Sleep Aid Prescribing Practices during Neoadjuvant or Adjuvant Chemotherapy for Breast Cancer

Background

Breast cancer is the most common cancer in women in the United States. Many women with early-stage breast cancer receive adjuvant chemotherapy, frequently anthracyclines, which reduces the annual breast cancer death rate by 20% to 38%. ¹

Sleep disruption is common in patients receiving chemotherapy. Insomnia is defined as difficulty falling or staying asleep, early awakening, or nonrestorative sleep occurring at least three nights per week and causing distress or impairment of daytime functioning. ² Approximately 30% to 50% of cancer patients have sleep difficulties, ³ which is higher than the general population (estimated to be 20% of adults and 30% of the elderly). ⁴ In a study of 823 subjects receiving chemotherapy for a variety of cancers, 36% reported insomnia symptoms, with 43% meeting diagnostic criteria for insomnia syndrome. ² Among them, breast cancer patients had the highest number of overall sleep complaints.

Sleep disturbances in breast cancer patients receiving chemotherapy have been reported in several studies and are associated with significant morbidity. In a prospective study, women with Stage I or II breast cancer receiving their third cycle of adjuvant chemotherapy experienced poor sleep quality and daytime sleepiness. ⁵ Chemotherapy leads to impaired sleep-wake activity rhythms that are progressively worse and longer lasting with repeated cycles. ⁶ Disrupted circadian rhythms during chemotherapy are associated with fatigue and depressive symptoms. ⁷ In one study, 65% of women reported poor sleep before chemotherapy began and actigraphy showed sleep fragmentation during the first 3 nights after chemotherapy. ⁸ A pre-treatment symptom cluster of sleep disturbances, fatigue, and depression is associated with a significant worsening of symptoms during chemotherapy. ⁹ A meta-analysis showed that breast cancer patients receiving chemotherapy had poor sleep with frequent awakenings and daytime sleepiness. ¹⁰ Thus, not only are sleep disturbances during chemotherapy for breast cancer common, they are associated with poorer quality of life, increased morbidity, and increased risk of psychiatric disturbances.

Behavioral sleep interventions in breast cancer patients undergoing adjuvant chemotherapy improve sleep, which correlates with less fatigue, fewer depressive symptoms, and higher performance status.^7,11,12 Nevertheless, many breast cancer patients receive sleep aid medications. Medications commonly used include short-acting nonbenzodiazepene hypnotics (zolpidem, zaleplon, eszopiclone), benzodiazepenes (temazepam, lorazepam), melatonin receptor agonists (ramelton), or anti-histamines. The National Institutes of Health (NIH) State of the Science Conference on insomnia concluded that benzodiazepine receptor agonists were efficacious in the short-term treatment of insomnia. ¹³

Hot flashes and postmenopausal status, which are common in women with breast cancer, affect sleep quality. Hot flashes are associated with less efficient and more disrupted sleep in breast cancer patients after completion of chemotherapy. ¹⁴ In a secondary analysis of 219 women with breast cancer, postmenopausal women had less total sleep time as well as worse objective sleep when compared with premenopausal women, and pre- and perimenopausal women had more sleep interruptions due to hot flashes. ¹⁵ We therefore hypothesized that menopausal status effects sleep aid use during chemotherapy.

To our knowledge, there have been no studies evaluating the sleep aid medication prescribing practices of breast cancer medical oncologists. This study aimed to describe the sleep aid prescribing practices of oncologists treating women with breast cancer receiving adjuvant or neoadjuvant chemotherapy at the University of California, San Diego.

Methods

This study was approved by the University of California, San Diego Human Research Protection Program. Patients who received four cycles of neoadjuvant or adjuvant Adriamycin® and cyclophosphamide (AC) at the University of California, San Diego from April 1, 2008 to March 31, 2010 were evaluated by retrospective chart review. Our primary objective was to characterize the prescribing patterns of sleep aid medications for patients receiving neoadjuvant or adjuvant chemotherapy for breast cancer. Secondary objectives included the effects of menopausal status on sleep aid use and the local feasibility of a chart-based retrospective review using the University of California, San Diego electronic medical record (EpicCare).

Demographics and clinical data pertinent to sleep disorders were collected. A report of electronic prescriptions for sleep aids was generated utilizing the electronic medical record (EMR). Medications were considered sleep aids only if described as being prescribed or used for sleep. Descriptive statistics and χ² test were used for all data.

Results

Demographics and tumor descriptions are listed in Table 1. Of the 124 subjects, 52.4% discussed sleep with their provider while on active chemotherapy. Whereas 13.7% of patients reported sleep aid use prior to the start of chemotherapy, 32.3% were prescribed sleep aids during chemotherapy (a total of 51 prescriptions). Sleep aids most commonly prescribed were lorazepam (31.4%) and zolpidem (29.4%) (Table 2).

Women prescribed sleep aids during chemotherapy were significantly more likely to discuss sleep with their provider (p < 0.0001), more likely to have been taking sleep aids previously (p < 0.0001), and were more likely to be taking psychiatric medications (p = 0.04). There was a trend toward increased sleep aid use in those with psychiatric diagnoses (p = 0.068). Menopausal status was not significantly associated with sleep aid use during chemotherapy.

Conclusions

We found that sleep disturbances during AC chemotherapy for breast cancer are common at the University of California, San Diego and are frequently treated with sleep aids. We successfully used the EMR reporting function to identify subjects and collect sleep aid prescribing data.

Though menopausal status had no effect on sleep aid use in our study, prior reports suggest that hot flashes are associated with poorer sleep in breast cancer patients.^14,15 It is possible that the high proportion of premenopausal patients (57.3%) may have minimized our reported prevalence of sleep aid use.

Our study suffers from the limitations of a retrospective chart review. It is possible that sleep disturbances were discussed with providers, but not documented in the EMR, or that subjects took sleep aids that were never reported. There may also be provider-specific variation in discussing sleep with patients and/or willingness to prescribe sleep aid medications. However, our data are consistent with prior studies of insomnia in breast cancer patients.^5–8,10

Although over half of subjects discussed sleep with oncology providers, it is possible that the prevalence of sleep aid use (and by extension, sleep disturbance prevalence) would have been higher if providers were encouraged to ask about sleep. In patient-reported surveys demonstrating that cancer-related fatigue is prevalent, fatigue was not reported to the provider 9% to 52% of the time,^16,17 perhaps because nearly half of the subjects believed that nothing could be done to relieve their symptoms. ¹⁷ Sleep disturbances are significantly correlated with cancer-related fatigue ¹⁸ and other morbidities, which presents opportunities for providers to discuss and intervene. In the future, the EMR could prompt providers to ask about sleep or to assist in identifying sleep interventions.

This study demonstrated that women with prior sleep aid use and concurrent use of psychiatric medications were more likely to need sleep aids during chemotherapy. Thus, this population should be considered at risk for sleep disturbances during chemotherapy. Our findings imply that preemptive discussion of sleep and earlier consideration of sleep aids or other interventions may be appropriate in these higher-risk individuals. We plan to target this high-risk population for a prospective study.

A better understanding of sleep disturbances and appropriate interventions will favorably impact quality of life for early-stage breast cancer patients receiving chemotherapy, including less fatigue, fewer depressive symptoms, and higher performance status. Furthermore, at least one study showed that decreases in quality-of-life scores correlate with early chemotherapy discontinuation, ¹⁹ implying that preemptive and effective management of sleep disturbances may improve compliance with planned treatment regimens and perhaps improve long-term outcomes. Sleep interventions may also be applicable to breast cancer survivors and to those with metastatic breast or other types of cancer. Further investigation in this area is needed.