Factors that Impact Duration of Antibiotic Therapy from Phase 3 Studies of Eravacycline for Intra-Abdominal Infection

Author(s):
Kenneth Lawrence; Erin Mathias; Philip Barie; Larry Tsai

Background:

Results from 2 trials (STOP-IT, DURAPOP) showed that, for patients with adequate source control (SC) for complicated intra-abdominal infection (cIAI) short DoRx was associated with similar outcomes compared to longer DoRx. This post-hoc analysis of 2 pooled Phase 3 studies assessed the impact of identified parameters on DoRx and clinical success.

Hypothesis:

Parameters associated with DoRx differed for patients with and without complicated appendicitis (cA), but that clinical outcomes stratified by DoRx are similar.

Methods:

Patients with cIAI were randomized (1:1) to eravacycline or a carbapenem. DoRx was discretionary up to 14d. Clinical outcome (at test of cure) was the primary endpoint in the microbiological-intent-to-treat (micro-ITT) population. Groups were categorized based on DoRx: <5d, 6-8d and >8d. Statistical analysis assessed the association of several collected patient variables, treatment allocation, and operative diagnosis (multi-group C², p<0.05). Logistic regression controlled for all parameters having a univariate association (p<0.20) with DoRx

Results:

Clinical success rates were higher for patients with complicated appendicitis (cA) and unaffected by DoRx. Amongst all patients, those who received longer DoRx were older (p=0.004), sicker (p<0.001), less likely to have cA (p<0.001) and underwent open surgery (p<0.001). For cA patients, open surgery and polymicrobial infection were associated with DoRx (p<0.001), whereas non-cA patients with longer DoRx were more likely to have open surgery and prior antibiotics (both, p=0.001), and to be sicker (p=0.029). Independent predictors of longer DoRx (logistic regression) are shown (Table) (odds ratio, 95% confidence interval; age was not an independent factor).

Parameter                          Overall                            cA                             Non-cA

 

APACHE II >10             1.52 [1.07-2.15]                                                 1.47 [1.01-2.14]

Open procedure            2.11 [1.60-2.79]             2.79 [1.75-4.45]             1.90 [1.34-2.70]

Non-cA                         2.68 [2.68-3.59]

Polymicrobial isolate                                          3.02 [1.62-5.62]

Conclusions:

Patients receiving longer DoRx for cIAI were generally sicker and underwent open surgery for non-cA. It is unclear why polymicrobial bacterial isolates (a laboratory artifact, as all cases are polymicrobial) drive longer DoRx of cA. With adequate SC, longer DoRx for cA demonstrably does not affect clinical success. Further research may elucidate why longer DoRx is still chosen and whether certain populations (e.g., failed SC) would benefit from longer DoRx.