Intravenous Versus Oral Acetaminophen Use in Febrile Neurocritical Care Patients

Introduction

Acetaminophen (APAP) is frequently used in health care settings for pain and fever management (Young et al., 2015). Regarding pain management, APAP is recognized for its long-established efficacy, low risk of drug–drug interactions, and minimal side effect profile compared with that of nonsteroidal anti-inflammatory drugs (NSAIDs) and opioids. NSAID induced gastrointestinal and cardiovascular side effects, as well as their potential antiplatelet effect limits the use of NSAIDs in neurocritical care (NCC) patients.

Fever in neurologically injured patients is associated with poor outcomes, and treatment to normothermia is common practice (Azzimondi et al., 1995; Bohman and Levine, 2014). Fever can exacerbate blood–brain barrier disruption, increase free radicals, increase cerebral metabolic rate of oxygen, aggravate cerebral edema, and lead to an increased intracranial pressure, which can all cause deleterious effects and lead to worse outcomes in a neurologically injured patient (Meier and Lee, 2017).

Pharmacokinetic studies demonstrate earlier and higher peak levels of APAP in both plasma and cerebrospinal fluid for intravenous (IV) versus oral (PO) or rectal APAP, likely due to lack of first-pass metabolism and no absorption concerns (Picetti et al., 2014; El-Mashad et al., 2017). The absorption of PO APAP is affected by changes in gastrointestinal motility and may be reduced by fasting, anesthesia, and stress experienced by patients after surgical procedures (Petring et al., 1995; Singla et al., 2012; De Ridder et al., 2017). Despite these pharmacokinetic advantages of IV APAP over PO APAP, clinically significant differences in the antipyretic effect need to be assessed.

To the best of our knowledge, there are limited studies comparing PO APAP with IV APAP for fever control in NCC patients. Therefore, the objective of this study was to determine whether IV APAP is associated with a greater reduction in body temperature than PO APAP in febrile NCC patients.

Methods

This is a single-center retrospective study of NCC patients admitted to the Virginia Commonwealth University Health Neuroscience Intensive Care Unit from May 1, 2012, to April 30, 2013. Patients were included if they received at least one dose of IV APAP for an oral body temperature of ≥38°C and were between the ages of 18 and 88 years. APAP doses were included if the patient had a predose pain score of 0 and oral body temperature of ≥38°C before APAP administration, suggesting that the dose was used as an antipyretic and not as an analgesic. Patients were excluded if they were on another antipyretic medication or were pregnant or incarcerated. Data were obtained from the electronic medical record. The study protocol was approved by the Virginia Commonwealth University Institutional Review Board.

The primary aim of this study was to compare the reduction in body temperature (RIT) associated with IV versus PO APAP, calculated as the change in the oral body temperature before each APAP dose and after the dose at 0.5, 1, 2, 3, and 6 hours. Many patients received a combination of IV and PO products throughout their intensive care unit stay, so each dose was evaluated independently. Only doses that were administered 6 or more hours after the last dose of APAP were included in the evaluation. Doses of IV APAP were 1000 mg, and a majority of PO APAP doses were 650 mg, which are the standard of care dosages at our institution based on product availability.

Data collection included patient demographics, diagnosis codes, date and time of APAP dose, number of doses administered, frequency of APAP use, and body temperature before and after each APAP dose. Body temperatures were most frequently measured orally during the study period, followed by rectal and axillary measurements. Bladder and esophageal temperature probes were not commonly used. Body temperatures were standardized to oral temperature using conversion factors of 1° below oral measurements for axillary and 1° higher for rectal measurements (Barnett et al., 2011). Reduction in body temperature (RIT) was calculated as the change in the body temperature before and at 0.5, 1, 2, 3, and 6 hours after administration. If multiple temperatures were obtained, the last recorded temperature in each time period was used to calculate the RIT.

Subgroup analyses were performed based on NCC disease state to determine if differences existed in RIT based on injury type. Subgroups included acute ischemic stroke, epidural hematoma, intracerebral hemorrhage, neoplasm, subarachnoid hemorrhage, spinal cord injury, subdural hematoma, and traumatic brain injury.

Descriptive statistics were used for dose characteristics, and Kruskal–Wallis and Mann–Whitney U tests were used to analyze between-group differences. It was assumed that all data points (doses) were independent, including those that came from the same patient. A p-value of <0.05 was considered statistically significant. All statistical analyses were performed using SPSS (2013, version 22.0; IBM Corp., Armonk, NY).

Results

There were 142 NCC patients who received APAP during the study period, and a total of 658 doses evaluated independently. The mean age was 57 years, 59% were male, and 47% were Caucasian (Table 1). Four hundred five doses of IV APAP and 253 doses of PO APAP were included in the RIT comparison. All IV APAP doses were 1000 mg, 96% of PO APAP doses were 650 mg, and 4% of PO APAP doses were 325 mg. IV and PO APAP predose temperatures were 38.8°C and 38.6°C, respectively (p < 0.01).

The percentage of doses achieving a temperature of <38°C within 6 hours of administration was similar for IV and PO APAP at 70.4% and 67.6% (p = 0.5), respectively (Table 2). The average time to oral body temperature <38°C was 1.7 hours for IV and 2 hours for PO (p = 0.22). The RIT was significantly greater among those receiving IV APAP compared with those receiving PO APAP at 0.5 (0.25°C vs. 0.2°C), 1 (0.4°C vs. 0.2°C), 2 (0.7°C vs. 0.5°C), 3 (0.9°C vs. 0.6°C), and 6 (1°C vs. 0.8°C) hours after administration (p < 0.05 for all comparisons) (Table 3).

The data were further analyzed by neurological disease state subgroup. In patients with an ischemic stroke, IV APAP was associated with a significantly greater RIT than PO APAP at 0.5 (0.5°C vs. 0.1°C, p = 0.03) and 1 (0.4°C vs. 0.2°C, p = 0.02) hour after administration. Similarly, in patients with an intracerebral hemorrhage, IV APAP was associated with a significantly greater RIT than PO APAP at 2 (0.7°C vs. 0.5°C, p = 0.04) and 3 (0.9°C vs. 0.6°C, p < 0.01) hours after administration. There were no significant differences found in any of the other subgroups.

Discussion

In our study, NCC patients treated with IV APAP had a significantly greater reduction in oral or equivalent body temperature and a trend toward achieving faster fever control than the PO dosage form despite higher predose body temperatures in the IV group. The RIT achieved with IV APAP varies in the literature from 0.14°C to more than 1°C, and our result lies within this range (0.4°C [interquartile range 0.1–0.8]) (Kallmünzer et al., 2011; Kett et al., 2011; Tsaganos et al., 2017; Roy and Simalti, 2018). This large variability may be due to small sample sizes in most of the trials, patient population, or baseline temperature. Due to this variability, we believe that these results are impactful, especially in the NCC population. The rapid onset of effect of IV APAP along with its sustained effect over 6 hours may be particularly beneficial in NCC patients where fever control is associated with better clinical, functional, and economic outcomes (Greer et al., 2008). Therefore, even though the observed median differences in RIT were small in our study, it may be associated with significantly better clinical outcomes.

The results in our study are consistent with published pharmacokinetic differences between the two routes of administration, for which the time to maximum plasma concentration for IV APAP is ∼15 minutes compared with ∼45 minutes for PO APAP in patients with normal gastrointestinal motility. IV APAP also leads to earlier peak cerebrospinal fluid levels, greater drug exposure, and overall faster onset of efficacy compared with the PO and rectal dosage forms (Mallinckrodt, 2017). Among healthy individuals given an endotoxin, IV APAP demonstrated a faster onset and greater fever reduction than PO APAP or placebo, with the onset of fever reduction occurring 15 minutes after the end of the 15-minute infusion (Frank Peacock et al., 2011).

Based on injury type, there were no significant differences found between patients with traumatic brain injury, subdural hematoma, spinal cord injury, subarachnoid hemorrhage, neoplasm, or epidural hematoma, possibly due to the small sample sizes in these subgroups. However, patients with ischemic stroke and intracerebral hemorrhage did show significant RIT differences between IV and PO up to 3 hours postdose.

Despite potentially better efficacy, the safety of IV APAP should be considered. IV APAP is associated with hypotension, although mixed results exist in the literature (Picetti et al., 2014; Saxena et al., 2015; De Ridder et al., 2017; Schell-Chaple et al., 2017). There are currently limited safety data for IV APAP in NCC patients, but the IV formulation contains 3.8 g of mannitol per 1000 mg dose, which may harmfully reduce blood pressure via osmotic diuresis. This quantity of mannitol is much less than the 1 g/kg body weight of mannitol given for intracranial hypertension, but IV paracetamol has been associated with a reduction in mean arterial pressure in acute brain injury patients (Picetti et al., 2014). Despite this, a different trial of 41 patients with traumatic brain injury found no difference in mean arterial pressure in patients administered IV paracetamol 1 g every 6 hours compared with a placebo group (Saxena et al., 2015). Another study of high-dose paracetamol in acute stroke patients with a body temperature of ≥36.5°C concluded that high-dose paracetamol was safe (De Ridder et al., 2017). Yet another study found that IV APAP significantly reduced core temperature, blood pressure, and heart rate up to 2 hours postdose, which may be clinically significant in certain patients. The authors suggest that in patients where normal or high blood pressure is essential for optimal outcomes, frequent monitoring after administration of IV APAP must occur (Schell-Chaple et al., 2017). Although our study did not evaluate blood pressure or heart rate, the impact of IV APAP on these parameters must be considered in NCC patients where adequate cerebral perfusion pressure is critical.

With increasing interest in therapeutic hypothermia for various disease states, establishing the efficacy and safety of temperature reducing agents is important. In addition to these safety concerns, IV APAP is significantly more expensive than PO APAP ($57 per 1 g vial of IV APAP vs. $0.02–$0.36 for a 650 mg PO dose [United States actual wholesale price]) (Lexi-comp, 2017).

Limitations of our study include its retrospective design, specifically accuracy and completeness of medical record, which resulted in varied sample sizes at the time points analyzed. Also, this is a single-center study with data from 2013. The dose of IV APAP was greater than that of PO APAP, which was a variable that was not able to be standardized because it reflects real-world practice. Another limitation is that, besides pharmacological management with other antipyretics, data regarding other mechanisms to manage temperature were not collected (e.g., cooling blankets). No clinical outcomes were assessed in this trial. Finally, for statistical analysis, we assumed that all data points (doses) were independent, even though many came from the same patient.

Conclusions

IV APAP was associated with a greater RIT than PO APAP at 0.5, 1, 2, 3, and 6 hours after administration in NCC patients. There was no significant difference found in the overall ability to control fever to a goal of <38°C within 6 hours. Based on the results of this study, IV APAP may be useful when rapid fever reduction is needed, when enteral administration is not tolerated, and when the risk of reducing blood pressure is low. Further prospective studies are needed to evaluate if rapid fever control leads to improved clinical outcomes in NCC patients.