Penicillin allergy labels associated with worse outcomes in bacterial pneumonia
Click Here to Manage Email Alerts
Key takeaways:
- Within 30 days of a bacterial pneumonia diagnosis, 69.2% of patients with a penicillin allergy label and 56.2% of those without a label were hospitalized.
- 40.1% of patients with a label and 35.3% of those without a label were diagnosed with acute respiratory failure within 30 days of their bacterial pneumonia diagnosis.
- Rates of mortality within 30 days of a bacterial pneumonia diagnosis included 6.2% of those with a penicillin allergy label and 5.7% of those without a label.
Patients with penicillin allergy labels experienced worse outcomes with bacterial pneumonia than patients who did not have these labels, according to a study published in The Journal of Allergy and Clinical Immunology: In Practice.
Clinicians should then consider risk mitigation strategies in treating these patients, Lauren W. Kaminsky, MD, PhD, a fellow in the section of allergy, asthma and immunology, department of medicine, Penn State College of Medicine, and colleagues wrote.
“The impact of penicillin allergy label on outcomes of specific infections, including bacterial pneumonia, is not well-studied,” Kaminsky told Healio.
“Because penicillin allergy label may delay the start of antibiotics and result in the use of alternative and broader-spectrum antibiotics, we investigated clinical outcomes, antibiotic use, and possible antibiotic adverse effects in patients with bacterial pneumonia and penicillin allergy label,” she continued.
The study design and results
The researchers examined data from 68,748 patients aged 18 years and older with bacterial pneumonia and a penicillin allergy label and 68,748 matched patients with bacterial pneumonia who did not have a penicillin allergy label.
Within 30 days of the diagnosis of bacterial pneumonia, 69.2% of those patients with the label were hospitalized, compared with 56.2% of those without the label (RR = 1.23; 95% CI, 1.22-1.24).
Similarly, 40.1% of the patients with the label and 35.3% without it were diagnosed with acute respiratory failure within 30 days of their bacterial pneumonia diagnosis (RR = 1.14; 95% CI, 1.12-1.15).
Intubation occurred among 8.6% of the patients with a label and 7.3% of those without one (RR = 1.18; 95% CI, 1.13-1.22), whereas intensive care was needed by 14% of the label group and 12.6% of the non-label group (RR = 1.11; 95% CI, 1.08-1.14), both within 30 days of the diagnosis, as well.
Thirty-day mortality rates included 6.2% for the patients with a penicillin allergy label and 5.7% for those without the label (RR = 1.08; 95% CI, 1.04-1.13).
Also within the 30-day window, 8.9% of patients with a label and 24.4% without a label received penicillin (RR = 0.37; 95% CI, 0.35-0.38). The patients with a label also had lower rates of macrolide (RR = 0.94; 95% CI, 0.92-0.95) and cephalosporin (RR = 0.92; 95% CI, 0.91-0.93) use compared with those without a label.
The patients with a label had greater uses of fluoroquinolones (RR = 1.53; 95% CI, 1.51-1.55), carbapenems (RR = 2.44; 95% CI, 2.33-2.55) and monobactams (RR = 11.3; 95% CI, 10.4-12.4), among other antibiotics, than the patients without a label within 30 days of diagnosis as well.
Further, 5.75% of patients with a label and 5.12% without developed a Clostridioides difficile infection (RR = 1.12; 95% CI, 1.07-1.17), whereas 0.83% of those with a label and 0.39% of those without developed a drug-induced rash (RR = 2.15; 95% CI, 1.86-2.48).
The researchers additionally found that 0.076% of the label group and 0.022% of the non-label group had severe cutaneous adverse reactions (RR = 3.47; 95% CI, 1.95-6.16). With a 23.6% rate, there were more cases of acute kidney injury among the label group than among the non-label group, which saw a 20.4% rate (RR = 1.16; 95% CI, 1.13-1.18).
Finally, the label group experienced a 0.32% rate for toxic liver disease, whereas the non-label group had a 0.22% rate (RR = 1.46; 95% CI, 1.19-1.8).
Conclusions and next steps
“In patients with bacterial pneumonia, penicillin allergy label was associated with worse clinical outcomes (hospitalization, acute respiratory failure, intubation, need for intensive level of care, and mortality), modified antibiotic regimens, and a higher risk of adverse drug reactions and Clostridioides difficile infections,” Kaminsky said.
These greater risks for worse clinical outcomes including mortality reinforce the need for risk mitigation strategies among patients with a penicillin allergy label who develop bacterial pneumonia, the researchers said, such as active surveillance and allergy de-labeling.
“As penicillin allergy label was associated with worse clinical outcomes in bacterial pneumonia, risk mitigation strategies should be considered for patients with penicillin allergy label, including history-taking, skin testing, and/or drug challenge, to actively address penicillin allergy label,” Kaminsky said.
Next, the researchers will expand their studies.
“As this study focused on adult patients, we are planning to investigate clinical outcomes of bacterial pneumonia in pediatric patients with penicillin allergy label,” Kaminsky said.
Reference:
- Bacterial pneumonia patients with a penicillin allergy label face worse clinical outcomes. https://www.aaaai.org/About/News/News/bacterial-pneumonia. Published Sept. 28, 2022. Accessed Sept. 29, 2022.
For more information:
Lauren W. Kaminsky, MD, PhD, can be reached at lkaminsky@pennstatehealth.psu.edu.
Perspective
Back to Top
This is a really interesting “big data” study that provides some novel, but not surprising, information. In this retrospective analysis, the authors found that patients with a penicillin allergy label and pneumonia were more likely to be admitted to the hospital, require ICU-level care, develop respiration failure, require intubation and die than patients with pneumonia without a penicillin allergy label.
This type of study adds to existing knowledge because this specific data surrounding pneumonia has been limited overall. This study adds more evidence to support that patients should be evaluated for penicillin allergy de-labeling. Ideally, this should be done prior to them becoming ill since time is of the essence in treating with appropriate antibiotics in severe pneumonia.
Perspective
Back to Top
This is a great paper that highlights the need for further intervention on penicillin allergy labels at a societal level. With the large-scale data in this report, the bad effects of reporting a penicillin allergy on the outcomes of pneumonia treatment are fairly clear.
We have known for a while that 95% or more of the patients who report an untested penicillin allergy or “label” do not actually turn out to be allergic when tested. We also know that penicillin allergy labels cause disruptions in optimal health care delivery because a penicillin allergy reported by a patient will result in the health care provider feeling pressured to prescribe something that is dissimilar to a penicillin, which often leads to an inferior antibiotic being chosen.
Research has shown that choosing an alternative antibiotic can then cause inadvertent increases in multidrug resistant infections, surgical site infections and other bad outcomes, because many of the alternatives are less effective.
In that context, with this new paper looking at an enormous database of pneumonia treatments that were given, we now see that reporting a penicillin allergy increases a patient’s risk for hospitalization, intensive care and death from pneumonia and that this risk is related to getting antibiotics that just do not work as well.
The best outcomes from treating any bacterial infection like pneumonia happen when patients get the right treatment at the right time, which is the hallmark of good antimicrobial stewardship.
So, my main comment on this paper is that it really reinforces how penicillin allergy labels are causing big problems for individual patients, health care providers, hospitals and society when we are up against bacterial pneumonia.
Patients want to have treatments that work when they get sick, but their untested penicillin allergy label frequently results in the wrong treatment at the wrong time. Health care providers want to provide good care for their patient’s infection, but an untested penicillin allergy is like making them do so while wearing a pair of handcuffs.
Hospitals are heavily scrutinized on their pneumonia outcomes, and penicillin allergies are clearly contributing to some of the failed treatments that they will get criticized about. Also, insurers are having to pay extra money for more expensive antibiotics that do not even work as well! Finally, our health care systems are not getting good outcomes for these patients who have untested penicillin allergies compared with their counterparts.
Any time you get bad outcomes, you have increased sorrows and costs for everyone, from patients to payers.
The only real solution to this particular problem of inferior treatment outcomes in those who report penicillin allergies is for us to really scale up our efforts to remove unnecessary penicillin allergy labels through testing.
It is going to need to happen at a societal level, and it is going to need to be integrated into every health care system to some degree due to the scale of the issue, as 8% to 15% of most populations report a penicillin allergy label.
The good news is that currently validated strategies can already identify around 60% to 70% of patients with an untested penicillin allergy who are low risk enough that they can safely disprove their penicillin allergy by just taking a dose under medical observation, which is called a direct challenge.
Studies are popping up all over the world showing that a risk-stratified approach to low-risk allergies with direct challenges can be safely and readily combined with more intensive consultation and allergy skin testing prior to challenges for patients who report concerning or severe reactions. This is called risk-stratified management of penicillin allergy labels.
For the future, I also think that a second key is to prevent unnecessary penicillin allergies from being added to a patient’s chart to begin with. Low-severity, self-limited rashes, isolated gastrointestinal upset, family history of a penicillin allergy and other factors do not warrant an allergy label, and we health care providers may be setting our patients up for a disaster down the line when we put them into a patient’s chart and set them on the path of unnecessary avoidance.
As I have sometimes remarked before, a penicillin allergy is frequently a seed planted in childhood that then grows into a large tree that dies while overhanging your house. When a storm (or infection) comes later, that tree then falls on your house and causes additional unnecessary damage that you could have avoided by removing it beforehand.
Collapse
Read more about
Click Here to Manage Email Alerts