DailyDoctor Archive Specialties Get app
Back to the 23 September 2026 edition

Research · 02 of 05

Malignant ascites: the 250 PMN cut-off borrowed from cirrhosis does not fit

Do not diagnose or exclude bacterial peritonitis in malignant ascites on the 250 PMN cut-off alone; culture every tap and judge the whole picture.

Design
Retrospective single-centre observational study, ROC analysis
Population
261 patients with malignant ascites, 337 paracenteses (19 culture-positive)
Primary outcome
Fluid characteristics associated with positive ascitic culture
Effect
PMN ≥250: sensitivity 73.7%, specificity 72.4%; PMN >1,318: 68.4% and 93.0%

A retrospective single-centre study in the Journal of Hospital Medicine (18 September) reviewed 337 paracenteses in 261 patients with malignant ascites and a full fluid work-up. Nineteen samples, from 17 patients, grew organisms.

The cirrhosis threshold for spontaneous bacterial peritonitis — 250 PMN cells/mm3 — had a sensitivity of 73.7% and specificity of 72.4% here. So roughly one in four culture-negative taps crossed it. The best thresholds found were a PMN percentage of 63% or more (sensitivity 68.4%, specificity 90.8%) and a PMN count above 1,318 cells/mm3 (sensitivity 68.4%, specificity 93.0%).

Neither new threshold is ready for use: they were derived from only 19 positive cultures and have not been validated. What the study does support is caution. In malignant ascites a PMN count just over 250 is not the diagnosis it would be in cirrhosis, and a count below it does not exclude infection.

  • Always send cultures in blood culture bottles at the bedside, not only a cell count
  • Record the PMN percentage as well as the absolute count
  • Weigh the clinical picture — fever, pain, rising inflammatory markers — above the 250 cut-off
  • Review empirical antibiotics once cultures return rather than completing a full SBP course by default

Why it matters

It challenges the reflex of applying a cirrhosis rule to every ascitic tap on the ward.

Don't overread it

The proposed cut-offs come from 19 positive cultures at one centre and are not validated for clinical use.

The statistics, in plain English

Specificity of 72% means about 28 of every 100 uninfected samples would be called infected at 250 cells/mm3. The better-looking thresholds came from the same small dataset they were tested on, which flatters them; with only 19 positives, the sensitivity estimates carry wide uncertainty.

Read the rest in the app

You have read your two free briefings this month. The app carries all 27 specialties, every morning, free — and this finding is waiting in it.

QR code to install Daily Doctor
Get Daily Doctor — free

Scan to keep reading on your phone. No account needed to start.

transitionsinpatientevidence

Tomorrow morning, before your first patient

One edition a day for internal medicine, written by the desk, every claim tied to its paper. Six minutes.

Get the app — free
Daily Doctor All 27 specialties, every morning. Free.
Get the app