A 54 year old man had an elective reversal of Hartman’s procedure originally performed for perforated diverticular disease. The post operative course was stormy and he had a large pelvic abscess drained but settled quite quickly after the drainage procedure. His cultures were not surprising, 2 types of E. coli both quite susceptible, 2 types of Enterococci reported only as “not VRE” and 4+ mixed anaerobes including B. fragilis group.
He was initially treated with imipenem for 8 days and then switched to oral TMP/SMX 1 DS BID. He still had a drain in but there was markedly less drainage than initially. The surgeon wants to send him home and I am not sure that TMP/SMX alone is appropriate therapy. Should be be on something more broadly active? Maybe amoxicillin-clavulanate?
Your question provides opportunity to discuss several aspects of infections associated with disruptions of the colon.
Firstly, any leak of colon contents as happened in this case leads to local peritonitis and eventually abscess. The primary therapy for abscess is drainage, often guided by CT scan. It may be important to repair the colonic leak but in this case, as often happens, conservative management allows the leak to heal.
Empiric therapy for colon-associated infections is the same no matter what the origin – appendicitis, diverticulitis, malignancy, trauma. If it an acute process that has not been preceded by hospitalizations, the likelihood of colonization with antibiotic-resistant bacteria is low and the “top line” agents piperacillin/tazobactam and imipenem should be reserved for severe cases only. Ceftriaxone/metronidazole combination is a good IV choice and I would have recommended it in this case.
Sequential IV to oral therapy is appropriate and it appears that TMP/SMX was selected based on the reported susceptibilities for E. coli. However, there were many types of bacteria in this culture including enterococci and many different anaerobes. This is completely typical. Empiric therapy for anaerobes is always indicated and the addition of metronidazole to the TMP/SMX is very reasonable. 500 mg PO BID is appropriate and convenient for dosing. There is ample evidence that BID metronidazole is as efficacious as the more commonly recommended TID. The principle is to select an agent or agents that are likely to be active against E. coli and Bacteroides fragilis. Specific therapy aimed at enterococci is generally not required and reserved for complicated and severe cases.
A single agent that is commonly recommended is amoxicillin/clavulanate as you suggested. It is an excellent agent in this clinical setting.