Skip to content

Trick of the Trade: IV-Push Antibiotics in the ED

IV_arm5 copyLimited intravenous access is a common conundrum in the Emergency Department, with heavy implications for medication administration. Of particular concern, are the profoundly septic patients that necessitate multiple timely therapies, which require tying up a line – fluids, pressors, several antibiotics, etc. The shift away from less central line (i.e. triple lumen) placement for initial resuscitation, may serve to further exacerbate this issue.

The Problem

Since most of these septic patients will require more than one antimicrobial for empiric coverage, the exact timing of therapy and desired order of administration are details that may not be adequately communicated. How many times have you noticed after ordering vancomycin and cefepime, that vancomycin has been administered first, allowing several hours to go by without having received that broad-spectrum, gram-negative coverage with cefepime?

Despite these challenges, the Surviving Sepsis Campaign (SSC) guidelines currently recommend administration of appropriate empiric antibiotics within 1 hour after recognition of severe sepsis.1 This is largely based upon a retrospective analysis of 2,731 patients with septic shock, which demonstrated that administration of effective antibiotics within the first hour of documented hypotension was independently associated with increased survival to hospital discharge.2 Currently however, a lack of guidance exists regarding the best way to achieve this important aspect of the resuscitation bundle.

Trick of the Trade – IV Push Antibiotics

Administration of antibiotics via intravenous push (IVP) may be one approach to hasten antibiotic administration without tying up lines. Additionally, while communicating the desired order of administration is still important, when faced with limited IV access, nurses may be more inclined to initiate the IVP antibiotic first, due to shorter administration times. Further, IVP administration of antibiotics may be a more economically beneficial alternative as compared to more common methods.3 Through cost avoidance of both pharmacy preparation and nursing administration time, as well as eliminating the need for minibags and IV tubing, one study estimated savings of $184,000 per year.4

Listed below are various antibiotics that have been shown to be safe when given as an IVP to adults:5–8

Antibiotic Concentration*, Diluent Rate of Administration Osmolality (mOsm/L)**
Cefazolin 1 g/10 mL, SWFI 1-2 min 340
Cefuroxime 750 mg/10 mL, SWFI 1-2 min 447
Cefoxitin 1 g/10 mL, SWFI 2-4 min 525
Cefotaxime 1 g/10 mL, SWFI 1-2 min 440
Ceftriaxone 1 g/10 mL, SWFI 1-2 min 423
Ceftazidime 1 g/10 mL, SWFI 1-2 min 435
Cefepime 1 g/10 mL, SWFI 2-4 min <400
Meropenem 1 g/10 mL, SWFI 3-5 min <500
Aztreonam 1 g/10 mL, SWFI 2-4 min 487

SWFI, sterile water for injection;

* The concentrations stated above do not necessarily represent recommended doses; these should be used to determine the volume required for higher/lower doses (e.g. cefepime 2 g/20 mL SWFI)

** Substances with an osmolality less than 600 mOsm/L are generally acceptable for administration via a peripheral line9

Important Considerations

  • When implementing IVP antibiotics in your ED, standardize the process. This may require substantial changes in nursing practice, updating policies and procedures, departmental education, adjustments to electronic order sets, adjustments to product stocking and inventory, as well as implementing oversight for unforeseen adverse effects
  • It is important to note, that the use of sterile water for injection (SWFI) is to help minimize osmolality; reconstituting with NS or D5W may produce significant phlebitis, and increase the risk for extravasation injury. Read more information (extravasation injuries PDF) on this topic.
  • As with any IV preparation, remember to properly label your syringe. Refer to Dr. Bryan Hayes’ post on The Art of Syringe Labeling in the ED.
  • Beta-lactam antibiotics are often administered as prolonged infusions in order to take advantage of their pharmacokinetic/pharmacodynamic profiles (T > MIC). However, this is usually employed for subsequent maintenance doses, and is also not practical for initiation in the ED. Moreover, time to administration of the initial dose is a more important factor in septic patients.
  • The literature used to support the aforementioned IVP antibiotics did not include pediatric patients. Therefore, the safety and feasibility of implementing this practice within the pediatric population is uncertain.

Take-Home Points

  • Various beta-lactam antibiotics may be safely administered via IVP.
  • IVP administration may be one strategy to help facilitate timely administration of antibiotics, and to prevent tying up multiple lines.
  • To date, no published literature exists to support the potential benefits (i.e. improved time to administration; improved outcomes) of IVP antibiotics in the Emergency Department; future studies are warranted.
Dellinger R, Levy M, Rhodes A, et al. Surviving Sepsis Campaign: international guidelines for management of severe sepsis and septic shock, 2012. Intensive Care Med. 2013;39(2):165-228. [PubMed]
Kumar A, Roberts D, Wood K, et al. Duration of hypotension before initiation of effective antimicrobial therapy is the critical determinant of survival in human septic shock. Crit Care Med. 2006;34(6):1589-1596. [PubMed]
Ambrose P, Quintiliani R, Nightingale C. Pharmacoeconomic analysis of administration of famotidine i.v. push vs.intermittent slow i.v. infusion. Ann Pharmacother. 1997;31(5):645. [PubMed]
Garrelts J, Smith D, Ast D, Peterie J. A comparison of the safety, timing and cost-effectiveness of administering antibiotics by intravenous bolus (push) versus intravenous piggyback (slow infusion) in surgical prophylaxis. Pharmacoeconomics. 1992;1(2):116-123. [PubMed]
Norrby S, Newell P, Faulkner K, Lesky W. Safety profile of meropenem: international clinical experience based on the first 3125 patients treated with meropenem. J Antimicrob Chemother. 1995;36 Suppl A:207-223. [PubMed]
Garrelts J, Wagner D. The pharmacokinetics, safety, and tolerance of cefepime administered as an intravenous bolus or as a rapid infusion. Ann Pharmacother. 1999;33(12):1258-1261. [PubMed]
Nowobilski-Vasilios A, Poole S. Development and preliminary outcomes of a program for administering antimicrobials by i.v. push in home care. Am J Health Syst Pharm. 1999;56(1):76-79. [PubMed]
Garrelts J, Ast D, LaRocca J, Smith D, Peterie J. Postinfusion phlebitis after intravenous push versus intravenous piggyback administration of antimicrobial agents. Clin Pharm. 1988;7(10):760-765. [PubMed]
Intravenous Nurses Society. Position paper: midline and mid-clavicular catheters. J Intraven Nurs. 1997;20:175-178.

ALiEM Copyedit

Reviewer Comments:

  1. Made some minor wording changes
  2. This is a great point, great article, but you might make it hit home harder with physicians by personalizing it and opening it with provider frustration. The biggest annoyance to any doc (or pharmacist) is when the antibiotics are ordered at 12pm but not administered till 3pm. It ties into our innate impatience, need to feel in control, and secret fear that our orders aren\'t being carried out, resulting in slapping our foreheads (not to mention looking silly when ICU asks when antibiotics were given and we find out they weren\'t). \"How many times have you ordered Vanc and Zosyn only to find out that because of how you placed your orders and lack of communication with nursing staff, the Vanc will run over an hour and the patient won\'t get that nice broad spectrum Zosyn for over an hour.\" I know you\'re a great EM pharmacist and maybe the main audience is for EM pharmacy, but a quick clinical scenario will resonate with readers. Save lives, save money, save time is the reason people need to read this.
  3. May emphasize the complication that multiple antibiotics places on timing of treatment. Many septic patients get 2 or 3 antibiotics. You briefly mention \"multiple therapies\"
  4. I\'d delete the second sentence in paragraph 2 \"While not explicitly studied\" along with reference 3. The article referenced provides no data to support IVP abx and it\'s just author conjecture, which is what you\'re writing now also. Find a better article which actually shows IVP abx improve adherence to SCC guidelines, or admit there\'s no data yet.
  5. Not sure everyone knows these, please establish abbreviations SWFI and CPOE.
  6. What you\'re really alluding to is damage to peripheral veins as well as extravasation injury in hyperosmolar solutions. Can you clarify that point without making this too long? Some learners are very familiar with this concept, others are less knowledgeable. Is there an online reference you could link to for background (similar to the labeling syringes reference)?
  7. Citations for table are unclear. Does 5 and 6 apply to all antibiotics in table? Excluding Cefepine and Meropenem which have their own? Clarify table citations.
  8. Under Considerations, first point, might mention \"oversight for unforeseen adverse effects\". Whenever you want to go from relatively small studies to standard of care there\'s potential for unforeseen consequences.
  9. Your first two \"important considerations\" are repeated under Take-home points. Is it really necessary to repeat bulleted points so close to each other? Consider shortening, combining, deleting.
  10. In trying to forecast responses to this article: One shortcoming to this article is that it seems to try to overcome lack of access in septic patients, but some will say these patients should have multiple peripherals for fluid resuscitation anyway. Also, you mention pressors tying up a line, but most will say if the patient is on pressors there should be a triple lumen central catheter. Anyway to address this?
Matthew Zuckerman, MD
ALiEM-AAEM Social Media and Digital Scholarship Fellow
Assistant Professor of Emergency Medicine
University of Colorado, Anschutz Medical Campus

Hi Matthew,

  1. First and foremost, thank you for your timely review/copy-edits!
  2. I completely agree, and I have added a brief, informal scenario to drive this point home.
  3. Agreed. This ties in with the need for a brief scenario. I have addressed both of these issues in point #2 under \"The Problems\" section.
  4. Good point. Gone. Frankly, I was just surprised I could find anything that related IVP antibiotics and sepsis, since it makes a lot of sense.
  5. Thank you. Done.
  6. Add a brief blurb about extravasation, and linked out to a nice, recent review article on the topic.
  7. Thank you, perhaps the referencing was a bit confusing. I simply compiled them all together after the preface statement before the table.
  8. Good point. This has been added.
  9. I have cleaned this up a bit, and I think it is less redundant.
  10. In my mind, this point is relative and debatable. Multiple peripherals would certainly be advocated for, but not always achievable; it may help to work through a lack of access in these instances. Otherwise, it may simply be just one component among a comprehensive strategy to optimize antibiotic therapy in sepsis - more or a less a convenience factor.Also, the utility of IVP antibiotics in severe sepsis is likely to be greatest during the very initial part of resuscitation in the ED. This is also the point where we have seen a push towards earlier vasopressors, a shift towards less central line placement for initial resus, and this continued goal for early abx within 1 hour (which is pretty tough). In reality, many patients are being titrated on pressors peripherally during this initial period, with multiple competing priorities and evolving strategies for care. IVP antibiotics may simply help in these initial situations.
Adam Spaulding, PharmD, BCPS
Emergency Medicine Pharmacist & Pharmacy Residency Program Director
Waterbury Hospital Health Center
Adjunct Assistant Professor - UCONN School of Pharmacy
Contributor to Emergency Medicine PharmD Blog

Expert Peer Review

Thank you for this practical post! Overall it’s very succinct and I think readers will find it useful. My thoughts are below.

  1. Second sentence in the first paragraph is a bit of a run-on and becomes difficult to follow. Consider cutting out extraneous words/commas to be more concise.
  2. Third sentence in the first paragraph; rather than saying “the push towards less” consider saying “the shift away from…” or something that doesn’t sound quite as contradictory.
  3. For your fourth reference (mentioned in the second paragraph) consider adding in a tiny blurb as to how IVP administration may be more economically beneficial. This doesn’t need to be exhaustive and it may add some leverage to your point. A more thorough pharmacoeconomic evaluation of these two methods can be found here to give you some actual dollar figures, albeit in 1992 dollars.
  4. Would agree with Dr. Zuckerman’s point regarding clarification of the “SWFI” and “CPOE” abbreviations.
  5. For your table, are you using all of the doses mentioned as recommended doses? How would you handle higher doses than those listed (e.g. a cefepime 2 gram dose, would you recommend dilution as mentioned in reference 7)?
  6. Consider tweaking structure of the post to be similar to other Trick of the Trade posts. For example, start off with a section outlining “The Problem,” you could potentially use a case to illustrate the obstacle that you propose your trick could overcome (this really isn’t too far off from what you currently have). Then move on to “Trick of the Trade: Administer Antibiotics via IVP” and close with your considerations/take-home points sections. See Bryan Hayes “Trick of the Trade: Rapid Oral Loading Dose of Phenytoin” as an example.
  7. There has been some PK data showing administration of beta-lactam antibiotics over a longer period of time is beneficial (e.g. piperacillin/tazobactam infused over 4 hours) to take advantage of the time above MIC. In fact, in my institution our standard administration time for pip/tazo is 4 hours, built into the CPOE system and smart infusion pump software. This seems to be the opposite from what you’re recommending (although I don’t disagree with your point). You might consider clarifying the difference between administration of the first dose (get it in fast) versus subsequent doses (keep concentration above MIC).
  8. “Pushing” an antibiotic such as meropenem over 5 minutes may not translate into realistic/actual bedside practice. Would you recommend placing this on a syringe pump?
  9. Would you feel comfortable implementing these recommendations in pediatric patients?
Meghan Groth, PharmD, BCPS
Emergency Medicine Pharmacy Clinician
Fletcher Allen Health Care
Contributor for Emergency Medicine PharmD Blog

Hi Meghan,

  1. Thank you for reviewing the post! This first paragraph has been revised.
  2. Agreed. Also revised in the first paragraph.
  3. Excellent reference and I agree! I have added a bit more on the economic benefit as shown from the study, and included the study in my referencs. Thanks!
  4. Yes. Clarified.
  5. Thank you for this needed clarification. I have added an asterisk for the \"concentration\" column which refers to a little note at the bottom of the table to clarify this. These are not necessarily recommended doses; doses should be individualized. This column is simply the recommended concentration. As in my example below the table, if you wanted to give 2g of cefepime, you would double the volume (20mL SWFI) and administer over 4-8 min, or give two back to back 1g doses. I do acknowledge the concentration used in reference 7, however simply adjusting the volume of the diluent to maintain the same concentration as in the table will allow you to maintain the same osmolality.
  6. Thank you for this advice. I have tweaked the post accordingly.
  7. Excellent point. I have added this clarification as a separate point under \"important considerations\".
  8. Yes, you could administer this on a syringe pump over 5 minutes, certainly if you have a pump around. However, I would contend that a 5 min IVP is not unreasonable, even if it is uncommon, and is still much quicker than administering via IV piggyback. While push back from nursing may occur initially, there are actually many other common medications which should ideally be administered over a similar rate, whether or not it is actually done. In my previous post on D10 vs D50 for severe hypoglycemia, I acknowledge that even a medication as common as D50 should be ideally administered over 4-5 min. Same thing with an amp of bicarb.
  9. This is a really important point you bring up. While I have seen several protocols from pediatric facilities that allow various antibiotics to be given via IV push to neonates/children, none of the studies that I have referenced actually included children, specifically. In fact, the Garrelts, et al. 1988 article above excluded those < 18 years of age. I have clarified that the safety of the listed IVP antibiotics have been shown in \"adults\" in the statement above the table, and also added this as a point under \"important considerations\".
Adam Spaulding, PharmD, BCPS
Emergency Medicine Pharmacist & Pharmacy Residency Program Director
Waterbury Hospital Health Center
Adjunct Assistant Professor - UCONN School of Pharmacy
Contributor to Emergency Medicine PharmD Blog

Expert Peer Review

Thank you for the opportunity to review this post. It provides an important tool, that many clinicians may not be aware of, that could potentially improve patient care. This knowledge could also have implications in under-served/under-resourced care environments (such as global health missions or frontier medicine) where supplies and equipment may be limited. My comments are as follows:

  1. I agree with the previous reviewer that the first section be titled \"The Problem\" to adhere to the format of previous posts
  2. The first section has a number of important points however I feel as though the 3 sections blend together and feel slightly bulky. Perhaps this info would be more clearly presented in outline format. Here are the takeaway points as I understand them (specific comments or suggested additions in parenthesis):
    • Intravenous access is a common conundrum in the ED
      1. (Patients may be \"hard sticks\" with limited peripheral access options)
      2. (Ultrasound and central venous access lines may take time and add to delays. This could address comment from a previous reviewer arguing that many of these patients will need central access anyway. Clarifies that IVP ABX may not prevent the need for a CVC, but gives you other options while better access is attained).
    • Septic patients are particularly vulnerable (suggest reorganizing first sentence so it does not begin with \"Because\")
      1. Great real world example of Abx not given when ordered/communication issues
      2. Shift away from CVC placement (could this shift be referenced to the updated surviving sepsis guidelines?)
    • Need for multiple antibiotics, pressors, fluids, ect. (especially in sepsis patients)
    • Timeliness of antibiotic administration associated with increased survival
      1. Supporting data from Surviving Sepsis Guidelines
  3. In the Trick of the Trade section, I agree that a nurse will be more likely to initiate an IVP antibiotic first, however it would not necessarily be due to simplicity. It would likely take more steps (albeit short steps), to reconstitute, draw up, and slowly push the medication as compared to a ready to use IVPB that is put on a pump. I would state that when faced with limited IV access, a nurse may be motivated to administer the medication with the shortest administration time first.
  4. The chart of antibiotics is great and including the osmolality is an important bit of supporting background information.
  5. In the Important Considerations section, it can\'t be stressed enough that this would require standardization and could be a significant diversion from how antibiotics are administered elsewhere in the hospital. The ongoing utilization of IVP dosing in an admitted patient who should be receiving prolonged infusions after the first dose could be potentially detrimental and systems should be in place to prevent this from occurring.
  6. In the Important Considerations section, paragraph 4, initial sentence is very long. Suggest re-wording to not start with \"While\" and also separating statements.
  7. In the Take home Points section, first line, suggest mentioning safety
  8. As the previous reviewer stated, are there data supporting this practice in the pediatric population as well? They are another population who frequently have limited IV access.
  9. After reading this post, I find myself asking, \"What other things should I be doing to help with this problem, particularly with septic patients?\" Answering this question may be outside the scope of this post and may also make it too lengthy, but the pharmacist in me wants to tackle this problem from every angle. Things that come to mind are:
    1. Having antibiotic compatibility charts readily accessible in the ED
    2. Educating staff on the importance of starting the gram negative agent first and timely antibiotic administration in sepsis
    3. Addressing the dogma that antibiotics must be separated to identify the offending agent in the case of allergic reaction
    4. Assure sepsis medications are stocked in the ED whenever possible
Robert S. Pugliese, PharmD BCPS
Clinical Specialist, Emergency Medicine
Clinical Assistant Professor
Thomas Jefferson University and Hospital

Hi Rob,

Thank you for taking the time to review and improve the quality of my post. See my responses below:

  1. This has been revised.
  2. I do see your point here. However, this section was previously in narrative format exclusively, and has since been broken up into 3 parts, as shown. I feel that its current format adequately conveys the various issues brought forth, namely: limited access and tying up lines; multiple abx and order of administration; and time to abx therapy. While perhaps a bit bulky, the three points do revolve around a common theme - the severely septic patient. I think the argument for IVP antibiotics can be made, and the value proposed, particularly within this patient population.
  3. Yes, good point. I have revised the statement to reflect your suggestion.
  4. Thank you!
  5. Kudos to that! I tried to reflect the importance of this within the first point.
  6. Good point. Revised.
  7. Another nice touch. Done.
  8. Yes, thank you. I have addressed this. See above comments and revised post.
  9. Completely any excellent EM pharmacist should. Also agreed however, that these other important points are outside the scope of this post, which is meant to be a focused, concise, trick of the trade.
Adam Spaulding, PharmD, BCPS
Emergency Medicine Pharmacist & Pharmacy Residency Program Director
Waterbury Hospital Health Center
Adjunct Assistant Professor - UCONN School of Pharmacy
Contributor to Emergency Medicine PharmD Blog
Adam Spaulding, PharmD BCPS

Adam Spaulding, PharmD BCPS

Emergency Medicine Pharmacist,
Pharmacy Residency Program Director,
Waterbury Hospital Health Center,
Adjunct Assistant Professor - UCONN School of Pharmacy,
Contributor to Emergency Medicine PharmD Blog
Adam Spaulding, PharmD BCPS

Latest posts by Adam Spaulding, PharmD BCPS (see all)