Perioperative Hypersensitivity: Recognition and Evaluation to Optimize Patient Safety

David A. Khan, MD; Kimberly G. Blumenthal, MD, MPH; Elizabeth J. Phillips, MD

Cefazolin is currently the most commonly identified cause of anaphylaxis in the United States, occurring in 1 in 10,000 surgeries; however, it is often overlooked.1 If anaphylaxis is suspected, serum tryptase drawn within 2 hours may be helpful to identify or differentiate the episode from other causes. Cefazolin is a first generation cephalosporin with R1 and R2 side chain groups that are distinct from other beta-lactms, and most patients with cefazolin allergy can tolerate penicillins and other cephalosporins.

What Does Cefazolin Anaphylaxis Look Like? A Real Life Case Example

Figure 1: Illustration of positive prick and intradermal skin tests to cefazolin (Cp) following two discrete episodes of anaphylaxis temporally associated with cefazolin two months prior. Prick testing to histamine (H) is positive as a control. Prick and Intradermal testing to saline is negative as a negative control. Intradermal skin tests to other reagents including ampicillin 25 mg/ml, benzyl penicillin 1000 and 10,000 IU/ml, minor determinant mixture (MDM) and major determinant (Pre-pen) were negative.

Figure 1: Illustration of positive prick and intradermal skin tests to cefazolin (Cp) following two discrete episodes of anaphylaxis temporally associated with cefazolin two months prior. Prick testing to histamine (H) is positive as a control. Prick and Intradermal testing to saline is negative as a negative control. Intradermal skin tests to other reagents including ampicillin 25 mg/ml, benzyl penicillin 1000 and 10,000 IU/ml, minor determinant mixture (MDM) and major determinant (Pre-pen) were negative.

A 50-year-old African-American woman presented to an Allergy and Immunology clinic following two suspected discrete episodes of anaphylaxis prior to an intended right hip replacement at an outside hospital. She gave no other significant past medical history. Her medication history was significant for longstanding use of rosuvastatin 20 mg orally and acetaminophen 500 mg orally as needed. On the first occasion, she received vancomycin 2g and cefazolin 1g prior to the planned joint replacement. No sedation or anesthetic agents had been given. Within minutes of the cefazolin infusion, she had skin flushing, facial and lip swelling, and hypotension. She received a single dose of epinephrine 0.3 mg intramuscularly, diphenhydramine 50 mg and hydrocortisone 125 mg both intravenously, and was transferred to the intensive care unit under observation for an additional day. Ten days later, preparation for right hip replacement was again attempted, and this time she received cefazolin 2 g (without vancomycin) and within minutes developed facial swelling and flushing; epinephrine 0.3 mg intramuscularly and diphenhydramine 50 mg intravenously were immediately given, and she was observed for several hours without recurrence. Two months later, she was seen in allergy clinic where she underwent skin prick testing to cefazolin followed by intradermal testing to penicillins, cefazolin, and ceftriaxone (Figure 1). Skin tests were weakly positive to cefazolin prick testing and strongly positive to intradermal cefazolin but negative to all other reagents. She tolerated challenges with amoxicillin and cephalexin, both 250 mg orally. Based on these data, she was diagnosed with cefazolin anaphylaxis and given advice that it was safe for her to take penicillins and cephalosporins other than cefazolin and also safe for her to take vancomycin. We emphasized that cefazolin should be clearly documented as a severe reaction (anaphylaxis) in all electronic health records and pharmacy records and she should wear a medic alert bracelet.

How to Identify Perioperative Hypersensitivity?

Perioperative hypersensitivity (POH) reactions are unexpected and unpredictable events that present suddenly without warning. The severity of reactions can range from mild reactions to severe anaphylaxis, which in some cases may be fatal. The incidence of POH ranges widely and by country of origin with recent studies suggesting an incidence of 1 in 10,000.2 The majority of cases of POH are thought to be allergic, caused by IgE-mediated mast cell activation. However, non-IgE-dependent mechanisms that activate mast cells may also occur in association with many drugs given perioperatively. Recently, the Mas-related G-proten-coupled receptor X2 (MRGPRX2) has been shown to be a cause of reactions to certain medications such as neuromuscular blocking agents, vancomycin, fluoroquinolones, and opioids. Radiocontrast dyes may also cause non-IgE mediated mast cell activation.

POH reactions typically present with cardiovascular and/or respiratory involvement including signs of hypotension, tachycardia, bronchospasm, and cardiac arrest. Mucocutaneous reactions such as erythema, urticaria, or angioedema can also occur, but may be missed due to draping of the patient. Cardiorespiratory symptoms are not specific for POH and may occur for a variety of other reasons such as medications, hypovolemia, underlying respiratory disease, and multiple attempts at intubations. Recently, an expert panel of anesthesia professionals and allergists developed a clinical scoring system to assist with determining the likelihood of a reaction being due to POH (Table 1).3 A weighted scale with points for or against POH are tabulated based on clinical parameters which produces a score yielding the likelihood of an immediate hypersensitivity reaction. This scoring system underwent content, criterion, and discriminant validity but has not undergone independent external validation.3

Table 1: Clinical scoring system for suspected perioperative hypersensitivity reactions.*

Cardiovascular System (CVS) (Choose hypotension, severe hypotension or cardiac arrest if appropriate, then any other items that apply) Points
Severe hypotension 6
Hypotension 4
Cardiac arrest 9
Tachycardia 2
A poor or unsustained response of hypotension to standard doses of sympathomimetics used to treat pharmacological hypotension during anaesthesia (e.g., ephedrine, phenylephrine, metaraminol) 2
A point-of-care echocardiogram showing a hyperdynamic and poorly-filled heart 2
Recurrence or worsening of hypotension after a further dose of a drug given prior to the initial event 1
Cardiovascular Confounders (choose any that apply)
Excessive dose of anaesthetic drug or drugs -2
Surgically induced hypovolemia or relative hypovolemia from prolonged fasting/dehydration -1
Acute illness predisposing to hypotension -1
Medications affecting cardiovascular responses during anaesthesia -2
Neuraxial regional anaesthesia (epidural/spinal) -1
Onset of hypotension after development of increased peak airway pressure during mechanical ventilation of the lungs -2
Respiratory System (RS) (Choose bronchospasm or severe bronchospasm if appropriate, then any other items that apply)
Bronchospasm 2
Severe Bronchospasm 4
Recurrence or worsening of bronchospasm after a further dose of a drug given prior to the initial event 1
Bronchospasm occurring before airway instrumentation (having excluded airway obstruction) 2
Respiratory Confounders
Respiratory disease associated with reactive airways -1
Prolonged or multiple attempts at tracheal intubation -1
Inadequate dose of drugs to obtund airway responses prior to airway instrumentation -1
Dermal/mucosal (D/M) (Choose any items that apply)
Generalised urticaria 4
Angioedema 3
Generalised erythema 3
A generalised rash is itchy in the awake patient who has not received epidural/spinal opioids 1
Angioedema in a patient taking an ACE inhibitor -3
Combinations (Choose a maximum of one item)**
CVS>2 & RS > 2 5
CVS>2 & D/M >2 5
RS>2 & D/M >2 5
CVS>2 & RS>2 & D/M >2 8
Timing (Choose a maximum of one item)
Onset of cardiovascular or respiratory features within 5 min of possible IV trigger 7
Onset of cardiovascular or respiratory features within 15 min of possible IV trigger 3
Onset of cardiovascular or respiratory features within 60 min of possible non-IV trigger 2
Onset of cardiovascular or respiratory features more than 60 min after possible non-IV trigger -1
Likelihood of immediate hypersensitivity reaction Total (net) score
Almost certain >21
Very likely 15 to 21
Likely 8 to 14
Unlikely < 8
*Table modified from Hopkins PM, Cooke PJ, Clarke RC, et al. Consensus clinical scoring for suspected perioperative immediate hypersensitivity reactions. Br J Anaesth. 2019;123(1):e29–e37.

**For a score from one of the 3 organ systems (CVS, RS, D/M) to contribute to a combinations score, the net score for that system must be > 2. The net score is the sum of scores for positive features minus the sum of scores for confounders within scores for that system.

The most useful laboratory test for helping to confirm an immediate hypersensitivity reaction is a serum tryptase level. Tryptase is a protease released by mast cells during anaphylaxis and is specific for evidence of mast cell activation. A tryptase level is ideally obtained within 2 hours of a reaction and will not be affected by medications used to treat a reaction. Elevations in tryptase have high positive predictive values (82–99%) for anaphylaxis in suspected perioperative reactions.4 However, patients may have anaphylaxis without an elevated tryptase level (> 11.4 ng/ml). A tryptase > 7.35 ng/ml has a 99% positive predictive value for POH in patients with severe cardiovascular collapse or cardiac arrest. An acute serum tryptase level greater than ([1.2 x serum baseline tryptase] + 2) can help confirm anaphylaxis (especially in those with normal acute tryptase levels) and has a positive predictive value for POH of 94%. Severe reactions can occasionally occur even with non-IgE mediated mast cell activation such as with vancomycin infused rapidly.5 In about 10% of cases, serum tryptase may be elevated in the setting of non-IgE mediated mast cell activation.

While there are many causes of POH, antibiotics, neuromuscular blocking agents, and disinfectants are among the most common causes documented internationally.2 In the U.S., however, cefazolin is the most commonly identified cause of POH, reported to be associated with >50% of cases of POH, and may occur with the first exposure to cefazolin.6 The pathway of sensitization to cefazolin has not been determined. It is important to recognize that currently most episodes of anaphylaxis to cephalosporins are thought to be related to the R1 side chains. Cefazolin has R1 and R2 side chains that are distinct and not shared with any other beta-lactams used in North America. In keeping with this, patients allergic to cefazolin are typically not allergic to penicillin or other cephalosporins.6

Why is Testing Important in the Setting of Perioperative Hypersensitivity?

After a suspected allergic reaction in the perioperative setting, it is critical to use all possible diagnostic investigational tools to identify the culprit drug. First, it is possible that the event occurred prior to the surgery and the surgery was aborted at the time of the allergic reaction, but is still needed or recommended. Second, even if that specific surgery was completed despite the reaction, most patients will require subsequent anesthesia in the future. International consensus recommendations by an expert panel of anesthesia professionals and allergists recommend a comprehensive allergy evaluation, ideally with collaboration between anesthesia professionals and allergists, for all patients with perioperative allergic reactions.7 The ideal time for allergy evaluation is thought to be approximately 4 weeks after the event; however, this is not evidence-based and testing up until at least 6 months after the event is still useful.8 For patients allergic to cephalosporins, approximately 60–80% will lose skin test reactivity 5 years after acute anaphylaxis.9

It is important that the allergist have access to the full anesthesia record and operative note that includes timing of administration and any other detail related to potential exposures including disinfectants, latex, lubricants, contrast, dyes, gelatin sponges used for hemostasis, foreign devices, and local anesthetics and their timing in relation to the suspected allergic event. This will help guide the appropriate assessments. After a negative skin test, observed challenges whereby a full dose of a drug is given to the patient in an observed setting in 1 or 2 steps is performed to all potential culprit medications that are possible in an outpatient clinic setting. Most allergy practices do not do intravenous challenges and challenges cannot be safely performed in outpatient allergy practices to opioids, benzodiazepines, neuromuscular blocking agents, and propofol. As such, negative skin testing to agents not challenged by the allergist warrant a “test dose” to be given immediately prior to subsequent anesthesia. The best evidence to date from the U.S. suggests that we will identify at least one-third of culprits with this strategy by finding patients with positive skin tests.8 Approximately 9 in 10 of the patients assessed by an allergy specialist, regardless of whether they have a positive skin test, will tolerate their subsequent anesthesia without a recurrent allergic reaction.8

Several patients presenting for surgery report allergies to penicillin or cefazolin without a formal allergist evaluation. Therefore, perioperative providers often seek alternative prophylactic antibiotics to administer to these patients. However, alternative perioperative antibiotics for surgical site infection prophylaxis can result in an increased risk of prophylaxis failure and infection.10 Furthermore, perioperative clindamycin and vancomycin are associated with an increased risk of Clostridium difficile colitis and acute kidney injury. Additionally, a patient with a cefazolin allergy documented in their electronic health record without specific allergist guidance may avoid all beta-lactam antibiotics, which can have additional clinical implications in the future.11 If cefazolin is identified as the causative drug, tolerance to other beta-lactams can be confirmed through specialized allergy testing.

Most patients in the U.S. who are labeled as being allergic to penicillin report histories of benign rashes, remote reactions, or unknown reactions. Patients with these low-risk histories can safely receive cefazolin for surgical prophylaxis.6 Recently, a multi-disciplinary group at Emory University developed a simple algorithm for administration of cefazolin or cefuroxime for perioperative antibiotic prophylaxis to patients with histories of penicillin allergy.12 If patients did not have histories of severe drug reactions such as Stevens-Johnson syndrome or toxic epidermal necrolysis, drug reaction with eosinophilia and systemic symptoms, liver or kidney injury, anemia, fever, or arthritis following pencillin, they were allowed to receive cefazolin or cefuroxime. After implemnation of this simple algorithm, cephalosporin use increased nearly 4-fold in penicillin allergic subjects and none of the 551 penicllin allergic patients who received a cephalosporin had an immediate allergic reaction. Another study revealed that even in patients with confirmed anaphylactic penicillin allergy, the risk of cefazolin allergy is < 1%.13

For patients with POH, we recommend a standard approach that optimizes patient safety and includes anesthesia professionals working with an allergist/immunologist to identify the culprit or compose a logical plan for subsequent anesthesia that minimizes allergic risk, while considering the importance of antibiotic prophylaxis.1 If the anesthesia professional does not have an allergist to work with, allergists can be found through a professional society link ( Additionally, with the expansion of telemedicine that accompanied the COVID-19 pandemic, virtual appointments with allergists at the major academic medical centers that routinely evaluate these patients, for formal triage and risk stratification preceding specialized testing is now possible.

In summary, POH reactions are rare events with many potential culprits with cefazolin being the most frequent in the U.S. Collaboration between anesthesia professionals and allergists can help to identify the culprit and develop a plan for safe administration of anesthesia in the future. While most patients with a label of penicillin allergy are not truly allergic, the vast majority of patients with an unverified penicillin allergy can receive cefazolin for surgical prophylaxis without increased risk of an allergic reaction.


David A. Khan, MD, is professor of medicine in the Division of Allergy & Immunology at the University of Texas Southwestern Medical Center, Dallas, TX.

Kimberly G. Blumenthal, MD, MPH, is assistant professor of Medicine at Harvard Medical School in the Division of Rheumatology, Allergy and Immunology at Massachusestts General Hospital, Boston, MA.

Elizabeth J. Phillips, MD, is a professor of medicine and John A. Oates Chair in Clinical Research in the Department of Medicine at Vanderbilt University Medical Center, Nashville, TN.

The authors have no conflicts of interest.


  1. Saager L, Turan A, Egan C, et al. Incidence of intraoperative hypersensitivity reactions: a registry analysis. Anesthesiology. 2015;122:551–559.
  2. Mertes PM, Ebo DG, Garcez T, et al. Comparative epidemiology of suspected perioperative hypersensitivity reactions. Br J Anaesth. 2019;123:e16–e28.
  3. Hopkins PM, Cooke PJ, Clarke RC, et al. Consensus clinical scoring for suspected perioperative immediate hypersensitivity reactions. Br J Anaesth. 2019;123:e29–e37.
  4. Volcheck GW, Hepner DL. Identification and management of perioperative anaphylaxis. J Allergy Clin Immunol Pract. 2019;7:2134–2142.
  5. Alvarez-Arango S, Yerneni S, Tang O, et al. Vancomycin hypersensitivity reactions documented in electronic health records. J Allergy Clin Immunol Pract. 2020.
  6. Khan DA, Banerji A, Bernstein JA, et al. Cephalosporin allergy: current understanding and future challenges. J Allergy Clin Immunol Pract. 2019;7:2105–2114.
  7. Garvey LH, Dewachter P, Hepner DL, et al. Management of suspected immediate perioperative allergic reactions: an international overview and consensus recommendations. Br J Anaesth. 2019;123:e50–e64.
  8. Banerji A, Bhattacharya G, Huebner E, et al. Perioperative allergic reactions: allergy assessment and subsequent anesthesia. J Allergy Clin Immunol Pract. 2020;S2213–2198.
  9. Romano A, Gaeta F, Valluzzi RL, et al. Natural evolution of skin-test sensitivity in patients with IgE-mediated hypersensitivity to cephalosporins. Allergy. 2014;69:806–809.
  10. Berríos-Torres SI, Umscheid CA, Bratzler DW, et al. Centers for Disease Control and prevention guideline for the prevention of surgical site infection, 2017. JAMA Surg. 2017;152:784–791.
  11. Blumenthal KG, Peter JG, Trubiano JA, Phillips EJ. Antibiotic allergy. Lancet. 2019;393:183–198.
  12. Kuruvilla M, Sexton M, Wiley Z, et al. A streamlined approach to optimize perioperative antibiotic prophylaxis in the setting of penicillin allergy labels. J Allergy Clin Immunol Pract. 2020;8:1316–1322.
  13. Romano A, Valluzzi RL, Caruso C, et al. Tolerability of cefazolin and ceftibuten in patients with IgE-mediated aminopenicillin allergy. J Allergy Clin Immunol Pract. 2020;8:1989–1993.e1982.