Hypokalemia Measurement and Management in Patients With Status Asthmaticus on Continuous Albuterol

This study was a retrospective chart review approved by the institutional review board at our university. We included patients admitted for status asthmaticus, with a PAS ≥12, and started on continuously inhaled albuterol in our PICU or intermediate care unit (IMU) between November 2017 and April 2019 after implementation of our standardized asthma pathway. We included patients 2 to 18 years of age who arrived at the PICU or IMU from the emergency department or outlying hospitals. Patients <2 years of age are excluded from our asthma pathway and, therefore, were not included in this study. Per our institutional asthma pathway, all patients on continuous albuterol received maintenance IVF containing 40 mEq/L potassium chloride.

To minimize confounding variables, we excluded intubated patients, patients on continuous terbutaline or magnesium infusions, and patients who received repeat continuous albuterol treatment separated by >4 hours off treatment. Of note, our PICU does not admit patients with unrepaired primary congenital heart lesions.

Patients were identified through a query of our institution’s Epic electronic health record. Patient demographics, duration of continuous albuterol administration, and albuterol dose were recorded. The primary outcome measure was the treatment of hypokalemia in relation to the number of laboratory draws for potassium levels. We defined treatment of hypokalemia as oral or IV replacement of potassium or increasing the potassium concentration in IVFs. The secondary outcome measure was the hypokalemia frequency and relation to the duration (hours) and initial dose (mg/kg per hour) of continuous albuterol. We considered all potassium values when determining the number of laboratory draws, including single and bundled (eg, basic metabolic panel) laboratory orders. We included laboratory draws for potassium levels while the patient was on continuous albuterol therapy. Potassium levels obtained from i-STAT samples were not included. When referencing hypokalemia, a patient’s lowest potassium value was used. Other variables examined were patient demographics, initial IVF concentration of potassium, and any changes in IVF potassium concentration. Hypokalemia was defined as a serum potassium level <3.5 mmol/L, the lower limit of normal for our hospital laboratory. Study data were collected and managed using REDCap electronic data capture tools hosted at our institution.^16,17 

Continuous variables were summarized as mean and SD or median and quartiles. Normality was checked using the Anderson-Darling test. The t test was performed where variables were normal or transformed to normality. Wilcoxon rank sum test was used when normality was impossible. Categorical variables were summarized by frequency and percent and analyzed using χ² or Fisher’s exact test. All statistical analyses and data visualizations were conducted using SAS version 9.4 software (SAS Institute Inc, Cary, NC). All tests were 2-sided, assuming P ≤ .05 indicated statistical significance.

A total of 156 patients (83 males, 53%) with status asthmaticus met the inclusion criteria during the study period. One hundred nine patients were admitted to the IMU (69.9%), and 47 were admitted to the PICU. Four patients were initially admitted to the IMU then transferred to the PICU during their stay. Our cohort had a median age of 7 years (interquartile range [IQR], 4–10) and weight of 24.2 kg (IQR, 17.7–38.6). Other demographic variables compared with frequency of hypokalemia are described in (Table 1).

These patients had 420 laboratory draws for serum potassium levels, averaging 2.7 per patient. One patient had 21 separate laboratory draws while on continuous albuterol for 125 hours. A lowest recorded potassium value of <3.5 mmol/L was found in 87 (55.8%) of the 156 patients. Twenty-two (14.1%) patients had a value of <3 mmol/L. The lowest potassium level recorded in our cohort was 2.3 mmol/L, with a median value of 3.4 mmol/L (IQR, 3.2–3.7).

Of the 156 patients, 46 had a highest potassium value of >4.7 mmol/L, the defined reference range laboratory value for hyperkalemia. This group had potassium levels checked an average of 4.33 times (IQR, 3–5) per patient. Comparatively, patients with a highest potassium value of ≤4.7 mmol/L had potassium levels checked, on average, 1.97 times (IQR, 1–2) per patient. In patients with hyperkalemia, only 6 had their IVF potassium concentration decreased from 40 to 20 mEq/L.

Thirteen of the 156 patients received treatment of hypokalemia. Most patients (117, 75%) were initially started on IVF containing 40 mEq/L potassium chloride. Patients started on IVF containing 20 mEq/L potassium had an average lowest potassium value of 3.67 mmol/L (IQR, 3.3–3.9) compared with 3.43 mmol/L (IQR, 3.1–3.7) in patients started on IVF containing 40 mEq/L of potassium. Patients started on IVF containing 20 mEq/L of potassium were more likely to receive treatment than those started on IVF containing 40 mEq/L (P = .0001) (Table 2). Most patients with hypokalemia (74 of 87, 85%) did not receive treatment for hypokalemia.

No correlation was found between initial weight-based albuterol dosage and the lowest potassium value recorded (P = .52) (Fig 1). There was also no association with initial albuterol dosage (not weight adjusted) and the lowest potassium value recorded (Table 3). The majority of patients in our cohort received initial doses between 10 and 20 mg per h (Table 3).

Our cohort had a median albuterol duration of 8.25 hours (IQR, 3.9–13.1). Albuterol duration did not significantly affect hypokalemia. Patients with hypokalemia had a mean albuterol time of 12.32 (SD, 15.76) hours, with a median time of 8.75 hours (IQR, 4.2–14.1), whereas patients without hypokalemia had a mean albuterol time of 11.50 (SD, 12.53) hours (P = .29) with a median time of 7.25 hours (IQR, 3.6–12). Median albuterol duration was not associated with need for treatment of hypokalemia (Fig 2).

Our results reveal that hypokalemia, defined as serum potassium <3.5 mmol/L, was present in approximately half of the patients who received continuously nebulized albuterol for status asthmaticus in our PICU or IMU. Our cohort had 420 laboratory draws for serum potassium levels, with few patients receiving treatment of hypokalemia. Hypokalemia was not correlated with duration or initial dose of continuous albuterol.

Although a little more than half of our cohort (51.5%) had a low potassium value of <3.5 mmol/L, very few patients received potassium replacement or increased potassium content in their IVFs. Even fewer patients had a lowest serum potassium level of <3.0 mmol/L. We recognize that our definition of hypokalemia (<3.5 mmol/L) is likely conservative. In otherwise healthy children without cardiac defects, at what level of hypokalemia would lead to arrhythmias is unknown, but that level is likely much <3.5 mmol/L. In adults, hypokalemia is generally considered severe if <2.5 mmol/L or if the patient is symptomatic.¹⁹  In our cohort, there was no guideline for hypokalemia treatment. Some patients had potassium levels lower than those whose levels were treated, suggesting that we can set treatment thresholds lower than previously thought. However, all our patients received IVFs containing either 20 or 40 mEq/L potassium while on continuous albuterol, which may have blunted the amount and severity of hypokalemia in our cohort. Although not strictly enforced, most patients remain on full maintenance IVFs while on continuous albuterol. We typically reduced IVFs once patients are taking by mouth. The amount of potassium-containing IVF being received is an important variable to consider when managing patients on continuous albuterol. Additional studies would be required to determine appropriate treatment guidelines for albuterol-associated hypokalemia.

Also important for clinicians to remember is to monitor for hyperkalemia if using higher concentrations of potassium in IVFs, especially as the dose or duration of albuterol decreases. We had 46 patients with hyperkalemia with a highest potassium value of >4.7 mmol/L. Only 6 of these patients had their IVF potassium concentration decreased from 40 to 20 mEq/L. These patients also had potassium levels checked, on average, twice as much as those without hyperkalemia, which shows that repeat laboratory draws were performed without other intervention, likely because the clinical team believed that these values were falsely elevated.

Albuterol, a β-adrenergic agonist, binds to β2-adrenergic receptors in the bronchiole smooth muscles, causing an intracellular shift in potassium via the Na,K-ATPase on skeletal muscle and leading to smooth muscle relaxation.⁵  There is not a total body reduction of potassium.⁶  Continuously nebulized albuterol can be safely used for long periods and at doses up to 20 to 25 mg/h.⁵  Reports exist of safely using doses up to 150 mg/h for refractory status asthmaticus.¹⁹  Our institution’s asthma pathway recommends continuous albuterol dosing of 5 to 30 mg/h on the basis of patient weight (unpublished data). In a study examining the safety of continuous nebulized albuterol in a non-ICU setting, Kenyon et al²⁰  reported that 12% of their cohort had hypokalemia with values <3.0 mmol/L, but the finding was not statistically different from patients receiving intermittent albuterol. The authors also concluded that because of the low levels of clinical deterioration, continuous albuterol is safe for use in children outside the ICU, which is a practice also adopted at our institution.

We measured serum potassium levels an average of 2.7 times per patient, with 1 patient receiving 21 laboratory draws. Pain with pediatric venous access is well studied in the literature. Especially for children, needlesticks associated with venipuncture can be traumatic and painful and can provoke anxiety.²¹  This pain may be short term but has been proven to have lasting effects, such as adulthood avoidance of medical care. Both parents and children report that needles are a large source of anxiety surrounding medical procedures.^21–23  Even healthy children receive numerous needlesticks during vaccinations, and this pain is not insignificant. This pain often causes children to fear health care settings, and this fear begins to develop around age 5 years, peaking at age 10.²⁴  Most of our study’s cohort fell within this age range, underscoring the risk of adverse effects, such as pain and anxiety, with increased potassium laboratory draws leading to minimal clinical benefit.²⁵  Our institution currently attempts to obtain laboratory draws through peripheral IV lines, if able, but if unsuccessful, the sample is obtained by venipuncture. Our results reveal that it may be possible to safely reduce the number of potassium laboratory draws in patients on continuous albuterol and potassium-containing IVFs, leading to fewer painful procedures a patient would experience while hospitalized. Other benefits to decreased laboratory testing include reduced staff burden needed to obtain samples and perform testing and decreased risk of iatrogenic anemia.^26,27 

Clinical pathways for the treatment of status asthmaticus have been shown to have a positive impact on patient outcomes.^1,4,14,28,29  However, these clinical pathways are not consistent with regard to the potassium content of IVFs and the number or frequency of laboratory draws necessary for patients on continuous albuterol. Most of these pathways do not address potassium laboratory draws or hypokalemia. Our research helps to clarify this aspect of the status asthmaticus clinical pathway at our institution and can be used to assist with quality improvement of asthma clinical pathways.

Because of the retrospective nature of this study, there are several limitations. First, adherence to our institution’s asthma protocol with regard to laboratory draws depends on the attending physician or clinician. Thus, not all patients had a similar number of laboratory draws based on length of continuous nebulized albuterol received. Second, our institution allows for the procurement of laboratory draws from existing peripheral IV lines, which can potentially increase the risk of hemolysis. We did not examine rates of potassium levels obtained secondary to falsely elevated readings from hemolysis. An additional limitation is that in our practice, the decision to treat hypokalemia depends on the clinician. We noticed that there was no standardized potassium level deemed low enough to require treatment and no standardized route or dosage of treatment. In future studies, a standardized hypokalemia treatment plan with dosages and routes of administration would lead to more clarity regarding the need for treatment of hypokalemia in status asthmaticus. One of our defined treatments for hypokalemia involved increasing the IVF content of potassium, which is relatively simple to do and which most would consider a benign intervention in otherwise healthy patients. Considering an increase in IVF potassium content a “treatment” skewed the results for patients started on IVF with lower concentrations of potassium. However, not all patients who started on lower IVF potassium concentration received this intervention. Finally, another limitation was the exclusion of patients with congenital heart disease. These patients are at increased risk of cardiac arrhythmias even from mild hypokalemia, so these results may not be applicable to this specific patient population.

Our patient cohort on continuous albuterol had a high number of blood draws for potassium levels and low levels of treatment for hypokalemia, despite 56.8% having a documented potassium level of <3.5 mmol/L. However, hypokalemia cannot be completely excluded in patients on continuous albuterol, and clinicians should be aware of this potential adverse effect. All our patients received IVFs containing potassium while on continuous albuterol, which may have blunted the amount and severity of hypokalemia seen in our cohort. The length of treatment with continuous albuterol and initial dosage of albuterol were not found to be statistically significant factors that lead to hypokalemia. Serial laboratory measurements may be decreased to reduce health care costs, with an anticipated reduction in pain and anxiety surrounding needlesticks.