Medical Myths: Fever = blood cultures now!

When I was an intern in a big old hospital one of the tasks we did was taking blood cultures off of patients who spiked fevers on the wards.  In fact thinking back on it, I think I was taking cultures to appease the senior ward nurses and not really stopping to think about the why?  Probably stabbed any number of patients with early sepsis and happily continued onto the next job without even worrying about the BP, lactate, line sepsis etc etc…  It was an annoying task as it was “time critical” – you had to get the cultures ASAP as the fever spiked – or else the boss would be grumpy, we might “miss the sepsis”.

Were we right to worry about getting the cultures done in a timely fashion?  Did it have to be “during the rigors?”

My colleague pointed out this paper by Riedel et al, Journ Clinical MIcro (April 2008).  This case-review looked at 1436 patients and their fever – culture timing profile and found that it really didn’t matter when you took the cultures in realtion to the spike of a fever – the majority of positive came from cultures taken hours after a fever.  This study cannot comment on the numbers of “missed” bacteremias as a result of wrong / delayed timing.

So what does it all mean for you, next time the nurse calls to tell you Mr Jones has spiked a temp….

  1. Walk, don’t run – it really doesn’t matter if you do it now or in an hour.
  2. Go there, look at the patient and think “what is going on here?”  Does he have a reason to be febrile already documented, on appropriate therapy?  Or, is he someone who needs to be reassessed for sepsis – does he have a line infection?  Does he have a wound infection?  Is he in SIRS or worse?  Do you need other sites cultured – urine, chest etc?
  3. Take cultures as clean as possible – aseptic approach – try and avoid those frustrating false positive BCs
  4. Contemplate changing the plan – look at the AB cover – is is sensible? Do you need to yank a line or call a surgeon to drain some pus?

Thanks Ben for the article.  Interns – relax, but do a more thorough job!    Casey


  1. Glad you are debunking this one! A bit like the instant troponin demanded by the ward nurses back in Ireland immediately after chest pain,under threat of “we’ll report you to the professor tomorrow if you don’t.” Loved that!

    My favourite addition to your debunk is that nurses would refuse to give paracetamol to these poor febrile punters unless we did cultures first, as somehow the bacteraemia would be miraculously cleared by paracetamol bringing down the temperature!! I truly think ward nurses in Ireland thought paracetamol was a powerful antimicrobial which could clear sepsis! If so,perhaps it should be slotted in to the EGDT paradigm…???

    • Matthieu G. says:

      “My favourite addition to your debunk is that nurses would refuse to give paracetamol to these poor febrile punters unless we did cultures first”

      Same in Belgium, mate!

      By the way, another medical myth that should be debunked is the need to bring down the temperature of ALL febrile patients.

  2. Yep, that last one is definitely worth a debunk.

    Isnt temperature a normal response? I thought we had data from ICU populations to demonstrate longer illness in those for hom temp more aggressively controlled? Can’t rmember the rfernce though.

  3. Michael says:

    Further to technique of taking cultures (as all the above regarding fever thought process should be considered each time).

    Ref: Clin Infect Dis. (2004) 38 (12): 1724-1730. doi: 10.1086/421087 [Looked at results of 37,500 blood cultures done at Mayo]

    -- Increased likelihood of isolating bug with 40ml vs 30ml vs 20ml vs 10ml of blood collected --> should collect at least 20ml of blood each venepuncture, and split between aerobic anaerobic bottles.
    -- Should collect 2x sets of 20ml from different venepunctures at initial culture time
    -- Should collect further 20ml sets of cultures within first 24 hours if still exhibiting signs of septicaemia, up to 4 sets of 20ml. After this the increased yield of further cultures is minimal.

    Does any of this change antibiotic practice?
    No, not as much as it should. And blood culture cost/benefit seems to lean towards the cost arm.
    [ doi: 10.1378/chest.116.5.1278 CHEST November 1999 vol. 116 no. 5 1278-1281 ] Only 20% of regimes were narrowed from 3rd Gen Cephalosporin to Penicillin in detection of penicillin-sensitive pneumococcus
    [ Arch Intern Med. 1994;154(23):2641-2645. ] 50% of antibiotic regimes changed after culture results known; documentation (or not documenting culture results) is a factor in this.

    Someone has to follow-up that culture that the ward intern collects at 10:30 pm.

  4. Casey Parker says:

    Thanks Michael for those points -- yes, you should try to get the best sample, clean and adequate volume to catch the bugs.. Can be tough in Paeds, but in adults it should be done properly.

    For reference to the concept of “fever is good” don’t treat it -- check out my previous posts on “paracetamol: mostly harmless” and it sequel.

    Good points all round

  5. Always believed this to be true. The number and volume of blood samples taken is important for yield (see Michael’s post above), the timing isn’t.

    Also, Procalcitonin is less useful for determining the presence of a bacterial infection, and is better employed to guide duration of therapy -> “An ESICM systematic review and meta-analysis of procalcitonin-guided antibiotic therapy algorithms in adult critically ill patients” (

  6. Nick Heath says:

    Speaking from a novice point of view ( junior doc in Scotland) my understanding is that cultures would need to be taken before new Antibiotic regimen started (if indicated). Antibiotics presumably the sooner the better so therefore cultures the sooner the better. Am I missing something?

    • Yes. Early culture and ABs as indicated
      But the patients whom continue to spike fevers, recurrence on the ward is what this myth is about

      The idea is that we need to get blood cultures again at the peak of fever
      Junior doctors are berated by nurses to do this when I was a lad!

      • Nick Heath says:

        That makes sense. So as long as cultures are off before AB change (pref multiple from multiple sites), life is good.

Speak Your Mind