In my patient with sepsis, is administration of proper antibiotics within an hour compared to 1-3 hours associated with better outcome?

The weight of the evidence based on observational studies suggests that the earlier the antibiotics are administered even within the first 3 hrs of the diagnosis of sepsis,  the better the patient outcome.

A 2017 study analyzing data from 37 studies (primarily observational) involving ~20,000 patients with severe sepsis and/or shock found a 10% increase in hospital mortality for every 1 hr delay in initiation of antibiotic therapy1. Two multicenter studies (1 in Pennsylvania2 and another in California3) and a New York State data base study involving patients with severe sepsis or septic shock4 similarly reported decreased survival with each 1- hr delay in antibiotic therapy. Another study of patients with severe sepsis found that each hour delay in first antibiotic dose administration was associated with an 8% increased risk of progression to shock5.

Despite the emphasis on the timing of the first dose of antibiotics, let’s not forget that the second dose of antibiotics should also be given on time in sepsis; a >25% delay is associated with increased mortality, length of stay and requirement for mechanical ventilation6.

So, yes, antibiotics should be given within 3 hours of diagnosis of sepsis, but within an hour followed by a timely second dose is even better!

Final Pearl: Did you know that sepsis is the 3rd leading cause of death in the US and contributes to 1 in every 2 to 3 hospital deaths7?



  1. Kalil AC, Johnson DW, Lisco SJ, et al. Early goal-directed therapy for sepsis: a novel solution for discordant survival outcomes in clinical trials. Crit Care Med 2017;45:607-14.
  2. Seymour CW, Kahn JM, Martin-Gill, et al. Delays from first medical contact to antibiotic administration for sepsis. Crit Care Med 2017;45:759-65.
  3. Liu VX, Fielding-Singh V, Greene JD, et al. Th timing of early antibiotics and hospital mortality in sepsis. Am J Respir Crit care Med 2017; 196;858-63.
  4. Seymour CW, Gesten F, Prescott HC, et al. Time to treatment and mortality during mandated emergency care for sepsis. N Engl J Med 2017;376:2235-44.
  5. Whiles BB, Deis AS, Simpson SQ. Increased time to initial antimicrobial administration is associated with progression to septic shock in severe sepsis patients. Crit Care Med 2017; 45:623-29.
  6. Leisman D, Huang V, Zhou Q, et al. Delayed second dose antibiotics for patients admitted from the emergency department with sepsis: prevalence, risk factors, and outcomes. Crit Care Med 2017;45:956-65.


If you enjoyed this pearl, sign up to get “Fresh Pearls from Oceans of Knowledge!” with each new post!

In my patient with sepsis, is administration of proper antibiotics within an hour compared to 1-3 hours associated with better outcome?

Is iron therapy contraindicated in my patient with active infection?

In the absence of randomized-controlled trials of iron therapy in patients with active infection, the harmful effects of iron therapy (IT) in this setting remains more theoretical than proven. 1,2

Although many pathogens (eg, E. coli, Klebsiella, Salmonella, Yersinia, and Staphylococcus species) depend on iron for their growth2,3, and iron overload states (eg, hemochromatosis) predispose to a variety of infections, studies evaluating the risk of infection with iron therapy have reported conflicting results.1-4 A 2015 systematic review and meta-analysis of 103 trials comparing IV iron therapy  with several other approaches, including oral iron therapy or placebo, found no increased risk of infections with IV iron.5 In contrast, an earlier systematic review and meta-analysis involving fewer number of trials found an increased risk of infections with IV iron. 6

These varied results are perhaps not surprising since the effects of iron therapy on the risk of infection is likely to be context-specific, depending on the patient’s preexisting iron status, exposure to potential infections and co-infection and genetic background. 4 Of interest, mice with sepsis have worse outcomes when treated with IV iron.7

Perhaps the most prudent approach is to hold off on iron therapy until the active infection is controlled, unless the benefits of urgent iron therapy is thought to outweigh its theoretical harmful effects.



  1. Daoud E, Nakhla E, Sharma R. Is iron therapy for anemia harmful in the setting of infection? Clev Clin J Med 2011;78:168-70.
  2. Hain D, Braun M. IV iron: to give or to hold in the presence of infection in adults undergoing hemodialysis. Nephrology Nursing Journal 2015;42:279-83.
  3. Jonker FAM, van Hensbroek MB. Anaemia, iron deficiency and susceptibility of infections. J Infect 204;69:523-27.
  4. Drakesmith H, Prentice AM. Hepcidin and the iron-infection axis. Science 2012;338:768-72.  
  5. Avni T, Bieber A, Grossman A, et al. The safety of intravenous iron preparations: systematic review and meta-analysis. Mayo Clin Proc 2015;90:12-23.
  6. Litton E, Xiao J, Ho KM. Safety and efficacy of intravenous iron therapy in reducing requirement for allogeneic blood transfusion: systematic review and meta-analysis of randomized clinical trials. BMJ 2013;347:f4822.
  7. Javadi P, Buchman TG, Stromberg PE, et al. High dose exogenous iron following cecal ligation and puncture increases mortality rate in mice and is associated with an increase in gut epithelial and splenic apoptosis. Crit Care Med 2004;32:1178-1185.
Is iron therapy contraindicated in my patient with active infection?

Are two sets of blood cultures adequate for evaluation of bacteremia in my febrile patient?

For great majority of patients, more than 2 sets of blood cultures is not likely to significantly improve the yield of detecting bacteremia. 

Although a 2004 report suggested that 2 sets of blood cultures over 24 h period had a sensitivity of only 80% for bacteremia, several other studies have found much higher sensitivities, ranging from ~90%- 99% 2-3. When broken down by organism, sensitivity of 2 sets of blood cultures may be highest for Staphylococcus aureus (97%), followed by E. coli (91%), and Klebsiella pneumoniae (90%) 2.  The Clinical and Laboratory Standards Institute guidelines recommend paired blood culture sets (each set with 2 bottles, 10 ml of blood in each) to detect about 90-95% of patients with documented bacteremia, and 3 sets for 95-99% detection rate 4.

It seems prudent to strike a balance between drawing more than 2 sets of blood cultures—with its attendant risk of picking up contaminants— and what may be a definite but small incremental increase in the rate of detection of true bacteremia.

Whatever you do,  please don’t order only 1 set of blood cultures! Aside from its generally low yield, when positive it may be difficult to distinguish contaminants from true invaders.



  1. Cockerill FR, Reed GS, Hughes JG, et al. Clinical comparison of BACTEC 9240 Plus Aerobic/F resin bottles and the Isolator aerobic cultures. Clin Infect Dis 2004;38:1724-30.
  2. Lee A, Mirrett S, Reller LB, et al. Detection of bloodstream infections in adults: how many cultures are needed? J Clin Microbiol 2007; 45:3546-48.
  3. Towns ML, Jarvis WR, Hsueh PR. Guidelines on blood cultures. J Microbiol Immunol Infect 2010;43:347-49.
  4. Weinstein MP, Reller LB, Murphy JR, et al. The clinical significance of positive blood cultures: a comprehensive analysis of 500 episodes of bacteremia and fungemia in adults. I. Laboratory and eipidemiologic observations. Rev Infect Dis 1982;5:35-53.
Are two sets of blood cultures adequate for evaluation of bacteremia in my febrile patient?

My hospitalized patient has developed acute nausea, vomiting, and diarrhea. Is there an association between proton pump inhibitors and acute gastroenteritis?


Proton pump inhibitors (PPIs) have been associated with increased risk of Clostridium difficile infection, as well as acute gastroenteritis (AG) caused by Salmonella, Campylobacter, and most recently, norovirus. 1,2

A recent prospective study1 of over 38,000 patients (mean age ~ 70 y) found a significant association between PPI use and AG leading to hospitalization with a dose-response relationship.  PPI use increased the risk of Salmonella, Campylobacter, and C. difficile infections.  Of note, H2 receptor antagonists were not associated with AG-related hospitalization in this study.

A 2017 retrospective case-control study also showed an association between PPI use and norovirus infection in hospitalized patients (mean age ~80 y in both groups). Most cases occurred during epidemic years with a median hospital stay of 5 days before onset of symptoms. Given the usually short incubation period of norovirus AG (typically 12-48 h), many of these cases likely acquired the infection during their hospital stay.

Besides reducing the acidity of gastric juice, PPIs may increase the risk of AG by causing an overgrowth of bacteria in the GI tract, reduce its motility and adversely affect the immune response, including neutrophil chemotaxis. 3

Does your patient really need a PPI?



  1. Chen Y, Liu B, Glass K, et al. Use of proton pump inhibitor and the risk of hospitalization for infectious gastroenteritis. PLoS One 2016;11:e0168618. Doi:10.1371/journal.pone. 0168618.    
  2. Prag C, Prag M, Fredlund H. Proton pump inhibitors as a risk factor for norovirus infection. Epidemiol Infect 2017;145:1617-23.
  3. Wandall JH. Effects of omeprazole on neutrophil chemotaxis, super oxide production, degranulation, and translocation of cytochrome b-245. Gut 1992;33:617-21.
My hospitalized patient has developed acute nausea, vomiting, and diarrhea. Is there an association between proton pump inhibitors and acute gastroenteritis?

My patient with low back pain was just diagnosed with a lumbar spinal epidural abscess. Should I order an MRI of the rest of the spine?

First, look closely for any signs or symptoms which may suggest cord involvement due to spinal epidural abscess (SEA) at other levels of the spine (in this case cervical or thoracic) which would necessitate an urgent MRI. Be particularly on the lookout for new pain (particularly radicular) or paresthesias involving the abdomen, chest or upper extremities (with or without weakness)1.

Otherwise, whether an MRI of the entire spine should be routinely obtained after a diagnosis of SEA in the absence of any suggestive signs or symptoms is less clear, in part related to lack of properly designed studies.1-4

Nevertheless, a retrospective study involving 233 patients with SEA may shed some light on the subject. Based on 22 cases of noncontiguous SEA (9.4% of total), the following independent risk factors were identified3:

  • Delay in presentation (≥1 week of symptoms)
  • Concomitant area of infection outside the spine and paraspinal region
  • ESR > 95 mm/h at presentation

Probability of non-contiguous SEA based on the number of risk factors was as follows:

  • 3 risk factors: 73%
  • 2 risk factors: 13%
  • 1 risk factor: 2%
  • Zero risk factor: 0%

Despite several shortcomings and the need to confirm its findings2,3, this study helps raise awareness of the potential for concurrent but asymptomatic SEA elsewhere in the spine whenever SEA is diagnosed.



  1. Bond A, Manian FA. Spinal epidural abscess: a review with special emphasis on earlier diagnosis. BioMed Res International 2016;Volume 2016, Article ID 1614328.
  2. Schoenfeld AJ, Hayward RA. Predicting modeling for epidural abscess: what we can, can’t, and should do about it. Spine J 2015;15:102-104.
  3. Ju KL, Kim SD, Melikian R, et al. Predicting patients with concurrent noncontiguous spinal epidural abscess lesions. Spine J 2015;15:95-101.
  4. Pfister HW, vonRosen F, Yousry T. MRI detection of epidural spinal abscesses at noncontiguous sites. J Neurol 1996;243:315-7.
My patient with low back pain was just diagnosed with a lumbar spinal epidural abscess. Should I order an MRI of the rest of the spine?

In my critically ill patient with infection, is capillary refill time greater than 2 seconds indicative of septic shock?

The data on the performance of capillary refill time (CRT) in adults is quite limited and what’s available does not suggest that the commonly cited 2 seconds cutoff is useful in assessing peripheral perfusion in critically ill adults1,2.

For example, a large study involving 1000 healthy adults reported that 45% of participants had a CRT > 2 seconds3.  Age also affects CRT with its 95 percentile upper limits reaching 4.5 seconds among healthy adults >60 y old3

Among patients with septic shock, a baseline median CRT of 5 seconds has been reported.  Values <5.0 seconds within 6 hours of treatment of septic shock has also been highly associated with successful resuscitation even before normalization of lactate levels4.

For these reasons, if CRT is used as a measure of peripheral perfusion in critically ill adults, a cut off of 5 seconds, not 2 seconds, may be more appropriate. But just like many other diagnostic tests, CRT should never be interpreted in isolation from other clinical parameters. 


  1. Lima A, Bakker J. Clinical Assessment of peripheral circulation. Critical Care 2015:21: 226-31.  
  2. Lewin J, Maconochie I. Capillary refill time in adults. Emerg Med J 2008;25:325-6.
  3. Anderson B, Kelly AM, Kerr D, et al. Impact of patient and environmental factors on capillary refill time in adults. Am J Emerg Med 2008;26:62-65.
  4. Hernandez G, Pedreros C, Veas E, et al. Evolution of peripheral vs metabolic perfusion parameters during septic shock resuscitation. A clinical-physiologic study. J Crit Care 2012;27:283-288.
In my critically ill patient with infection, is capillary refill time greater than 2 seconds indicative of septic shock?

Do most patients with mycotic aneurysms have endocarditis?

No! In fact, the great majority of patients who develop mycotic aneurysm (MAs) in the postantibiotic era have no evidence of endocarditis1-3.

MAs are thought to be related to microbial arteritis due to blood stream infection of any source with implantation of circulating pathogen (usually bacterial) in atherosclerotic, diseased, or traumatized aortic intima. Plus, MAs may develop due to an adjacent infectious process (eg, vertebral osteomyelitis), either through direct extension or via lymphatic vessels, pathogen seeding of vasa vasorum, or infection of a pre-existing aneurysm1,2.  All these factors may occur in the absence of endocarditis.

Many of your patients may be at risk of MA such as those with advanced age or history of diagnostic or therapeutic arterial catheterization, illicit intravascular drug use, hemodialysis and depressed host immunity1-3..  Staphylococcus aureus, Salmonella sp, S. epidermidis and Streptococcus sp are common culprits in descending order1-3.

So think of MA in your patient with recent blood stream infection,  particularly due to S. aureus or Salmonella sp, in the setting of persistent signs of infection  with or without evidence of endocarditis.

Final Fun Fact: Did you know that the term “mycotic aneurysm” is a misnomer, having been first introduced by Sir William Osler to describe aneurysms of the aortic arch in a patient with (you guessed it) bacterial not fungal endocarditis?


  1. Gomes MN, Choyke PL, Wallace RB. Infected aortic aneurysms: A changing entity. Ann Surg 1992;215:435-42.
  2. Muller BT, Wegener OR, Grabitz K, et al. Mycotic aneurysms of the thoracic and abdominal aorta and iliac arteries: Experience with anatomic and extra-anatomic repair in 33 cases. J Vasc Surg 2001;33:106-13.
  3. Mukherjee JT, Nautiyal A, Labib SB. Mycotic aneurysms of the ascending aorta. Tex Heart Inst J 2012;39:692-5.
Do most patients with mycotic aneurysms have endocarditis?