Antibiogram test explained, from one culture to a cumulative report

An antibiogram test is the susceptibility panel run on one isolate: the laboratory grows the organism against a set of antibiotics and reports each as susceptible, intermediate or resistant. A cumulative antibiogram is a different thing built from many of those panels, summarising a year of isolates for empiric prescribing.

The individual panel

When a culture grows a significant organism, the laboratory identifies it and tests it against a panel of agents chosen for that organism and that specimen type. Most laboratories use an automated system that measures growth across a set of antibiotic concentrations and derives a minimum inhibitory concentration, the lowest concentration that inhibits visible growth.

The MIC is then compared against a breakpoint, and the result is reported as S, I or R. That letter is what reaches the chart, and it is what most clinicians ever see. The panel is selective by design: agents with intrinsic resistance are not reported, and some agents are suppressed unless the isolate is resistant to the preferred options, which is a cascade reporting rule and a stewardship intervention in its own right.

The cumulative report

A cumulative antibiogram takes every isolate from a period, normally twelve months, and reports for each organism and drug pair the percentage of isolates that were susceptible. It exists to answer a different question from the individual panel: not what to give this patient whose results are back, but what to start in the next patient whose results are not.

Because it is an aggregate, it needs rules that an individual panel does not.

• Count each patient once per organism, normally the first isolate in the period
• Do not report a cell with fewer than 30 isolates, per CLSI M39 guidance
• State how intermediate results were handled, since they can be excluded or counted as resistant
• State the period, the population and the breakpoint source on the report itself

Worked example, three rows to one cell

Suppose three urine isolates of Escherichia coli tested against ciprofloxacin: two susceptible, one resistant, and one of the susceptible ones is a second culture from a patient already counted. Under the first isolate rule the duplicate is dropped before anything is counted, leaving one susceptible and one resistant, which is 50 percent on a denominator of two.

In a real report that cell would carry several hundred isolates, and a cell with two would be suppressed. The example is only there to show that the transform is counting under stated rules, which is exactly why it is worth checking rather than trusting.

Reading a cumulative antibiogram without being misled

• Read the n before the percentage. Ninety percent of 400 and 90 percent of 32 are different claims.
• Check the period and whether it is the whole facility or a cut.
• Remember that the columns are not comparable across organisms unless the same drug was tested on all of them.
• Do not pick the highest percentage. A narrow-spectrum agent scores well precisely because it is only tested where it works.
• Look for the method note. If there is not one, the report cannot explain its own trend.

Common questions about the difference

Can you use the cumulative antibiogram to treat a specific patient? No. Once that patient has their own susceptibility results, those results are better evidence than any aggregate. The cumulative report is for the window before results exist.

Does a cumulative antibiogram tell you resistance is rising? Only with care. A change between two periods is a finding when the method was identical and the denominators are large enough. Otherwise it is a change in the report, not in the organisms.

---

Related on this site: how a MIC becomes an S, I or R, CLSI and EUCAST breakpoints.