Neonatal and Pediatric Medicine
Open Access

Osteomyelitis can present unique diagnostic and management challenges in neonates and young infants. Neonates are defined as less than 28 days-of-life (DOL). Infants up to 2 months of age were included. We describe the use of magnetic resonance imaging (MRI) to aid in the timely diagnosis of two of our own patients. Labs and vitals for these cases are presented (Table 1).

Table 1: Laboratory results for our young infancy osteomyelitis patients.

We also review case reports published over the past 41 years (Figure 1) to observe trends in symptoms, diagnostic methods, etiologic pathogens, and radiography to identify osteomyelitis (Tables 2 and 3). We asked: 1) since magnetic resonance imaging (MRI) technology was introduced into diagnostic protocol in 1998, has there been an improvement in diagnosis and outcomes? 2) What additional aspects of the disease and diagnostic protocol may be affecting outcomes?

Literature search was performed using PubMed, AAP, and Scopus. Of the total records found, 1,775 were from PubMed, 1 from AAP, 0 from Scopus. Of the total case reports found 411 were from PubMed and 0 from AAP. 66 total cases were accessible. 45 of these cases included children <2 months of age. Of the 45 cases, 37 pertained to a long bone. 34 of these cases contained no other significant disease (Figure 1). In the end, 34 cases from the literature and our two cases were reviewed. These cases were divided into two groups; 1) all young infancy osteomyelitis cases before first reported MRI usage in 1998 (Table 2).

Figure 1:Flow chart of the research inclusion and exclusion method.

2) All young infancy osteomyelitis cases after first reported MRI usage in 1998 (Table 3). Groups consisted of 15 and 21 patients, respectively.

4-week-old male with 5-day neonatal intensive care unit (NICU) stay for respiratory distress and polydactyly with supernumerary finger stub removal at birth presented with progressive upper extremity pseudoparalysis. After 7 symptomatic days, he was admitted to an outside hospital emergency department (ED). The x-ray and labs were normal, and he was sent home with an “arm sprain”, despite pseudoparalysis.

After 11 symptomatic days, the elbow and arm became edematous and the infant became irritable. Radiographs repeated by the pediatrician showed bony destruction within the left distal humeral metaphysis demonstrating osteomyelitis. Initial vitals and labs are in Table 1.

MRI with gadolinium was consistent with septic elbow and osteomyelitis of the distal humerus, proximal radius, and olecranon. The patient was taken to the operating room (OR) where he underwent elbow irrigation and debridement (I&D) with packing. The humeral supracondylar area contained yellow and lobulated pus.

He was treated with cefotaxime and vancomycin, postoperative vitals are shown in Table 1. I&D was repeated 3 days later with wound closure. Methicillin-resistant Staphylococcus aureus (MRSA) sensitive to clindamycin was identified. Blood cultures were negative. He was discharged on POD 4 and treated with 6-weeks intravenous clindamycin. At 3-year follow-up, the elbow lacked 15 degrees flexion. X-rays showed bone defect filling in.

Male at 27 3/7 week twin gestation was admitted to the NICU for late-onset Group B streptococcus sepsis at DOL 17. He was treated with a 21-day course of ampicillin and intubated for poor oxygenation. Leg swelling was observed on DOL 54. Labs and vitals are presented (Table 1). Initial radiograph was normal. Follow-up radiograph on DOL 58 showed metaphyseal lucency with periosteal reaction suggestive of proximal tibial osteomyelitis. He was started on cefotaxime and vancomycin.

Orthopedic surgery was consulted. MRI demonstrated abscess of the tibia and fibula, fluid crossing into the epiphysis. An abscess containing clear fluid with fibrinous material was debrided around the proximal tibia and distal femur. Drill holes were made into the tibia epiphysis and metaphysis. OR cultures were negative. He was given a 6-week course of ampicillin for presumed Group B streptococcus and discharged on DOL 95. At 5-year follow-up, plain radiographs show the growth plate functioning with no central irregularity.

Between the two groups, most common symptoms were pseudoparalysis, in 28 (77.8%), and swelling, in 22 (61.1%) patients. 14 (38.9%) patients experienced both symptoms; all experienced one. 11 (30.6%) were febrile. Patients were between 3 and 54 DOL (average 22 DOL).

In Group 1 (cases before first reported MRI usage in 1998; Table 2), 8 (53.3%) patients are male and 7 (46.7%) are female. In group 2 (cases after first reported MRI usage in 1998; Table 3), 12 (57.1%) patients are male, and 9 (42.9%) are female.

A comparison and analysis of key diagnostic variables between the 2 groups are presented (Table 4). The results indicated no statistical significance.

Table 4: Key diagnostic variables.

We asked: 1) since magnetic resonance imaging (MRI) technology was introduced into diagnostic protocol in 1998, has there been an improvement in diagnosis and outcomes? 2) what additional aspects of the disease and diagnostic protocol may be affecting outcomes?

Over the past 41 years, despite the introduction of MRI into diagnostic protocol in 1998, diagnosis and treatment of early infancy osteomyelitis have not significantly changed. There is a low incidence of early-infancy osteomyelitis. Signs and symptoms are difficult to discern [23-25]. Prompt diagnosis and treatment are crucial [26-28].

Earlier diagnosis mitigates serious sequelae, such as cartilage destruction, bone and growth plate deformity, joint dislocation, limited joint motion, and neurologic abnormalities [29-31]. A flow chart (Figure 2) highlights common time points for diagnostic delay.

Figure 2: Diagnostic flow chart for early childhood osteomyelitis. Potential for diagnostic delay is indicated for the following time points: presentation of symptoms, diagnostic workup, and negative results.

Duration of symptoms before admission remains about the same, which may be unavoidable since parents are responsible for presenting during this subtle phase. Of the reported cases, patients were admitted after an average of 10 symptomatic days and diagnosed after 10 additional days (Tables 2 and 3). Fever occurs in one third or fewer cases [23,29]. Pseudoparalysis or extremity swelling are present in 60-95% of infections and should alert the clinician to possible osteoarticular infection [23-25]. Pseudoparalysis present at birth is indicative of a neurological problem or Erb’s palsy; however, pseudoparalysis which develops after birth indicates infection. There was a decrease in patients misdiagnosed with neuropathy after the use of MRIs, although about the same number of patients was misdiagnosed in general.

Radiologic abnormalities may be absent, subtle, or only positive in chronic infections when lytic lesions, sequestrum, or involucrum are present [23]. MRI is useful in defining bone edema, soft tissue involvement, delineating bone and soft tissue abscesses, and showing co-existing joint pathology [32,33]. MRI is the most sensitive radiological investigation for diagnosing osteomyelitis and can detect osteomyelitis 1 to 3 days post-infection [24,32].

Radiographs cannot pick up signs of osteomyelitis until after 2 weeks when the infection is chronic [32,34,35]. To prevent delay, MRI should be ordered immediately once patients develop pseudoparalysis, regardless of “normal” radiographs. MRI should not delay diagnosis, but rather hasten it and improve certainty of the underlying disorder.

Neonates are often treated empirically with broad spectrum antibiotics. Ideally, pathogen identification should be sought to optimize treatment with antibiotics. Continued advancements in PCRbased testing will enable pathogen identification to occur with higher degrees of specificity, sensitivity, and rapidness [36]. More aggressive pathogen analysis and treatment is warranted and will be especially helpful for treating neonates.

Despite the evident value of using MRI technology, the natural history of early infancy osteomyelitis may be impossible to avoid. However, an in-depth understanding of the disease and common time points of diagnostic delay may give insight as to why outcomes are not improving as one might expect. Optimizing diagnostic protocol may be warranted if it will lead to clinicians lowering the threshold for attaining MRI and utilizing the technology earlier in their patient assessments. Emphasis on early diagnosis appears to be crucial for improving outcomes.

Both patients’ parents were informed that the case data would be submitted for publication, to which they agreed.

The authors have no disclosures or conflicts of interest.