Effects of ibuprofen on sleep quality as measured using polysomnography and subjective measures in healthy adults

Effects of Ibuprofen on Sleep Quality as Measured Using Polysomnography and Subjective Measures in Healthy Adults Francis G e n g o , P h a r m D

DENT Neurological Institute at Millard Fillmore Hospital, Buffalo, New York*; and Department of Pharmacy Practice, School of Pharmacy and Pharmaceutical Sciences, State Universi{y of New York, Buffalo, New York ABSTRACT

Background: Although some literature has suggested that NSAIDs may affect sleep physiology, this observation is not consistent with clinical use of these drugs and has not been verified using standard sleepresearch methodologies. Objective: This study was undertaken to determine whether ibuprofen 400 mg administered at 3, 7, and 11 PM (total daily dose, 1200 mg) produced any significant alterations in the character and quality of nighttime sleep as measured by standard sleep laboratory polysomnography (PSG) and subjective measures. Methods: This 4-day, multiple-dose, double-blind, randomized, placebo-controlled trial was conducted in a hospital-based, sleep laboratory in the United States (DENT Neurological Institute at Millard Fillmore Hospital, Buffalo, New York). Healthy subjects aged >18 years spent 3 consecutive nights in a sleep laboratory. Day 1/night 1 was for acclimation; day 2/ night 2, for baseline PSG and subjective sleep assessments; and day 3/night 3, for treatment effects on sleep character and quality. All subjects received placebo on days 1 and 2. On day 3, subjects received ibuprofen 400 mg or placebo TID. Results: All 30 subjects (15 per group) completed the study (18 men, 12 women; all white). The mean age (SD) was 28.6 years and mean body weight was 71.4 kg. In both groups, mean values for sleep efficiency and quality of sleep were significantly higher on night 3 compared with baseline; the mean (SD) changes from baseline were not significantly different between the ibuprofen and control groups (sleep efficiency, 0.4 [6.3] and 0.3 [6.2]; quality of sleep, 8.6 [26.8] and 3.3 [21.3]). Mean night-3 sleep efficiency in the ibuprofen group was 88.6%--substantially higher than the minimally acceptable sleep efficiency of 75% stated in the protocol. Three mild adverse events were reported in 2 subjects.

Conclusion: This study found that in these subjects a total daily dose of 1200 mg ibuprofen did not produce any clinically or statistically significant alterations in the character and quality of nighttime sleep as measured using standard sleep laboratory PSG and subjective measures. (Clin Ther. 2006;28:1820-1826) Copyright © 2006 Excerpta Medica, Inc. Key words: ibuprofen, polysomnography, sleep quality.

INTRODUCTION Over-the-counter (OTC) NSAIDs such as ibuprofen are frequently used for the temporary relief of minor aches and pains associated with the common cold, headache, toothache, muscular aches, backache, menstrual cramps, arthritis, and fever reduction. The usual nonprescription dose of ibuprofen for mild to moderate pain in adults is 200 to 400 mg every 4 to 6 hours, to a maximum total daily dose of 1200 rag. 1 With such a broad indication for common maladies, it is not surprising that OTC NSAIDs are among the most frequently used OTC medications. One potential action of NSAIDs that has not been documented fully is their effect on sleep quality. 2,3 This effect is of interest because prostaglandins (PGs),

Data were published in abstract form in Gengo F, Siegal B, Reitberg DP. The effects ofibuprofen on polysomnographic and subjective measures of sleep in healthy adults.J Clin PbarmacoL

1996;36:659. *Currently, Amherst, New York.

AcceptedforpublicationSeptember29, 2006. ExpressTrack online publication December1, 2006. doi:l 0.1016/j.clinthera.2006.11.018 0149-2918/06/$19.00 Printed in the USA.Reproductionin wholeor part is not permitted. Copyright© 2006 ExcerptaMedica, Inc.

.....

....

which are inhibited by NSAIDs, play a role in sleep regulation. PGD 2 is thought to induce sleep, whereas PGE 2 has been reported to induce wakefulness.4 A study by Murphy et al 5 reported that ibuprofen and acetylsalicylic acid (ASA) adversely affected polysomnographic (PSG) measurements while providing no significant confirmatory subjective findings. Methodologic issues may have contributed to the inconsistency between the objective and subjective findings. PSG is a multidimensional recording of biophysiologic changes that occur during sleep. Brain activity (measured using electroencephalography [EEG]), eye movement (electro-oculogram [EOG]), and heart rhythm (electrocardiography [ECG]) are most commonly monitored. PSG is used for evaluating sleep abnormalities as well as other disorders related to sleep and/or wakefulness. The objective of this study was to determine whether 400 mg of ibuprofen taken at 3, 7, and 11 I'm (total, 1200 mg/d) altered the character and quality of nighttime sleep as measured by standard sleep laboratory PSG and subjective measures. To further explore the clinical relevance of the PSG findings from Murphy et al, this study incorporated some refinements to the design used by Murphy et al, related to the acclimation period, number of PSG recordings, and dosing schedule.

SUBJECTS AND METHODS The study protocol was reviewed and approved by the institutional review board at Millard Fillmore Hospital, Buffalo, New York. Healthy volunteer subjects were recruited using advertisement. Written informed consent was obtained from all subjects before their study participation. Patients received a small financial stipend for participating. Inclusion and Exclusion Criteria Male and female individuals were eligible for the study provided they met all of the following criteria: (1) age 18 to 45 years; (2) use of a reliable method of birth control or surgical sterility (women); (3) regular monthly menstrual cycle, as confirmed by medical history (women); (4) good health, as confirmed by medical history, physical examination including vital-sign measurement, and routine clinical chemistry; (5) weight of >60 kg (men) or ->50 kg (women) and within 15% of ideal weight range listed in the Metropolitan Height and Weight Tables6; (6) agreement not to take any medication throughout the study; (7) agreement not :~ N o v e m b e r

2006 • ....

1

i

i •

:

F. G e ngo

to consume alcoholic beverages or xanthine-containing foods/beverages beginning 24 hours prior to the first dose of study medication through the end of measurements; and (8) pregnancy or breastfeeding in women. Individuals were not eligible to participate in the study if any of the following were noted: (1) abnormal sleep history, including problems with sleep initiation or sleep maintenance; (2) hypersensitivity or idiosyncratic reaction to ibuprofen, ASA, or any other NSAID; (3) regular use of any illicit drug; (4) consumed >3 cups of coffee per day (or equivalent); (5) received an experimental drug or device within 60 days preceding the study; (6) history of coagulation defect or anticoagulant therapy; (7) use of any medication within 3 days of admission; and (8) smoking habit.

Study Design This multiple-dose, double-blind, randomized, placebo-controlled, parallel-group trial was conducted in a hospital-based sleep laboratory in the United States (DENT Neurological Institute at Millard Fillmore Hospital, Buffalo, New York). All eligible subjects provided written informed consent prior to inclusion in the study. Standard medical assessments were performed within 14 days of initial dosing to ensure that all study subjects were in good health. A complete medical and sleep history was obtained and a complete physical examination was performed, including sitting heart rate and blood pressure measurements, a fasting venous blood sample for hematology and blood chemistry profiles, a urine specimen for urinalysis, and a 12-lead ECG. Each subject underwent a 10-day screening period of regimented sleep habits, which included a routine bedtime at 11:30 eM _+30 minutes and a standard wake time of 7:30 AM ± 60 minutes. Within a half-hour of bedtime, subjects were asked to void and/or have a bowel movement, if possible. Subjects maintained a sleep diary for recording bedtimes and wake times. Central nervous system-active medications, including antihistamines and ethanol, were prohibited during this 10-day screening period. Subjects also were to refrain from using any NSAID during this 10-day period. To reduce variability associated with the menstrual cycle in women, baseline PSG was performed during the early follicular phase, at 3 ± 2 days of the end of the menses. 7,8 For each subject, the evaluation was conducted over a 4-day period. Day 1/night 1 was an acclimation :•

• ~

:

i: iI

...... i 8 2 1

i

phase in the sleep laboratory to ensure that the subjects had normal sleep physiology. Day 2/night 2 was considered as the baseline day, followed by randomization on day 3/night 3. Subjects were released from the study on the morning of day 4. Subjects reported to the sleep laboratory at approximately 2 PM on days 1, 2, and 3. At 3 PM they were administered dose 1 of the assigned study medication. Subjects received a balanced, standardized, bland dinner and noncaffeinated beverage at 6 PM. At 7 PM, subjects received dose 2 of the assigned study medication. Subjects received a light snack at 11 PM and were administered dose 3 of the study medication. EEG electrodes and other monitoring equipment were attached to the subjects who were then escorted to a sound-attenuated bedroom at 12 AM and remained in bed until 8 AM. On day 1, electrodes were placed on subjects but the sleep monitors did not record any measures. On days 2 and 3, the following PSG sleep measures were recorded for 8 hours using the following: frontal, parietal, and occipital EEG; EOG; and electromyelography. Each sleep record was scored manually in 30-second epochs according to the methodology of Rechtschaffen and Kales. 9 The scorers were blinded to treatment assignments and data time points. Within 15 minutes of arising on days 2, 3, and 4, subjects were asked to complete the Stanford Sleep Scale 1° and 3 separate 100-mm visual analog scales (VASs) that assessed the level of alertness, general wellbeing, and quality of sleep. After the electrodes were removed on days 2 and 3, the subjects were discharged from the laboratory until 2 I'M and instructed to refrain from napping during the day. After removal of the electrodes on day 4, subjects were discharged from the study. All adverse reactions that occurred during the 4-day dosing and evaluation period were recorded.

Study Medications Subjects were randomly assigned according to a computer-generated randomization schedule to receive ibuprofen 400 mg (two 200-mg capsules) or identical placebo capsules TID (3, 7, and 11 PM) for 3 days. On days 1 and 2, all subjects received placebo on a singleblind basis. On day 3, subjects received ibuprofen or placebo on a double-blind basis.

Efficacy Assessments The 2 primary efficacy end points were sleep efficiency (the proportion of time spent asleep relative to

the time spent in bed) and the subject's evaluation of the quality of sleep, as measured on VAS. The secondary end points were number of awakenings; % of Stage Wake, Stage 1, Stage 2, Stage 3, Stage 4, and Stage REM; time of Latency to Stage 1, Stage 2, Stage 3, Stage 4, and Stage REM; number of stage shifts; and subjects' evaluation of alertness (Stanford Sleep Scale and VAS) and well-being (VAS). The hypothesis of this study was that ibuprofen would produce a statistically significant (P <_0.05) decrease in sleep efficiency compared with that observed following administration of placebo, with a value <75% efficient considered consistent with insomnia. Changes in sleep efficiency would be confirmed by a significantly altered response to the "Quality of Sleep" VAS (___30%poorer response relative to baseline score). The null hypothesis is that there would be no differences in sleep efficiency or there would be no complementary alteration in the Quality of Sleep VAS score.

Statistical Analysis Differences between means were considered statistically significant if tests indicated that the probability of a random occurrence of a difference was <0.05. Twenty-four subjects (12 per group) were needed to achieve 98% power to detect a significant difference between ibuprofen and placebo when the true difference in sleep efficiency is ->9% (5%) (mean [SD]). However, based on sample-size results in a study conducted by Murphy et al, 5 in which statistically significant discrimination for sleep efficiency was reported, 30 subjects (15 per group) were to be entered into the study to enhance the power of detecting significant differences for the subjective end points. Variables with continuous distributions were analyzed for background comparability using 1-way analysis of variance (ANOVA) with treatment as the main effect. Sex and race were analyzed using the Fisher exact test. All statistical tests were performed using SAS software (SAS Institute Inc., Cary, North Carolina). To calculate changes from baseline (A), night-2 evaluations were considered as baseline and were subtracted from all corresponding night-3 evaluations. All variables were analyzed using ANOVA with effects for treatment and sex. The treatment-by-sex interaction was also assessed for each of the variables by adding it to the main model. It was not retained in the main model because in general it was not significant.

:

.

.

.

.

:

:

:

i

. . . . .

.

.

.

.

.

.

.

:F. G -engo

.

The treatments were compared with respect to the incidence of adverse events using the Fisher exact test. All subjects were included in the tolerability analysis.

Table I. Baseline demographic and clinical characteristics of the study population.*

RES U LTS Study Population

Characteristic

Thirty subjects (15 in each group) participated in the study. All subjects received study medication, completed the study, and were evaluable. Within each treatment group, there were 9 men and 6 women. All subjects were white. The mean age, weight, and height of the sample were 28.6 years (range, 18-45 years), 71.4 kg (range, 50.8-93.4 kg), and 174 cm (range, 150-193 cm), respectively. The placebo and ibuprofen groups had comparable demographic and clinical characteristics (Table I).

Age, mean (SD), y Sex Male Female Weight, mean (SD), kg Height, mean (SD), cm

Baseline Evaluations The 2 groups were comparable with respect to both primary efficacy measures and all secondary efficacy measures except number of stage shifts (P = 0.038; Table II). % Awake and Latency to Stage 4 sleep were not significantly different between groups at baseline. These variables were numerically slightly different, so they were also analyzed using a model containing the corresponding baseline values in addition to treatment and sex.

Primary Efficacy Variables Both treatment groups had improved mean sleep efficiency and sleep quality during night 3 compared with baseline. The changes from baseline were similar between the 2 treatments (Table II). In the ibuprofen group, mean night-3 sleep efficiency was significantly higher than the minimally acceptable (as stated in the protocol) value of 75% (88.6%; P < 0.001 [1-sided t test]) and similar to that in the placebo group (87.3%). A significant treatment-by-sex interaction was present for sleep efficiency (P -- 0.039) due to women showing a numeric advantage in favor of ibuprofen, and men showing a numeric advantage in favor of placebo. When the treatment effect was adjusted for interaction, there was no significant difference between ibuprofen and placebo.

Secondary Efficacy Variables The treatment groups were comparable with respect to all parameters except % Stage REM, in which a significant treatment difference in favor of ibuprofen

I bu profen (n = 15)

Control (n = 15)

28.7 (8.0)

28.6 (7.3)

9 6

9 6

71.3(13.8)

70.4(12.1)

174.5 (13.5)

172.5 (10.2)

*No significant between-group differences were found.

was found (P = 0.02) (Table II). Repeat analysis was performed on the 3 variables that showed baseline incompatibility--number of stage shifts, % Awake, and Latency to Stage 4 Sleep. The treatment differences adjusted for baseline remained nonsignificant.

Adverse Events Two subjects in the ibuprofen group reported 3 adverse events and all were of mild intensity. One subject reported chest pain starting on day 3 at 8:30 AM, lasting 10.5 hours and considered possibly related to treatment. This subject also reported abdominal pain starting on day 3 at 9:30 PM, lasting 1 hour and considered unrelated to treatment. A second subject reported skin irritation starting at 8:30 AM on day 4, lasting 3 days and considered not related to treatment. DISCUSSION

Ibuprofen did not interfere with sleep in this study. Subjects treated with ibuprofen 400 mg TID (maximum daily OTC dose) had a mean sleep efficiency of 88.6% on night 3. This was not statistically different from placebo (87.3%), and was significantly greater than sleep efficiency <75%, which is consistent with insomnia. These objective data were confirmed by the absence of a significantly altered response on the subjectively rated Quality of Sleep VAS. The sleep-recording data compared favorably with normal values, u,12 The ibuprofen and placebo groups had data (night 3) within normal ranges for % Stage 1 (3.2% and 4.5%, respectively; normal, 3%-5%), % Stage 2 (49.8% and 50.0%; normal, 45%-60%),

r ...........................................................................................................................................................................................................................

•

J

C

I I I

I

I

I

I

II

I

I

I

I

0

0

~

~

0

~

0

~

~

~

5 c

Z

2~ °

;

m

~

c

~

i

~

~

~

~

0

~

~

~

"-

C O C O

~

~

i

i .....................................................................................................................................................................................................

1824

:::

i

....

;,

,::i::'

:=

iVolume28 INumber:l:i

Gengo

% Stage 3 + % Stage 4 (16.9% and 15.4%; normal, 10%-20%), % Stage REM (22.9% and 20.0%; normal, 20%-25%), and Latency to Stage REM (88.1 and 82.9 minutes; normal, 70-100 minutes). Stages 3 and 4 are often analyzed in combination because both are considered components of slow-wave sleep and individually are difficult to differentiate. Although a statistically significant difference was found in favor of ibuprofen compared with placebo for % Stage REM, the difference was considered clinically nonsignificant and not related to sleepiness or insomnia. A study conducted by Murphy et al5 concluded that ibuprofen and ASA, but not acetaminophen, had negative effects on sleep patterns. Statistically significant (P < 0.05) differences were found in favor of placebo compared with ibuprofen and ASA for sleep efficiency (96.2% vs 87.5% and 91.0%, respectively). The observations, however, may have been influenced by sleep deprivation from the lack of an appropriate acclimatization period. Typically, sleep efficiency ranges from 80% to 95% in adults aged <65 years. 13 Values >95% are generally associated with subjects who enter sleep following a period of relative sleep deprivation. Because of the lack of an extensive adaptation period, the investigators concluded that subjects likely experienced some sleep deprivation (first-night effect) on the night prior to PSG recording. Thus, on the night on which sleep was recorded, subjects were likely to be somewhat more sleepy than normal. % Stage 1, % Stage 2, % Stage 3 + 4, % Stage REM, and Latency to Stage REM were statistically similar between the ibuprofen and placebo groups and were within the normal ranges. Additionally, subjective sleep questionnaires did not reveal any significant differences between placebo, ibuprofen, aspirin, and acetaminophen, and no adverse effects were reported. Several study-design elements in the Murphy study were modified for the current study in an effort to standardize procedures with more appropriate sleepstudy methodology. First, the Murphy study had no prestudy regimented sleep schedule, whereas the current study had a 10-day regimented sleep period. In both studies, all subjects were instructed to sleep from 12 to 8 AM (normal sleep duration, 7.5-8.0 hoursll,12). The Murphy study employed only 2 nights in the sleep laboratory, with recordings only on night 2 (after treatment). The current study employed 3 nights in the sleep laboratory, with recordings on nights 2 (baseline) and 3 (after treatment). [ November

2006

:. [i[:.

.

.

:

Second, Murphy randomly assigned subjects to receive ibuprofen or placebo on the first day before any sleep recordings were made, with only historical assessment of usual sleep patterns. In the current study, randomization occurred only after a 10-day regimented sleep phase and 2 nights in the sleep laboratory, with night-2 sleep recordings confirming subjects' normal sleep patterns. Third, if acute effects of ibuprofen on sleep efficiency and sleep quality were to be seen, they would have been more pronounced in the current study because the total daily dose was administered within 9 hours prior to bedtime. In the Murphy study, double-blind administration of ibuprofen 400 mg or placebo was conducted at 11 em on night 1, 8:15 AM on day 2, and 11 I'm on night 2 (25, 15.75, and 1 hour, respectively, before the posttreatment sleep recording), and the first ibuprofen dose was administered immediately before the acclimatization night of sleep. However, the current study used single-blind placebo at 3, 7, and 11 PM on nights 1 and 2, and double-blind ibuprofen or placebo at 3, 7, and 11 I'M on night 3 (9, 5, and 1 hour, respectively, before the posttreatment sleep recording). The results of the present study found no clinically significant differences and virtually no statistically significant differences in sleep patterns between ibuprofen and placebo when compared in a sleep study with an adequate acclimatization period, an appropriate time of randomization, a rigorous clinical dosing schedule, and with sufficient power to detect treatment differences. CONCLUSION

This study found that in these subjects a total daily dose of 1200 mg ibuprofen did not produce any clinically or statistically significant alterations in the character and quality of nighttime sleep as measured using standard sleep laboratory PSG and subjective measures. ACKNOWLEDGM

ENTS

The research and publication of this study were financially supported by Wyeth Consumer Healthcare, Madison, New Jersey. The author acknowledges Terry Fullerton, PharmD; Donald Reitberg, PharmD; Vernice Bates, MD; B. Seigal; Tomas Holmlund, MD; Laszlo Mechtler, MD; and Donna Cwudzinski, MD (study coordinator), for their contributions to the research. :

•. . . . . .•

'....

:

:, :

....

82s

REFERENCES

American Medicine. New York, NY: WebMD; 1997. 13. Ohayon MM, Carskadon MA, Guilleminault C, Vitiello MV. Metaanalysis of quantitative sleep param-

1. Medical Economics Data. Physician's Desk Reference. 47th ed. Montvale, NJ: Medical Economics; 1993:2529-2530. 2. Badia P, Boeker M, Myers B. Bright light effects on temperature, SCR, cognition, and sleep. Paper presented at: 10th Congress of the European Sleep Research Society; May 20-25, 1990; Strasbourg, France. 3. Horne JA, Percival JE, Traynor JR. Aspirin and human sleep. Electroencephalogr Clin Neurophysiol. 1980; 49:409-413. 4. Hayaishi O. Molecular mechanisms of sleep-wake regulation: Roles of prostaglandins D2 and E2. FASEBJ. 1991 ;5:2575-2581. 5. Murphy PJ, Badia P, Myers BL, et al. Nonsteroidal anti-inflammatory drugs affect normal sleep patterns in humans. Physiol Behav. 1994;55:10631066. 6. 1983 Metropolitan height and weight tables. Stat Bull Metrop Life Found. 1983;64:3-9. 7. Hariharasubramanian N, Nair NPV, Pilapil C. Circadian rhythm of plasma melatonin and cortisol during menstrual cycle. In: Brown GM, Wainwright SD, eds. The PinealGland: Endocrine Aspects. Toronto, Ontario, Canada: Pergamon; 1985:31. 8. Webley GE, Leidenberger F. The circadian pattern of melatonin and its positive relationship with progesterone in wo men.J Clin EndocrinolMetab. 1986;63:323-328. 9. Rechtschaffen A, Kales A, eds. A

eters from childhood to old age in healthy individuals: Developing normative sleep values across the human lifespan. Sleep. 2004:27:12551273.

Manual of Standardized Terminology, Techniques, and Scoring System for Sleep Stages of Human Subjects. Washington, DC: US Dept of Health, Education, and Welfare Public Health Service; 1968. Publication 204. 10. Kryger MH, Roth T, Dement WC.

Principles and PracticeoflSleep Medicine. Philadelphia, Pa: Saunders; 1989. 11. Adams RD, Victor M. Principles of Neurology. 4th ed. New York, NY: McGraw-Hill; 1991. 12. Disorders of sleep. In: Dale D, Federman D, eds. WebMD Scientific

1826 :

:

~

:

Address correspondence to: F r a n c i s G e n g o , P h a r m D , D E N T N e u r o l o g i c a l Institute, 3 9 8 0 S h e r i d a n Drive, A m h e r s t , N Y 14226. E-mail: fgengo@ buffalo.edu

"

i i ii

•

i

....

ii:ii:: 'Volume28 Number !!ili : • Z:::

: :iq::'~i,: