Some say 24 hours (UK Minimum retesting interval), whereas others say 48 hours (Waldron JL, et al. An automated minimum retest interval rejection rule reduces repeat CRP workload and expenditure, and influences clinician-requesting behaviour. J Clin Pathol. 2014 Aug;67(8):731-3.).
Who is right?
Actually CRP is na acute phase reactant. That means it is a protein synthezised by the liver, in response to almost any type of infection / inflammation.
You can have a look of this paper:
https://www.researchgate.net/publication/239070163_The_titration_curve_of_native_C_reactive_protein
It is a rather inespecific though - meaning the test isn´t able to actually made a diagnosis. Critically ill patients, or patients with such conditions as Rheumatoid Artritis, or even people wit mild chronic respiratory illness, will have it chronically elevated - So there´s no sense in determining it too frequently, unless you want to investigate the recuring of the condition or infection. For Rheumatoid Arthitis, for instance, I´d have it determined on a monthly or bimonthly determination, just as a way to monitor treatment.
Another use of the determination of CRP is the ultrassensitive assay. This one should be determined only if the patient has been exempt of any cause of infection - inflamation for the last 1 to 2 weeks. In that case CRP should be close to zero, and any small amounts of CRP that may be detected can be associated with cardiovascular risk. also in this case it is useless to repeat it too often, since other conditions may produce a false positive result.
Anyway, I wouldn´t recommend repeated determinations (less than say at least a week or 2 apart) lest (since it is a sensitive, but inespecific test) you won´t really have the best information the test can be provide (and therefore has a low advantage / cost relationship)
Article The titration curve of native C reactive protein