exploration or drainage, and in 1 case a biopsy was done. In 3 cases, a change in antibiotics was recommended because the patient was not responding clinically. Twenty-eight patients (47%) required surgical intervention at some point during their treatment course.TABLE 1. Diagnostic MRI Findings of 60 Pediatric Patients With Acute Hematogenous OsteomyelitisTABLE 2. A Summary of Signal Changes in the 104 Repeat MRI StudiesThe indications for additional MRIs (n=104) included clinical factors (worsening or persistent symptoms, increased inflammatory markers) in 47 studies (45%), and as part of routine follow-up from the infectious disease clinic (43 cases; 41%). Other reasons included insufficient contrast on the index study and follow-up after abscess decompression. Compared with the 57 MRIs ordered for routine follow-up, post abscess decompression, or insufficient contrast on the index study, a change in management occurred after 10 of the 47 MRI studies (21%, P<0.001) ordered because of the persistent or worsening clinical course of the patient.Of the 104 subsequent MRI studies, 28 were conducted during the patients' hospital stay within 2 weeks of the index study. All of these studies were ordered because of a worsening clinical course. In children whose additional MRI occurred within 14 days of the diagnostic study, a change in treatment occurred after 8 of the 28 MRIs (29%, P=0.003) compared with management changes in 3 of 76 MRIs done after 15 days of the index study.Of the 11 MRI studies that resulted in a change in the treatment of the patient, the CRP levels were increasing in 7 and were elevated or unchanged in 4 patients. None reported decreasing or normal CRP levels. In patients with persistently elevated or increasing CRP levels, a change in clinical management occurred after 11 of the 21 MRI studies (52%, P<0.001) compared with 0 of the 93 studies in children with normal or decreasing CRP levels, which were not associated with a change in the treatment plan.Staphylococcus aureus was the most common causal organism, isolated in 23 children. Eight of those 23 children had methicillin-resistant S. aureus. Additional MRI studies did not have a statistically significant correlation between the results of the blood cultures and a change in treatment (P=0.338). Twenty-eight patients (47%) were treated surgically through biopsy or surgical exploration or drainage. Eight of those patients (29%) had a change in clinical management after a repeat MRI study compared with 3 patients (9%) who were just treated medically; however, this result was not statistically significant (P=0.058). The findings of the correlations between these clinical indicators and whether the repeat MRI resulted in a change in management are summarized in Table 3.TABLE 3. Factors Related to Imaging Studies Which Prompted a Change in Treatment PlanDISCUSSIONMRI plays an important role in the diagnosis of osteomyelitis, especially during the early phase of the disease before the development of plain radiographic findings. Our findings on the diagnostic studies show that MRI can help in clinical management by discerning abscesses (28%), identifying abnormalities in the surrounding soft tissues (87%), and diagnosing coexisting articular abnormalities such as effusion or synovitis (38%). It is interesting to note that the abscesses were only seen in 19% of axial cases, in comparison with 32% of infections involving the appendicular skeleton. We might have anticipated that abscesses would be more common in axial sites, as symptoms are often less specific and the initial diagnosis is often delayed.6The role of MRI in the evaluation of patients with known AHO remains unclear, and has not been addressed in the literature. The goals of this retrospective study were to characterize any new findings on repeat MRI studies, and to determine whether such findings informed a change in clinical management. Of the 104 follow-up MRI studies reviewed, only 16% showed a new finding. Although the numbers are small, a new finding did not necessarily imply the need for an intervention. In 89% of patients whose osteomyelitis resolved without any change in clinical management, these repeat MRIs provide a perspective on the evolution of imaging changes during successful treatment. Our findings are in agreement with anecdotal evidence suggesting that the MRI findings often persist (even after clinical improvement or resolution), and that improvements on MRI may lag behind the clinical course. Abnormal signal changes within the bone persisted or were increased in 62% of studies. Abscesses were increased or unchanged in 33%, although it should be recognized that even after surgical evacuation, fluid may reaccumulate and not necessarily be infected. These observations highlight the importance of clinical factors (physical examination, temperature, and laboratory studies) when deciding whether monitoring the response to treatment, as many patients clinically resolved despite persistent or worsening findings on MRI.We also investigated whether the findings on follow-up MRI studies were associated with changes in treatment. None of the 43 studies obtained during the course of routine follow-up were associated with a change in treatment, and we therefore suggest that MRI plays no role in the routine monitoring of patients with AHO who have responded appropriately to treatment. Overall, 10 MRI studies (21%) ordered for persistent or worsening clinical course were followed by a change in management. Treatment changes included surgical exploration for abscess drainage (2 new, 3 increasing) or biopsy (1), and a change in antibiotics (3). In the 2 cases of known abscesses, surgical drainage could have been offered at the time of the index study. These patients did not respond to medical management and, instead, required surgical decompression after the repeat MRI. In patients whose treatment changed, there was a statistically significant difference when comparing those patients who had an MRI within 14 days (8 of 28) of diagnosis compared with those having this study after 14 days (3 of 76) (P=0.003). As CRP level has been shown to be a useful inflammatory marker in monitoring the clinical course of a patient with acute osteomyelitis, we also attempted to correlate the findings on follow-up MRIs with CRP values obtained at the same time.7 Browne et al8 suggest that CRP levels greater than 3.6 mg/dL may indicate the need for an MRI to identify any abscesses not visible on bone scan or plain film radiography. The CRP level was normal or decreasing in the vast majority of cases (80%), and CRP levels had failed to decrease or were worsening in all 11 patients whose treatment was changed (P<0.001). The CRP level was increasing in only 8 of 104 studies and 7 of these patients had a change in treatment; despite that, no additional findings were identified on the follow-up MRI in 3 of these 7 patients. A delayed response, or protracted clinical course, may also be seen with more virulent organisms such as methicillin-resistant S. aureus.9 Our study, however, did not find a statistical correlation between the organism cultured and whether additional MRI studies influenced clinical decisions.We recognize that there are limitations to this retrospective review. Selection bias exists as we only looked at the subset of patients with multiple MRI studies, some of whom had their study ordered for routine surveillance, and some because of a failure to respond clinically. Owing to confounding variables, the numbers available, and the study design, it is impossible to prove that MRI findings specifically informed the changes in treatment or affected the outcomes. Although we did not evaluate the role of plain radiographs in this study, we suggest that radiographs are an important adjunct in following patients with AHO, especially after 10 to 14 days.Although most patients respond in a predictable manner to standard treatment for AHO, a subset will lag behind the “typical” clinical course. Findings suggestive of a nonresponder include persistent or prolonged pain, persistent fevers, and inflammatory markers, which are increasing or fail to decrease as might be expected. Possible reasons for a failure to respond include ineffective chemotherapy, an undiagnosed or untreated subperiosteal or intraosseus abscess, additional sites of involvement, or an incorrect diagnosis. Options for the management of these nonresponders include a change in antibiotics or surgical exploration for biopsy, repeat cultures or irrigation, and debridement. Our findings suggest that repeat MRI may potentially be of value in patients with a persistently elevated or an increasing CRP level, especially in the first 14 days after diagnosis. One could also argue that as the location and diagnosis are known, that a failure to respond clinically justifies either a biopsy (with or without debridement) or a change in antibiotics, rather than another MRI. As noted earlier, we are unable to prove that the MRI findings specifically informed the treatment change within the limitations of this retrospective study. There is the need to compare treatment course and outcomes in 2 patient cohorts, namely those with AHO who fail to respond clinically who are evaluated by repeat MRI, and are not reevaluated with this imaging modality. In addition to the costs associated with the study, the risks of sedation in younger patients, and family time off from work, must all be considered. Further study will be required to determine whether there is a subset of patients whose treatment outcomes will be improved by repeat MRIs.While acknowledging the limitations of this retrospective review, our findings suggest that there is no indication for repeat MRI for surveillance or monitoring in patients with known AHO who have responded appropriately to treatment. Repeat MRI may potentially play a role in patients who have a CRP value that plateaus or fails to decrease, especially during the first 14 days after the diagnosis is made. Further study will be required to determine whether clinical indicators alone (persistent or worsening pain, a CRP value that plateaus or fails to decrease, or persistent fevers) are sufficient to inform a change in treatment, and whether findings on repeat MRI are of added value.ACKNOWLEDGMENTThe authors thank the contributions of Rebecca Gaugler, MD, who assisted with data collection and reviewed the final manuscript.REFERENCES1. Lalam RK, Cassar-Pullicino VN, Tins BJ. Magnetic resonance imaging of appendicular musculoskeletal infection. Top Magn Reson Imaging. 2007;18:177–191 [CrossRef] [Full Text] [Medline Link] [Context Link]2. Jaramillo D, Treves ST, Kasser JR, et al. Osteomyelitis and septic arthritis in children: appropriate use of imaging to guide treatment. AJR Am J Roentgenol. 1995;165:399–403 [CrossRef] [Medline Link] [Context Link]3. Gylys-Morin VM. MR imaging of pediatric musculoskeletal inflammatory and infectious disorders. Magn Reson Imaging Clin N Am. 1998;6:537–539 [Medline Link] [Context Link]4. Mazur JM, Ross G, Cummings J, et al. Usefulness of magnetic resonance imaging for the diagnosis of acute musculoskeletal infections in children. J Pediatr Orthop 1995;15:144–147 [CrossRef] [Medline Link] [Context Link]5. Morrison WB, Schweitzer ME, Bock GW, et al. Diagnosis of osteomyelitis: utility of fat-suppressed contrast-enhanced MR imaging. Radiology. 1993;189:251–257 [Medline Link] [Context Link]6. Connolly SA, Connolly LP, Drubach LA, et al. MRI for detection of abscess in acute osteomyelitis of the pelvis in children. AJR Am J Roentgenol. 2007;189:867–872 [CrossRef] [Medline Link] [Context Link]7. Khachatourians AG, Patzakis MJ, Roidis N, et al. Laboratory monitoring in pediatric acute osteomyelitis and septic arthritis. Clin Orthop. 2003;409:186–194 [CrossRef] [Full Text] [Medline Link] [Context Link]8. Browne LP, Mason EO, Kaplan SL, et al. Optimal imaging strategy for community-acquired Staphylococcus aureus musculoskeletal infections in children. Pediatr Radiol. 2008;38:841–847 [CrossRef] [Full Text] [Medline Link] [Context Link]9. Arnold SR, Elias D, Buckingham SC, et al. Changing patterns of acute hematogenous osteomyelitis and septic arthritis: emergence of community-associated methicillin-resistant. Staphylococcus aureus. J Pediatr Orthop 2006;26:703–708 [CrossRef] [Full Text] [Medline Link] [Context Link]hematogenous osteomyelitis; pediatric;|01241398-201012000-00021#xpointer(id(R1-21))|11065213||ovftdb|00002142-200706000-00004SL0000214220071817711065213P41[CrossRef]|01241398-201012000-00021#xpointer(id(R1-21))|11065404||ovftdb|00002142-200706000-00004SL0000214220071817711065404P41[Full Text]|01241398-201012000-00021#xpointer(id(R1-21))|11065405||ovftdb|00002142-200706000-00004SL0000214220071817711065405P41[Medline Link]|01241398-201012000-00021#xpointer(id(R2-21))|11065213||ovftdb|SL00000386199516539911065213P42[CrossRef]|01241398-201012000-00021#xpointer(id(R2-21))|11065405||ovftdb|SL00000386199516539911065405P42[Medline Link]|01241398-201012000-00021#xpointer(id(R3-21))|11065405||ovftdb|SL000436441998653711065405P43[Medline Link]|01241398-201012000-00021#xpointer(id(R4-21))|11065213||ovftdb|SL0000469419951514411065213P44[CrossRef]|01241398-201012000-00021#xpointer(id(R4-21))|11065405||ovftdb|SL0000469419951514411065405P44[Medline Link]|01241398-201012000-00021#xpointer(id(R5-21))|11065405||ovftdb|SL00006907199318925111065405P45[Medline Link]|01241398-201012000-00021#xpointer(id(R6-21))|11065213||ovftdb|SL00000386200718986711065213P46[CrossRef]|01241398-201012000-00021#xpointer(id(R6-21))|11065405||ovftdb|SL00000386200718986711065405P46[Medline Link]|01241398-201012000-00021#xpointer(id(R7-21))|11065213||ovftdb|00003086-200304000-00024SL00003086200340918611065213P47[CrossRef]|01241398-201012000-00021#xpointer(id(R7-21))|11065404||ovftdb|00003086-200304000-00024SL00003086200340918611065404P47[Full Text]|01241398-201012000-00021#xpointer(id(R7-21))|11065405||ovftdb|00003086-200304000-00024SL00003086200340918611065405P47[Medline Link]|01241398-201012000-00021#xpointer(id(R8-21))|11065213||ovftdb|00006449-200808000-00003SL0000644920083884111065213P48[CrossRef]|01241398-201012000-00021#xpointer(id(R8-21))|11065404||ovftdb|00006449-200808000-00003SL0000644920083884111065404P48[Full Text]|01241398-201012000-00021#xpointer(id(R8-21))|11065405||ovftdb|00006449-200808000-00003SL0000644920083884111065405P48[Medline Link]|01241398-201012000-00021#xpointer(id(R9-21))|11065213||ovftdb|01241398-200611000-00002SL0000469420062670311065213P49[CrossRef]|01241398-201012000-00021#xpointer(id(R9-21))|11065404||ovftdb|01241398-200611000-00002SL0000469420062670311065404P49[Full Text]|01241398-201012000-00021#xpointer(id(R9-21))|11065405||ovftdb|01241398-200611000-00002SL0000469420062670311065405P49[Medline Link]17065930Clinical Indications for Repeat MRI in Children With Acute Hematogenous OsteomyelitisCourtney, Paul Maxwell BS, BA; Flynn, John M. MD; Jaramillo, Diego MD; Horn, Bernard David MD; Calabro, Kristen BS; Spiegel, David A. MDInfection830