Immediate Hypersensitivity to Chlorhexidine: Experience from an Allergy Center in China

This retrospective single center study involved all patients who underwent chlorhexidine allergy testing in the Allergy Center of West China Hospital, Sichuan University (Chengdu, China), in the period February 2018 to May 2022. In our center, chlorhexidine allergy testing is routinely performed in all patients with perioperative allergic reactions and who suffered allergic reactions to more than two kinds of infusion drugs or even saline or glucose IV infusion. We obtained clinical written informed consent from all subjects before testing. This retrospective study was approved by the Biomedical Research Ethics Committee of West China Hospital of Sichuan University, China (approval No. 2022[1937]).

Age, sex, clinical setting, detailed symptoms, management, diagnosis, and accidental re-exposure thereafter were obtained for all patients. The severity of allergic reactions was classified according to the Ring and Messmer system (I, cutaneous signs: generalized erythema, urticaria, angioedema; II, measurable but not life-threatening symptoms: cutaneous signs, hypotension, tachycardia, respiratory disturbances; III, life-threatening symptoms: collapse, tachycardia or bradycardia, arrhythmias, bronchospasm; IV, cardiac and/or respiratory arrest).¹²  The diagnostic procedure for chlorhexidine allergy is shown in figure 1.

A skin prick test was carried out using chlorhexidine (2%, Fujian Weizhenyuan Pharmaceutical Technology Link Company, China) diluted to 5 mg/ml¹³  with positive (histamine 10 mg/ml) and negative (diluent) controls (ALK-Abello, Denmark). A positive skin prick test reaction was defined as a wheal size of 3 mm greater than that of the negative control at 15 to 20 min. The coexistence of a flare and itch supported a positive result.¹⁴ 

If skin prick test was negative, an intradermal test was performed with 0.002 mg/ml¹³  chlorhexidine. An increase in wheal size of more than 3 mm from the initial papule with flare at 15 to 20 min was considered a positive result.¹⁴ 

Chlorhexidine serum specific immunoglobulin E was measured by the UNICAP System (Thermo Fisher, Sweden) for patients whose skin prick test or intradermal test result was positive. A specific immunoglobulin E level 0.35 kUA/l or greater was deemed positive.

No validated provocation model is available for chlorhexidine. We defined chlorhexidine allergy as one or more relevant clinical reactions to chlorhexidine in combination with two positive tests.⁵  Finally, patients diagnosed with chlorhexidine allergy were included in our study.

All other drugs, especially antibiotics and all the drugs used perioperatively, were also included in the allergy test in accordance with the position paper.^10,12,15  In our cohort, for patients diagnosed with chlorhexidine allergy, povidone iodine skin test was performed to choose alternatives.

In June 2022, all included patients were followed up by telephone and asked whether they had been accidentally re-exposed to chlorhexidine after the diagnosis.

The data were processed with IBM SPSS Statistics 21.0 (USA), and descriptive statistics were reported. Mean ± SD or median and interquartile range were used to describe continuously scaled variables, and frequency counts (%) were used for categorical variables.

Forty-three patients investigated in our center in the period underwent chlorhexidine allergy testing; of these, 10 patients (median age, 40.5 yr [interquartile range, 19.0 to 57.0 yr]; 5 [50.0%] female) were positive based on allergy testing (fig. 1). The clinical data and drug allergy testing results are summarized in table 1. Only one patient (10%) had a history of allergic rhinitis, while the others had no atopic comorbidities. The mean time interval between the last allergic reaction and the evaluation was 4.5 months (interquartile range, 1.0 to 13.5 months).

Ten patients experienced a total of 30 allergic reactions to chlorhexidine (mean ± SD, 3.0 ± 1.3), and the details are summarized in table 2. Two patients experienced two allergic reactions, and three patients experienced four allergic reactions during general and local anesthesia, respectively. During these six allergic reactions (6 of 30, 20%), the patients may have been exposed to chlorhexidine via different routes, such as skin disinfectants, central venous catheters, urethral gels, or their combinations. Only one allergic reaction (1 of 30, 3%) occurred with exposure via a mouthwash. The other 23 allergic reactions (23 of 30, 77%) occurred via a skin disinfectant; the route of exposure was IV cannulation in 22 allergic reactions (22 of 23, 96%) and broken skin in one allergic reaction (1 of 23, 4%).

The symptoms included a rapid onset and ranged in severity from skin rash to anaphylactic shock, and in some cases cardiac arrest. The percentages of grade I, II, III, and IV reactions were 8 (27%), 11 (37%), 5 (17%), and 6 (20%), respectively. Grade II to IV reactions accounted for 73% (22 of 30) of all reactions. However, epinephrine was used in only 14% (3 of 22), and 50% (11 of 22) of these reactions were treated with dexamethasone (table 2).

All included patients were diagnosed by skin prick and serum specific immunoglobulin E tests (fig. 1). The wheal diameters on skin prick test ranged from 6 mm to 14 mm (mean ± SD, 11.8 ± 2.7 mm); of these, the wheal diameters were larger than 10 mm in eight patients. The specific immunoglobulin E values ranged from 0.76 kUA/l to 43.5 kUA/l (mean ± SD, 12.06 ± 12.77 kUA/l), which were graded as class II (0.71 to 3.5 kUA/l) in four patients, class III (3.51 to 17.5 kUA/l) in four patients, and class IV (17.6 to 50 kUA/l) in two patients (table 1).

The other drug allergy tests and their results are shown in table 1; all of the additional allergy tests were negative. All the patients in our cohort completed the povidone iodine allergy test, and the results were all negative. Thus, povidone iodine could have been used as an alternative disinfectant in these patients.

All patients diagnosed with chlorhexidine allergy were followed up by telephone, and none were lost to follow-up. The time between diagnosis and follow-up ranged from 1 month to 52 months (median time, 21.5 months [interquartile range, 6.7 to 32.2 months]). Two patients (2 of 10, 20%) had been accidentally re-exposed to chlorhexidine after diagnosis. Among them, patient number 4 experienced anaphylactic shock after accidental exposure to a mouthwash containing chlorhexidine during dental treatment, even though she told her dentist that she was allergic to chlorhexidine. Patient number 6 experienced anaphylaxis during acupuncture treatment due to the skin disinfectant used before the puncture.

Our study showed that most of the allergic reactions to chlorhexidine were attributed to the skin disinfectants used before IV cannulation in China. Recently, a narrative review showed that topical chlorhexidine exposure before insertion of a peripheral venous cannula rarely caused systemic anaphylaxis.¹⁶  In 2016, Sharp et al. reviewed the literature about chlorhexidine allergy and found that urethral gels and central venous catheters were the most common culprits (more than 80%) in 68 published case reports.¹⁷  The reasons for the distinction were as follows.

First, skin disinfectants containing chlorhexidine used for skin preparation before IV cannulation are routinely used in medical institutions in China, which will increase the likelihood that more people will be exposed to chlorhexidine than with other methods, such as urethral gels and central venous catheters. In the long run, whether this will lead to more chlorhexidine sensitization and an increased incidence of allergies in China is a problem that needs attention. Although some studies have shown that chlorhexidine exerts a better effect than povidone iodine on infection related to intravascular catheters¹⁸  or clean-contaminated surgery,¹⁹  some experts have argued that chlorhexidine should not be used before short-term IV cannulation or for disinfecting needleless connector access points.²⁰  Based on our experience, we suggest reinforcing the awareness of the potential harm in the overzealous use of chlorhexidine in areas where there is little or no evidence that it reduces infection better than other, less allergenic antiseptics, especially used for IV cannulation.

Second, according to a previous study, the sensitization potential increases with increasing chlorhexidine concentrations.²¹  In China, the Health Ministry has set a maximum concentration of 4.5% on skin,²²  which is much higher than in many other countries, such as the United Kingdom (2%)²³  and Japan (0.5%),²⁴  and may lead to more sensitization to chlorhexidine. Decreasing the recommended concentration might reduce the possibility of sensitization.

Third, although standard clinical practice requirements in China recommend that the antiseptic should be dry before IV cannulation, it is possible that it does not have an adequate time to dry; alternatively, cannulation has occurred through pooled chlorhexidine. When a patient is sensitized to chlorhexidine, even a small amount to the circulation can cause an allergic reaction.

Although increasing numbers of cases of chlorhexidine allergy have been reported in the literature¹  and numerous governmental warnings have been issued during recent decades to warn of the risk of this allergy,^24–26  chlorhexidine is not often recognized as a potential allergen by healthcare workers and is easily overlooked. In our study, two patients (numbers 3 and 5) developed local skin rashes after blood sampling, but this minor allergic symptom was often ignored by medical staff and patients, and the patients subsequently suffered a more severe allergic reaction. A previous study also suggested that mild allergic reactions to chlorhexidine can precede anaphylaxis.²⁷  Moreover, most patients in the current study experienced repeated allergic reactions (mean ± SD, 3.0 ± 1.3) and even several severe life-threatening reactions before being referred and diagnosed. During the process, the infusion drug was always blamed as the culprit, especially if it was an antibiotic, at which point it was changed to another infusion drug, and the patient experienced a reaction again. Therefore, chlorhexidine allergy should potentially be considered for patients who complain of allergic reactions after different drug infusions after its use as a disinfecting agent. It is often used as part of surgical preparation and in the perioperative setting. International guidelines on perioperative hypersensitivity investigations recommend testing chlorhexidine in all patients with suspected perioperative hypersensitivity.^8–10 

For chlorhexidine allergy testing, Opstrup et al. reported that both the estimated sensitivity and specificity were high for the skin prick test (sensitivity 95% and specificity 97%) and serum specific immunoglobulin E test (sensitivity 100% and specificity 97%). Moreover, when the skin prick test and serum specific immunoglobulin E test were combined, the highest sensitivity and specificity were observed compared to combinations of other tests (such as intradermal test and histamine release test).⁵  In our study, all included patients were diagnosed by skin prick and serum specific immunoglobulin E tests, and the results were concordant with their clinical history. This further verified that the skin prick test and serum specific immunoglobulin E test both have a high sensitivity and specificity for chlorhexidine allergy diagnosis. However, to date, there have not been any commercially available chlorhexidine serum specific immunoglobulin E tests in China. If the clinical history indicates that the culprit is chlorhexidine, the skin prick test should be considered as an important screening and diagnostic method.

The chlorhexidine specific immunoglobulin E concentration is dynamic, increasing after exposure followed by a decline over time if further exposure is avoided.²⁸  In this study, specific immunoglobulin E was greater than 0.35 kUA/l in 22 of 23 patients at a median time of 10 weeks after the allergic reaction and declined to less than 0.35 kUA/l in 17 of 23 patients (most rapidly within 4 months) in the following months or years. Consequently, the optimal sampling time seems to be more than 1 month and less than 4 months. In another study, specific immunoglobulin E was determined in the sera of 14 patients at an average of 29 months (range 1 to 61 months) after diagnosis, and only six patients were positive.²⁹  In our cohort, all levels were markedly elevated at a median time of 4.5 months, even at 26 months after the last allergic reaction. The high levels might be due to their repeated exposure and the relatively severe systemic symptoms.

For patients diagnosed with chlorhexidine allergy, re-exposure should be avoided. However, a study from the Danish Anesthesia Allergy Center showed that 35% of patients were accidentally re-exposed after diagnosis.²⁸  Another study from the United Kingdom showed that 1 patient was re-exposed to chlorhexidine and developed anaphylaxis in 18 patients diagnosed with chlorhexidine allergy.⁶  In our study, two patients (20.0%) were also accidentally re-exposed. One reason for the difficulty in avoiding re-exposure is that chlorhexidine is widely used in the healthcare setting and may not be clearly labeled in products. Another reason is that the ingredients are not carefully checked by patients or healthcare workers. Therefore, avoiding re-exposure requires concerted efforts from multiple parties, including healthcare workers, patients, and manufacturers.

In cases of chlorhexidine allergy, tests against other allergens may also be positive, so when chlorhexidine is positive on testing, all other potentially relevant drug exposures should still be tested.^10,30  In our study, all other relevant exposures were tested, and any allergies to the associated agents were ruled out. Previous studies have shown that 25 to 35% of patients with chlorhexidine allergies were also positive for other potential culprits (such as latex, opiates, β-lactams, and neuromuscular blocking agents).^5,31,32  The reason for the difference may be attributed to the small sample size, and allergic reactions during the perioperative period were relatively rare in our study. More research on multiple sensitizations is needed.

There is an international consensus that timely administration of epinephrine is required for grade II to IV reactions.^33,34  In our study, although the majority of reactions were not life-threatening, a significant proportion were. Grade II to IV reactions accounted for 73% of the reactions. Only 14% of these reactions were treated with epinephrine, while 50% were treated with dexamethasone. Recently, Jiang et al. assessed the initial treatment for anaphylaxis in China by performing a systematic analysis of published case reports from 2014 to 2018 and found that only 14.2% of the patients were appropriately treated with epinephrine as the first-line intervention. These findings highlight the critical gap between clinical practice in China and the recommendations in international guidelines, and targeted training is urgently needed for healthcare providers in China.³⁵ 

Our study has some limitations. First, this was a single-center study with a small sample size and may not reflect the status of chlorhexidine allergy throughout China. Second, the retrospective nature of this study means that patient recall bias could not be avoided. Thus, further prospective, large-sample, multicenter clinical studies should be conducted in the future.

This study summarizes the characteristics of chlorhexidine allergy in an allergy center in China. The results showed that the majority of the reactions to chlorhexidine were attributed to skin disinfectants, and IV cannulation was the most common exposure route. Most patients were finally diagnosed after repeated allergic reactions, suggesting that chlorhexidine allergy was easily overlooked. Thus, we suggest increasing awareness of the potential allergenicity of chlorhexidine, especially after IV cannulation or during the perioperation. In addition, the skin prick test and serum specific immunoglobulin E test both had high sensitivity and specificity in diagnosing chlorhexidine allergy.

This work was supported by grants to Dr. Meng from the 1·3·5 Project for Disciplines of Excellence, the Clinical Research Incubation Project from West China Hospital of Sichuan University (Chengdu, China; 2018HXFH026), and the Chengdu Science and Technology Bureau of Technology Innovation R & D Project (Chengdu, China; 2021-YF05-00473-SN); and to H. Zhang from the West China Nursing Discipline Development Special Fund Project at Sichuan University (Chengdu, China; HXHL21036).

The authors declare no competing interests.