Comparison of Adverse Effect Profiles of Nafcillin and Oxacillin in Pediatric Patients

Introduction

Nafcillin and oxacillin are beta-lactam antibiotics widely used as first-line therapy for methicillin-susceptible Staphylococcus aureus infections. Common side effects reported in adults with these anti-staphylococcal penicillins include nephrotoxicity, hepatotoxicity, hypersensitivity reactions, and thrombophlebitis.¹ There are scarce data available in the literature comparing adverse effect profiles of nafcillin and oxacillin, and most publications reference adult populations. A 2016 study of adult patients led to our hospital system’s formulary change from nafcillin to oxacillin for the anti-staphylococcal penicillin of choice.¹ One pediatric study, assessing patients receiving nafcillin or oxacillin, reported adverse effects in 18% to 25% of the population.² Owing to the limited pediatric evidence available and the formulary change from nafcillin to oxacillin, this study was conducted to determine the frequency of adverse effects of nafcillin compared with oxacillin in pediatric patients.

Materials and Methods

This was a retrospective, single center study conducted at a children’s hospital located within an academic medical center. On August 1, 2018, the hospital system made a formulary change from nafcillin as the preferred anti-staphylococcal penicillin antibiotic to oxacillin. The study period was selected as August 1, 2016, to August 31, 2020, to incorporate 2 years of data for both the nafcillin and oxacillin groups. A data report was pulled from our electronic medical record containing all nafcillin and oxacillin administrations during this period. Patients were included if they were admitted to the children’s hospital and received either nafcillin or oxacillin for at least 48 hours during the study period. Patients were excluded if they were 18 years of age or older. Patients were included multiple times if they received a second course of therapy in a separate admission. All data were collected from the electronic medical record at initiation and throughout the duration of therapy. Baseline characteristics including age, sex, and race were collected for each patient. Additionally, nafcillin or oxacillin daily dose and duration of treatment, concurrent nephrotoxins (e.g., ibuprofen, ketorolac, chlorothiazide, furosemide, gentamicin, vancomycin, piperacillin-tazobactam), concurrent hepatotoxins (i.e., acetaminophen, rifampin, fluconazole), and all documented comorbidities were collected. All doses were dispensed in either a 20-mg/mL syringe or bag for doses >1000 mg. Drug concentrations in the bag varied by dose, but all were <40 mg/mL.

The primary outcome of the study was to compare the cumulative sum of adverse effects of nafcillin to oxacillin. The cumulative sum of adverse events was calculated as the total number of events occurring during therapy across all patient encounters. Secondary outcomes included the incidence of hypokalemia (defined as serum potassium <3.5 mEq/L), acute kidney injury (AKI; defined as 50% increase from baseline serum creatinine), hepatotoxicity (defined as liver transaminase values greater than 5 times the upper limit of normal), neutropenia (defined as absolute neutrophil count [ANC] <1500 cells/µL), infusion-related reactions, intravenous (IV) line replacements, and early discontinuation of therapy. Events occurring while the patient was receiving nafcillin or oxacillin therapy were collected. Infusion-related reactions were tracked through comments documented on the medication administration record by either the nurse or provider. Intravenous line replacements were collected via notation in daily progress notes from providers if replacement was due to thrombophlebitis or extravasation. Notably, line replacements were not included if the patient went from a peripheral IV line to a peripherally inserted central catheter for a planned extended duration of therapy. Early discontinuation of therapy must have been documented in the patient’s chart, explicitly stating that the antibiotic regimen was changed owing to an intolerance of nafcillin or oxacillin.

Baseline characteristics were compared via Fisher exact test, and drug therapy details were compared via Mann-Whitney U test. The primary outcome was analyzed through chi-square test to identify a significant difference between cumulative adverse effects among the groups. Secondary outcomes were analyzed for differences with chi-square or Fisher exact test, as appropriate for data type. To compare baseline and post-therapy laboratory values, a sub-analysis using paired Student t tests was conducted for serum potassium, serum creatinine, liver transaminases, and ANC in patients who had both values available. All statistics were performed by using VassarStats Website for Statistical Computation.

Results

From August 1, 2016, to August 31, 2020, there were 122 patient encounters identified as receiving either nafcillin or oxacillin during admission. Sixty-nine patient encounters were excluded from the study, most having received therapy for less than 48 hours (n = 60); the remaining were excluded for being 18 years or older. A total of 53 patient encounters were included, 32% in the nafcillin group (n = 17; 15 individual patients) and 68% of the population in the oxacillin group (n = 36; 31 individual patients). While not explicitly excluded, no included patients were pregnant or had active malignancy.

Baseline characteristics did not significantly differ between the 2 groups (Table 1). Patients were predominantly male (58%) and 49% of the study population was African American. The median age of participants was 2 years, and 25% of the population consisted of neonates (younger than 4 weeks). The median gestational age of included neonates was 25 weeks. Between the nafcillin and oxacillin groups, there was a significant difference in the median duration of therapy (4 days vs 11 days for nafcillin and oxacillin, respectively; p = 0.013) (Table 2). Total daily doses of nafcillin and oxacillin were consistent with traditional dosing of each medication. Four patient encounters (11%) in the oxacillin group received the drug as a continuous infusion. Seventy-five percent of patient encounters (n = 40) were on concurrent nephrotoxins, with half of these receiving more than 1 agent. The most common nephrotoxins were vancomycin (n = 11), ibuprofen (n = 10), chlorothiazide (n = 9), furosemide (n = 8), and gentamicin (n = 8). The median number of concurrent nephrotoxins per patient encounter was 1 for both groups. Fifty-five percent of patient encounters (n = 29) were on concurrent hepatotoxins, with the most common being acetaminophen (n = 28). Most patient encounters had comorbidities present (n = 42; 79%), with 33 of these patients having multiple comorbidities present (79%). Comorbidities included respiratory disorders (bronchopulmonary dysplasia, respiratory distress syndrome, chronic lung disease; n = 15; 28%), gastrointestinal disorders (short gut syndrome, G-tube dependence, failure to thrive; n = 15; 28%), congenital heart defects (n = 10; 19%), neurologic disorders (seizures, cerebral palsy, intraventricular hemorrhage, spinal cord lipoma; n = 7; 13%), and miscellaneous disorders (diabetes mellitus, chronic osteomyelitis, asthma; n = 5; 9%).

Table 1. Baseline Patient Characteristics

View Fullscreen

Table 2. Details of Nafcillin and Oxacillin Therapy and Patient Conditions

View Fullscreen

Data detailing all outcomes are presented in Table 3. Overall, 30 patient encounters (56.6%) were noted to have an adverse effect. The primary outcome, cumulative sum of adverse effects, was found to be 16 total events occurring in 9 patients from the nafcillin group and 45 total events occurring in 21 patients in the oxacillin group (p = 1). When assessing secondary outcomes, baseline and post-therapy laboratory values were compared, when available; because of this, smaller sample sizes were assessed for each outcome. There were no significant differences in hypokalemia (p = 0.46), AKI (p = 0.72), hepatotoxicity (p = 1), or need for IV line replacements (p = 0.12) between the nafcillin and oxacillin groups. One patient in the oxacillin group had baseline liver dysfunction that worsened during therapy. Notably, all but 1 patient who experienced AKI were receiving other nephrotoxin(s) during therapy. Of the 26 patients in the oxacillin group with white blood cell data available, 3 developed neutropenia (12%); no neutrophil comparison laboratory values were available for patients in the nafcillin group. Infusion-related reactions were noted in the nafcillin group (n = 2; 12%), but not the oxacillin group. There were no instances of early discontinuation of therapy due to drug intolerance documented in patient charts for either drug.

Table 3. Primary and Secondary Outcomes

View Fullscreen

A sub-analysis was conducted comparing pre- and post-laboratory values to examine the overall change for continuous variables in each group (Table 4). There was a significant decrease in serum potassium for both nafcillin (p = 0.004) and oxacillin (p < 0.0001) groups. Additionally, there was a significant increase in serum creatinine from baseline for nafcillin (p = 0.03) and oxacillin (p = 0.07) groups. Changes in liver transaminases from baseline were not found to be significant for either group. There was a significant decrease in ANC for the oxacillin group (p = 0.002). No data were available for neutropenia in the nafcillin group.

Table 4. Sub-analysis of Drug-Associated Adverse Effects*

View Fullscreen

Discussion

This study found no difference in the cumulative sum of adverse effects between the nafcillin and oxacillin groups. For our secondary outcomes, we also were unable to detect any differences in the incidence of hypokalemia, AKI, hepatotoxicity, neutropenia, infusion-related reactions, and IV line replacements. We did not find any documentation of early discontinuation of either nafcillin or oxacillin due to adverse effects and intolerance of the drug.

When analyzing the degree of adverse effects within each group, we discovered a significant decrease in serum potassium concentrations in both nafcillin and oxacillin groups to median values of 3.2 and 3.4 mEq/L, respectively. These values represent true hypokalemia and can be considered clinically significant. A statistically significant increase in serum creatinine concentrations was identified in both treatment groups; however, this would not be considered clinically significant in practice. It is reassuring that neither drug led to increased values of liver transaminases (p = 0.22 and 0.12), which has been identified in previous studies of these drugs.^2,3 About 21% of patients receiving oxacillin with reductions in neutrophils developed neutropenia (ANC <1500 cells/µL), all of whom were on prolonged therapy (≥14 days).

To our knowledge, this is the first study comparing adverse effect profiles of nafcillin and oxacillin in the pediatric inpatient population, and we found no difference in the sum of cumulative adverse events between the 2 groups. A prospective review of methicillin, nafcillin, and oxacillin use in pediatric patients found that 18.7% and 25% of patients receiving nafcillin and oxacillin, respectively, experienced adverse effects, and recommended regular monitoring.² In 2002, Maraqa et al³ performed a retrospective review of adverse effects from outpatient use of nafcillin, oxacillin, clindamycin, and several other antibiotics in 222 pediatric patients. In contrast to our results, they found the highest incidence of adverse effects to occur in the oxacillin group, specifically with hepatotoxicity (22%) and rash (31.7%). In 2016, Viehman et al¹ conducted a review in 224 adult patients. Patients receiving nafcillin were more likely to experience hypokalemia (51% vs 17%) and AKI (18% vs 6%) than patients receiving oxacillin. They did not find a significant difference in incidence of hepatotoxicity between the 2 groups, similar to our results. Importantly, patients receiving nafcillin were more likely to discontinue therapy early owing to adverse effects than patients receiving oxacillin, which differed from our study. Additionally, a review of adverse event reports submitted to the US Food and Drug Administration showed a higher incidence of nephrotoxicity (18.2% vs 5.1%) and hypokalemia (10.8% vs 0.7%) in patients receiving nafcillin as compared with oxacillin, differing from our results.⁴

Our results found that pediatric patients receiving either nafcillin or oxacillin did experience adverse effects associated with therapy; however, there was no difference in cumulative adverse effects, bringing new perspective to the literature. As such, we are unable to discern if one drug has a better safety profile compared with the other, and if one drug should be preferred for use in the pediatric population. At our institution, oxacillin remains the formulary agent of choice for patients of any age.

Our study is not without limitations. Owing to the retrospective nature of the study, we cannot directly attribute our findings to the use of nafcillin or oxacillin, because many other factors may have played a role in these effects. We accounted for concurrent nephrotoxins and hepatotoxins in the patient’s daily medication lists; however, our small sample size did not allow for completion of regression analyses to control for other confounders. Additionally, our definition of nephrotoxicity was not inclusive of all Kidney Disease: Improving Global Outcomes (KDIGO) criteria,⁵ as we did not include a serum creatinine increase of 0.3 mg/dL or more within a 48-hour period. For infusion-related reactions and IV line replacements, we relied on what was documented in the electronic medical record; this may have led to missed information in some patients. We had a small sample size from only 1 medical center with disproportionate representation among the 2 groups, making it possible that a difference does exist, but was not detected. The significant difference in median duration of therapy between the 2 groups may have also skewed our results. Finally, baseline data were not available for all patients, which may have caused us to exclude some instances of adverse effects in the populations. This also further reduced sample size, increasing risk for type II error.

Conclusions

This retrospective study did not detect a significant difference in adverse effects between nafcillin and oxacillin in pediatric patients. Both treatment groups experienced adverse effects, specifically AKI and hypokalemia. Larger, prospective studies are necessary to determine if there is a difference in safety profile between the 2 drugs in this patient population.