Osteomyelitis 05: Treat the children (with osteo) well

Let’s move onto the final part in the osteo series – treatment.

Editors note: small corrections were made on 3-14-14 with the addition of information of ortho consultation and the delay of antibiotics and changing cephalexin to cefazolin. Thanks to Dr. Indi Trehan for alerting me to the errata.

Choosing the right antibiotic

If you think its osteo, then start IV treatment – I can’t reiterate that enough. If you have an orthopedist willing to take the patient to the OR in situations where this is warranted you can delay initiation of therapy only if the kid is well appearing. Ortho may be able to take a child as an add-on or the first case in the morning.

In general the causative agent can be difficult to nail down. The choice of antibiotic depends on the most likely pathogen as well as the drug’s ability to penetrate the bone. In children that appear to have localized disease you can start monotherapy, with an eye towards choosing an IV agent that can easily be converted to an oral analogue. There is not a ton of evidence in comparing agents. Lazzarini et al performed a systematic review and at the time of publication there were few studies that compared agents in a manner that effectively staged disease and outcomes. Peltola et al noted in a study of children with culture positive osteo that clinda and first-generation cephalosporins both resulted in CRP normalization in 9 days. The clinda patients seemed to have less diarrhea. Intrestingly, 85% of their patients had MSSA.

Given the overall paucity of evidence, here are some good choices for initial therapy:

Clindamycin

40 mg/kg/day divided into 3 or 4 doses
Max 3 g/day
Great if MRSA prevalence is high where you work. Feigin et al (back in 1975 no less) indicated that bone penetration (bioavailability at site of infection) was about 75%.

First generation cephalosporin

Cefazolin ≥150 mg/kg/day in 4 divided doses
Max 2-4 g/day
Start here if the MSSA prevalence in your community is at least 90% – unfortunately this is becoming less common in the US where MRSA is super popular

Antistaphylococcal penicillin

Naficillin, oxacillin, dicloxacillin – under 200 mg/kg/day in 4 divided doses
Max 8-12 g/day
Bone penetration is only about 15-17%, hence the 4x a day dosing. in addition, these are good choices only if you have a very high MSSA prevalence (≥90%) and/or you know that the organism is sensitive.

Vancomycin

≤40 mg/kg/day in 4 equal doses
Dosing needs to be adjusted based on drug levels
Up to 67% bone penetration, use over Clinda if the local prevalence of Clinda resistant Staph is >10%

It is also important to consider the age of the patient. The only situation where monotherapy initially is not necessarily appropriate 100% of the time is in the febrile neonate (under 3 months). You’re going to want to consider not only Staph spp, but also gram negatives and group B strep. Going with a third generation cephalosporin (cefotaxime) plus an anti-staph agent (vancomycin or nafcillin). If MRSA is suspected and/or the infant has been in an ICU for >1 week go with vancomycin.

In older children who are seriously ill it is more appropriate to go with vancomycin as opposed to clinda or a cephalosporin. In cases not-responsive to vanc, you’ll have to pull out the big guns. Linezolid is a big gun. The dose is 600 mg/dose IV q12 hours for children ≥12 years of age and 10 mg/kg/dose IV q8 hours those <12 years.

Coverage for Kingella is worth considering in patients under the age of 3-4 years of age, especially those in daycare. Kingella is usually susceptible to cephalosporins but often resistant to vancomycin, clindamycin and antistaphylococcal penicillins. However, kingella is usually more of an indolent/milder presentation than staph so empirically considering treating for it ahead of other organisms is not necessarily warranted.

Finally, children with sickle cell should receive a third generation cephalosporin first because of the prevalence of Salmonella.

Length of therapy and the switch to oral

At this time the evidence suggests that it is best to start therapy with IV agents. Patients with rapid improvement get better in as short as 3-4 days, many by 7-10 days. Those with abscesses or septic arthritis obviously do not and merit surgical management. There have been recent trials that have indicated that there were no change in outcomes when the length of IV treatment was less than one week. Jagodzinski et al noted that in a prospective series of children with osteo that more than half could be converted to oral in as short as 3 days based on improving labs and clinical status. There were no MRSA cases though. There is still controversy regarding whether it is appropriate to go oral in MRSA cases. Overall MRSA cases probably need to be treated longer – up to 4-6 weeks, versus those that are not – 3-4 weeks. Dich et al noted that the risk of treatment failure is greater in patients treated less than 3 weeks. This study was from 1975, and MRSA was not a consideration at that time. Peltola et al also saw resolution in approximately 3 weeks – again, they saw mostly MSSA. It is clear then, that the evidence has not evolved to the point where the optimum length of treatment is unknown. Ultimately this is less important in the Pediatric Emergency Department, but still important to consider when initially managing the child with osteo.

Monitoring treatment response

Certainly it is important to continue to assess for fever, local inflammation, and the development of new sites of infection. Both ESR and CRP can be used to trend response to therapy. CRP responds to normal at a faster rate and seems to be more sensitive. also increases early in the infection but returns to normal sooner than ESR. In addition, the rate at which CRP returns to normal may be a sensitive indicator of a complicated clinical course. Peltola et al. noted that in 50 children with Staph osteo the CRP normalized in a mean of 9 days vs 29 days for ESR. Roine at al found that patients with a higher CRP on treatment day 4 had a higher likelihood of a complicated clinical course (need for repeat drainage, more extensive radiographic findings). WBC, if elevated usually normalizes within 7-10 days of effective treatment. Complete X-Ray resolution can take months. Therefore, at this time CRP is the best available option to trend response in addition to clinical parameters. ESR and CBC can be viewed as additive.

Complicated courses

Failure of clinical response will generally be evident by the one week mark. Complications can be related to local spread (thrombophlebitis, abscess or septic arthritis requiring surgical drainage), unusual pathogen (mycobacterium or fungus), or worsened systemic disease.

Summary & Recommendations

Initial antibiotic choice:

Neonate: 3rd generation cepahlosporin + vanc (if in ICU or MRSA prevalence is high) OR nafcillin/oxacillin
>3 months of age: Clinda if MRSA prevalence is high, anti-staphylococcal penicillin or first generation cephalosporin if MSSA is more prevalent in your area
Sickle cell: Third generation cephalosporin