Acute Cholangitis at 28 Weeks: Antibiotic Choice When the Patient Isn't the Only One at Risk
Acute cholangitis at 28 weeks, hours before ERCP achieves real source control — and choosing antibiotic coverage that protects both the mother's actual severity and the fetus at the same time.
Melissa H., 31, is twenty-eight weeks into her second pregnancy and two days into a fever and right-upper-quadrant pain that started, she says, "like the worst side stitch I've ever had, except it wouldn't stop." A previously uncomplicated pregnancy, previously healthy history. Ultrasound, the preferred modality in pregnancy, showed a dilated common bile duct with a visible stone, and her mild scleral icterus on exam matches: acute cholangitis. Her white count of 15,200 and a temperature of 39.2°C give her two of the Tokyo Guidelines' five severity criteria, which is what makes this Grade II rather than Grade I — a distinction that decides how urgently the duct has to be drained, and, the infectious disease service will argue, how broadly she has to be covered until it is. Her first pregnancy, four years ago, was uncomplicated and ended in a term vaginal delivery, and today's fetal growth measurements are appropriate for 28 weeks — a genuinely reassuring baseline against which any drug choice here has to be weighed. Her vitals remain reassuring, blood pressure 112/70, heart rate 96, and fetal heart tracing is reassuring with no contractions. ERCP with fetal shielding is arranged for later today to achieve source control. Before that happens, though, antibiotics need to start now, and the choice carries a second patient's safety alongside her own.
Ceftriaxone and metronidazole together cover the same organisms driving most community-acquired cholangitis — E. coli, Klebsiella, and typical biliary anaerobes — and both are among the agents Gomi's Tokyo Guidelines antimicrobial recommendations list for community-acquired Grade I and II disease, so choosing them is not a concession made for the pregnancy. The pregnancy is what rules out the alternatives: fluoroquinolones carry real theoretical fetal cartilage and tendon concerns, and extended aminoglycoside courses carry a genuine ototoxicity signal. Metronidazole's own reputation in pregnancy rests on Burtin's meta-analysis, which found no increased teratogenic risk — and which drew mainly on first-trimester exposures, so at twenty-eight weeks she sits later than the population that reassurance was actually built from, not earlier. Set against that reassurance is a real operational worry: cholangitis, left only partially covered, can decompensate from moderate to severe faster than a scheduled reassessment might catch, and some biliary-sepsis guidance leans toward broader empiric coverage — piperacillin-tazobactam specifically — for exactly that reason. The team has roughly six to twelve hours before ERCP achieves the actual source control that ends this question either way. Which leaves the grade itself as the weakest thing in the room. Two criteria is what she meets right now, on numbers drawn once, hours ago; a third would move her, and nothing about Grade II describes how long a patient stays there.
Before the first dose, with ERCP still hours away
Ceftriaxone plus metronidazole — both have long, reassuring safety records in pregnancy, adequate biliary tract penetration, and cover the same E. coli/Klebsiella/anaerobic spectrum that's actually causing most community-acquired cholangitis. It avoids fluoroquinolones, which carry real theoretical fetal cartilage/tendon concern, and avoids extended aminoglycoside exposure, which carries a genuine ototoxicity signal. At two of five criteria, there's no coverage gap this regimen leaves open.
Doesn't disagree with the pregnancy-safety reasoning in isolation. Disagrees that today's severity is the only number that matters here — Gomi's own recommendations move to piperacillin-tazobactam at Grade III, and cholangitis is an infection that can cross that line fast. She met two of five criteria this morning on a single set of labs; a rising white count or a falling pressure adds a third without anyone re-examining her.
"No coverage gap at her current severity" assumes her current severity is where this stays until ERCP actually clears the duct — that's the part not yet guaranteed.
Both of them are actually arguing about the same six-to-twelve hours between now and ERCP, not about today's regimen forever. Start ceftriaxone-metronidazole — it's correct for Grade II and it's the better choice for the fetus while it's adequate — but set an explicit trigger, not a vague "reassess later": systolic pressure under 90, any new altered mentation, or urine output under 0.5 mL/kg/hr for two hours automatically moves her to piperacillin-tazobactam without waiting for a formal team re-huddle. That gives the ID physician's real deterioration concern an actual mechanism instead of a hope that someone notices in time, and it doesn't cost the fetus anything today that today's severity doesn't already justify.
Agreed on ceftriaxone-metronidazole now with the explicit escalation trigger charted in the nursing orders, ERCP scheduled for later today with fetal shielding.
Not agreed: whether the trigger criteria should also include a specific lactate threshold, which the ID physician wanted added and the obstetrician thought would generate false escalations given how nonspecifically lactate can run in normal late pregnancy — left for the covering team to decide before shift change.