Improving Outcomes, Improving Throughput, Mythbusting

Dueling (dual-ing?) cellulitis therapy!

While my shop does not observe or admit a large number of cellulitis patients, though this study makes me pause to consider my future treatment plans.

A retrospective analysis of 269 patients admitted to a 941 bed Alabama hospital with an acute bacterial skin or skin-structure infection were evaluated based on administration of either vancomycin monotherapy or vancomycin & clindamycin combination therapy. The primary outcome was hospital length of stay, with 90 day readmission rate as a secondary outcome.

For all-comers, LOS was similar, though there were fewer 90-day readmissions (5.3% vs 15.3%) in the combination therapy arm. However, subgroup analysis revealed that, for patients with an abscess, combination therapy was associated with an 18.2% decrease in LOS, as well as an impressive decrease in 90 day readmissions (2% vs 24.3% ! – OR 14.6 !!!!).

First, this is a subgroup analysis of a retrospective analysis of a single center, so let’s hold our horses a bit. Second, a 90 day readmission rate is a long time, but at 24.3%, that seems quite high. Third, I am unsure of what the local antibiogram looks like for Northern Alabama, so I’m not certain if clindamycin would be the best choice everywhere. Lastly, why are we admitting these patients for so long? Why not consider nsaids or steroids to decrease LOS – or admissions?


One thought on “Dueling (dual-ing?) cellulitis therapy!

  1. Hey Patrick, thanks for your post. Although the results of this study are certainly thought-provoking, there are several limitations, including those you discussed above, that will prevent me from jumping on the dual therapy bandwagon so soon.

    A few more points to consider:
    -Significantly more patients in the dual therapy group received I&D compared to vanco monotherapy (94% vs 81%). I&D was found to be a predictor of decreased 90-day admissions in those with abscesses (table 5): “Patients with abscesses treated with combination therapy were 88.5% less likely to be readmitted than patients receiving monotherapy (P = .011). Patients who had I&D were 86% less likely to be readmitted than patients without I&D (P = .017).

    -Significantly more diabetic patients and higher rates of MRSA present in monotherapy group (although regression analyses demonstrated these factors did not influence outcome data).

    -Although the difference was not found to be significant, there were ~3x more vanc isolates with MICs of 2 mcg/mL in the monotherapy group vs dual tx (10.5% vs. 3%) which could potentially lead to treatment failure with vancomycin. In fact, these isolates were associated with a longer LOS compared to those <2 mcg/mL.

    -A significantly greater number of patients in the monotherapy group discharged on either vanc, linezolid, or daptomycin. These patients may represent a population with more severe infections unable to be transitioned to PO antibiotics and thus more susceptible to a greater LOS and more likely to readmitted.

    -Most patients were discharged with sulfamethoxazole-tmp, however a greater number of dual tx patients were continued on clinda outpatient (39% vs 15.5%). The authors did not comment on susceptibility rates for smx-tmp or evaluate the success of I&Ds.

    -Despite the 0% incidence of C. diff found in this study, we can't definitively trust this safety profile, as patients may have presented to an alternative hospital for further management following discharge. This incidence is inconsistent with existing data. Previously reported rates of C. diff associated with clindamycin exposure are quite terrifying (OR 16.8, 95% CI 7.48-37.76, and should prompt careful consideration prior to adding clinda on top of concomitant antibiotics.

    I'd like to echo your point that local antibiograms should play a pivotal role in the guidance of antibiotic therapy. Here in our shop, MRSA is nearly 100% susceptible to vanc, while clinda comes in much lower at only 62%. With the exception of anaerobic activity, we're not really adding any additional antimicrobial coverage with clinda that vanc doesn't already have. Some may argue in favor of the "Eagle effect" and potential benefits of a ribosomal mechanism, but solid evidence to support optimized outcomes associated with this theory is lacking. With the exception of a suspected toxin-producing infection such as necrotizing faciitis, I'd have a hard time recommending clinda in combination with vanc for an abscess until we see replication of these results from a prospective RCT…

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s