Date: September 30, 2026 

Dr. Aaron Skolnik

Guest Skeptic: Dr. Aaron Skolnik is an Assistant Professor of Emergency Medicine at the Mayo Clinic Alix School of Medicine and Vice Chair of Critical Care Medicine at Mayo Clinic Arizona. Aaron is a full-time multidisciplinary intensivist and is the medical student clerkship director for critical care.

Reference: ZhangJian et al. Incidence of Adverse Events in Peripheral Intravenous Vasopressor Use: A Systematic Review and Meta-Analysis. JAMA Netw Open. 2026

Case: A 72-year-old woman presents by EMS with fever, productive cough, confusion and progressive weakness. Her temperature is 39.2°C, HR 118 beats/min, RR 28 breaths/min, oxygen saturation 91% on room air and BP 78/44 mmHg. She has coarse crackles over the right lower lung field and cool extremities. Chest radiography demonstrates a right lower-lobe infiltrate. Her lactate is 4.1 mmol/L. You establish two ultrasound-guided 18-gauge peripheral IVs, administer appropriate antibiotics and begin judicious crystalloid resuscitation. Her blood pressure remains 82/46 mmHg with a MAP of 58 mmHg. The nurse asks: “Do you want me to start the norepinephrine through the peripheral IV while you put in the central line?”

Background: Shock is one of those diagnoses where time matters. At its core, shock is inadequate cellular oxygen utilization, resulting in tissue hypoperfusion. If left unchecked, this process leads to organ dysfunction, cardiovascular collapse, and death. For many of our patients, fluids alone will not restore adequate perfusion, and that is where vasopressors come in. They increase vascular tone and blood pressure, with norepinephrine generally being our first-line vasopressor in distributive/septic shock.

Traditionally, we were taught that vasopressors belong in a central venous catheter (CVC). The rationale makes physiologic sense. These are potent vasoconstrictors, and extravasation into surrounding tissue could theoretically produce local ischemia, tissue necrosis, or even limb-threatening injury. A catheter sitting in a large central vein provides rapid dilution of the medication and avoids the small peripheral vessels where infiltration might cause trouble.

The problem is that central lines are not benign and certainly aren’t instantaneous, as portrayed on TV shows. Placement takes time, equipment, expertise, and often ultrasound guidance. Complications include arterial puncture, bleeding, pneumothorax, arrhythmia, venous thrombosis, and catheter-related bloodstream infection. Meanwhile, the patient’s MAP is still circling the drain.

The alternative is sitting right there in the patient’s arm: a peripheral IV that may already be functioning. Starting norepinephrine through that line allows us to treat the shock now and sort out longer-term vascular access later. The 2021 Surviving Sepsis Campaign reflected this shift by recommending initiation of vasopressors peripherally rather than delaying therapy until central access is obtained.

That brings us to the bedside tension. We don’t want to delay a potentially life-saving drug while waiting for the perfect vascular access (perfect is the enemy of good). However, we also don’t want norepinephrine to infiltrate into someone’s forearm. The relevant question isn’t whether peripheral vasopressors are perfectly safe, because nothing we do in emergency medicine is perfectly safe. It is about how often clinically important adverse events occur, under what circumstances, and whether peripheral administration can spare patients the need for an invasive central line. You could say we are under a lot of pressure to make the right decision.

Previous systematic reviews suggested that complications from peripheral vasopressors were uncommon and predominantly minor. However, those reviews had important gaps, including limited agent-specific information, relatively little information about midline catheters, and limited data addressing whether peripheral administration avoids CVC placement.

Clinical Question: In critically ill adults requiring vasopressors, what is the incidence of adverse events associated with peripheral intravenous vasopressor administration, and how often does this strategy avoid placement of a central venous catheter?

  • Population: Adults ≥18 years with hypotension, shock, critical illness, or risk of shock requiring vasopressor therapy. Eligible designs were RCTs, prospective or retrospective observational cohort studies, and case series containing at least ten patients.
    • Excluded: Case reports, pediatric studies, conference abstracts and editorials. Studies were also excluded for failure to report adverse events, absence of critically ill patients, inappropriate study design, inability to extract data, uncertain/nonperipheral vasopressor administration, duplicate populations, bolus rather than infusion administration, or inclusion of pediatric data.
  • Intervention: Vasopressor administration through a peripheral intravenous catheter
  • Comparison: There was no conventional control group for the principal analysis.
  • Outcome:
    • Primary Outcome: Pooled incidence of localized adverse events associated with peripheral vasopressor administration, expressed per catheter and divided into minor and major adverse events.
    • Secondary Outcome: Pooled proportion of peripheral catheters for which CVC placement was avoided.
  • Type of Study: Systematic review and meta-analysis of predominantly observational incidence/prevalence studies

Authors’ Conclusions: “In this systematic review and meta-analysis of adult patients, AE incidence was low after short-term vasopressor administration through PIV catheters, particularly short catheters. These findings suggest that PIV administration might reduce the need for CVC placement with appropriate monitoring.”

Quality Checklist for Systematic Review:

  1. The main question being addressed should be clearly stated. Yes
  2. The search for studies was detailed and exhaustive. Yes
  3. Were the criteria used to select articles for inclusion appropriate? Yes
  4. Were the included studies sufficiently valid for the type of question asked? Unsure
  5. Were the results similar from study to study? No
  6. Were there any financial conflicts of interest? No apparent major financial conflict

Results: The review included 49 studies published from 1997 to 2025, representing 33,060 peripheral IV catheters used for vasopressor administration in critically ill adults. Patients’ mean or median age across studies ranged from 36 to 81 years. Most studies were conducted in ICUs, with others from EDs, operating rooms, wards, intermediate care units, or mixed settings. Most patients received vasopressors through short peripheral IV catheters; four studies used midlines, and one used both. Norepinephrine was by far the most commonly studied agent, used in 23,118 catheters (70%).

Key Result: Short-term peripheral vasopressor administration was associated with a low pooled incidence of adverse events, with major events being extremely rare, while central venous catheter placement was avoided in about 60% of cases.

  • Minor Adverse events
    • Among 31,212 catheters, there were 735 minor AEs
    • Pooled incidence across all vasopressors was 2.3% (95% CI; 1.5%–3.7%; I² = 95.2%).

  • Major Adverse events
    • There were 31 major AEs in total: 30 venous thromboembolic events and one case of tissue necrosis requiring fasciotomy.
    • Importantly, the results differed substantially according to catheter type. All 30 VTEs occurred in the four studies using midline catheters (1,126 catheters), giving a pooled incidence of 1.4% (95% CI 0.4%–5.4%; I² = 85.2%).
    • Among 43 studies using short PIV catheters (29,596 catheters), there was only one major AE, the tissue-necrosis event, corresponding to a pooled incidence of 0.0% (95% CI 0.0%–0.0%; I² = 0%).
    • When both catheter types were considered together, the pooled incidence of major AEs was 0.0% (95% CI 0.0%–0.2%; I² = 0%).
  • Secondary Outcome: Pooled proportion of cases in which central venous catheter placement was avoided while vasopressors were administered peripherally.
    o Reported in 38 studies involving 15,371 catheters, with a pooled CVC-avoidance proportion of 59.7% (95% CI 46.4%–71.7%).

1. Garbage In, Garbage Out: This is a big meta-analysis (33,060 catheters) but a big denominator does not necessarily mean high-quality evidence. Thirty-one of the 49 studies were retrospective, only four were RCTs, and 61% were rated as unclear or high risk of bias. That matters when you’re looking for adverse events. If an infiltration or minor extravasation wasn’t recognized and documented in the chart, a retrospective study probably won’t find it. In fact, retrospective studies reported fewer minor AEs than prospective studies and RCTs. Meta-analysis can improve precision, but it can’t make bias disappear.

2. Apples & Oranges: There was a lot of heterogeneity. The I² was 95% for minor adverse events and almost 99% for CVC avoidance. These studies included OR, ED and ICU patients receiving different vasopressors, doses and concentrations through different catheter sizes and locations for very different lengths of time. So that 2.3% number is an average of some very different clinical situations. Don’t assume it represents the risk for the patient in front of you.

3. What Exactly Are We Measuring? Only about one-third of the studies reported catheter gauge and a similar proportion reported catheter location. Infusion duration ranged from minutes to days. Even the outcome definitions were broad. “Minor adverse event” included everything from pain and swelling to extravasation, cellulitis and thrombophlebitis. Those aren’t necessarily equivalent events. Pooling them gives us a number, but we need to be careful about what that number means clinically.

4. Rare Doesn’t Mean Zero: There was only one major tissue injury among almost 30,000 short PIV catheters. That’s reassuring. But a pooled estimate of 0.0% does not mean the risk is zero. Rare events are exactly the events most likely to be missed by retrospective chart review or incomplete follow-up. There is another wrinkle: the denominator was catheters, not necessarily individual patients. That’s not quite the same question our patient is asking when they want to know, “What are the chances this will hurt me?”

5. Incidence Is Not Safety: This may be the most important point. There wasn’t a central-line control group for the main analysis. So, this review tells us how often complications were observed with peripheral vasopressors; it does not tell us that peripheral vasopressors are safer than, as safe as, or better than central administration. Likewise, the 60% CVC-avoidance number doesn’t prove that the peripheral strategy caused those central lines to be avoided. This is reassuring descriptive evidence, not comparative evidence.

Comment on Authors’ Conclusion Compared to SGEM Conclusion: We would be a little more cautious. This SRMA provides reassuring but low-certainty evidence that major complications are uncommon with short-term vasopressors through short PIVs. It supports starting a clinically indicated vasopressor through a well-functioning PIV rather than delaying treatment for central access. It does not establish that peripheral administration is universally safe or should replace central access.

SGEM Bottom Line: Don’t let the perfect line be the enemy of the good line. If a patient needs vasopressors now, starting them through an appropriate, closely monitored short PIV appears reasonable while you decide whether they need a central line.

Case Resolution: You don’t wait for a central line. Norepinephrine is started through the 18-gauge peripheral IV, the site is checked frequently, and the dose is titrated to her blood pressure and perfusion. Over the next several hours, her blood pressure and mental status improve. Her vasopressor requirement falls, and the norepinephrine is stopped without infiltration or extravasation. She never needs a central line and is admitted to the ICU.

Clinical Application: These data support starting a vasopressor through an appropriate short PIV rather than delaying treatment solely to obtain central access. They should not be interpreted as permission to run concentrated vasopressors through any peripheral line indefinitely. For us, this supports a pressors-first, central-line-if-needed approach rather than a rigid central-line-before-pressors approach.

What Do I Tell the Patient? Your blood pressure is dangerously low, and we need to give you medicine to bring it up. We can start it right away through the IV in your arm. Serious problems from doing this for a short time appear to be very uncommon, but the medicine can leak out of the IV and damage the surrounding tissue. We’ll check the IV often. If you need the medicine for longer, we may need to put an IV into a larger vein.

Keener Kontest: There was no winner last week. A double-dummy trial is when each participant receives both treatment routes, but only one contains the active medication.

Remember to be skeptical of anything you learn, even if you heard it on the Skeptics’ Guide to Emergency Medicine.