Conference Coverage

In sickle cell disease, osteomyelitis is a tough call


– Osteomyelitis is an especially challenging diagnosis in children with sickle cell disease (SCD) because the bone and joint signs, elevated white cell counts, and C-reactive protein levels that are commonly used to diagnose bone infection are frequently features of SCD as well.

As a result, most patients with SCD and suspected osteomyelitis are treated without a confirmation of the diagnosis.

Dr. Julie Weisman

Dr. Julie Weisman

Now, data from a retrospective review of the characteristics and outcomes of osteomyelitis in children with SCD suggest that changes in erythrocyte sedimentation rate (ESR) may be a better laboratory marker for osteomyelitis than either white blood cell counts (WBC) or C-reactive protein (CRP), reported Julie K. Weisman, MD, from Children’s National Health System in Washington, DC.

Among 30 patients with SCD who were followed at a single center over a decade, 29 patients had elevated ESR, but only 13 patients had leukocytosis, and only 13 had elevated CRP.

“Prior studies on sickle cell disease have shown that it is very difficult to differentiate between osteoarticular infection and bone infarction. Therefore, oftentimes, this diagnosis is very difficult to make,” Dr. Weisman said at the annual meeting of the American Society of Pediatric Hematology/Oncology.

Laboratory findings for osteomyelitis in SCD are often nonspecific, including leukocytosis, elevated CRP and ESR, and blood cultures positive for Staphylococcus aureus (the predominant pathogen in children with osteomyelitis), or, in children with hemoglobinopathies, salmonella.

In children with SCD, CRP levels can vary from normal to elevated. ESR is similarly variable, as low hematocrit values can result in higher ESR values. Additionally, sickle erythrocytes can fail to aggregate, which can lead to lower ESR values.

A decade of data

The researchers set out to get a better handle on the characteristics and outcomes of osteomyelitis in patients with SCD and to see which laboratory and imaging findings might prove most useful for diagnosing osteomyelitis in this population. They reviewed data on 59 patients who were identified with indeterminate or likely osteomyelitis over a 10-year span. Of those, 30 were diagnosed and treated for osteomyelitis, and 29 were tentatively diagnosed but not treated. The latter group likely had symptoms caused by a bone infarction or vaso-occlusive crisis, Dr. Weisman said.

Among the 30 treated patients, osteomyelitis was confirmed by bone biopsy in 3, and an organism was isolated from blood or an abscess in 6. In the other 21, osteomyelitis was presumed based on clinical, laboratory, and MRI findings.

The median patient age was 12 years (range, 8 months to 18 years), 18 were male, and all but three patients have the HbSS genotype. Of the remaining patients, two had the HbSC and one the HbSF genotypes.

Infections occurred in the lower extremities in 11 patients, in the upper extremities in 10, in the pelvis or vertebrae in 2 each, and in the scapula, clavicle, hand, rib, or mandible in 1 patient each.

Just 13 of the 30 patients (43%) had lab findings of leukocytosis (more than 15,000 cells/mm2), and an equal number had elevated CRP (greater than 10 mg/L).

In contrast, 29 patients had an ESR above 20 mm/hour, and, in three of these patients, the rate was higher than 100 mm/hour.

When the researchers compared white blood cell counts and CRP levels between the treated patients and the 29 untreated controls, they found no significant differences for either measure of inflammation. In contrast, ESR was significantly higher among treated patients (P = .03).

Looking at the receiver operating characteristic curve for ESR, they found that an ESR of more than 100 mm/hour had 100% specificity for osteomyelitis in this group of patients.

Only 6 of the 30 (20%) had bacteremia. In 9 patients, nontyphoidal salmonella was isolated from cultures of either bone biopsy (3), abscess (3), or blood (6), but no possible causative organism could be isolated in the remaining 21 patients.

All patients were treated with prolonged antibiotic therapy. Surgical drainage and/or debridement were required in 6 patients. Two patients developed chronic osteomyelitis, but infection eventually resolved in all patients.


Dr. Weisman recommended early consultation with infectious disease experts and orthopedists; labs studies with complete blood counts, CRP, and ESR; and imaging studies with MRI when there is clinical suspicion of osteomyelitis in patients with SCD.

The study was internally funded. Dr. Weisman reported no conflicts of interest to disclose.

Recommended Reading

FDA: REMS no longer necessary for epoetin, darbepoetin
MDedge Hematology and Oncology
Eltrombopag improves frequency, speed, robustness of hematologic recovery
MDedge Hematology and Oncology
FDA boxed warning leads to drop off in use of ESAs
MDedge Hematology and Oncology
Role of TET2 in hematologic malignancies
MDedge Hematology and Oncology
Combo improves response rates in treatment-naïve SAA
MDedge Hematology and Oncology
FDA: REMS for ESAs no longer needed, though risks persist
MDedge Hematology and Oncology
Study supports use of tPA in stroke patients with SCD
MDedge Hematology and Oncology
FDA clears direct-to-consumer marketing of genetic risk tests
MDedge Hematology and Oncology
Drug granted fast track designation for PNH
MDedge Hematology and Oncology
Drug induces remission in patient with severe TA-TMA
MDedge Hematology and Oncology