Sleep Disruption

The prevalence of sleep complaints in cancer patients has been studied primarily in cross-sectional studies using convenience samples with heterogeneous definitions and measures of sleep disturbances. In a large questionnaire study of over 900 patients with different types of cancer, fatigue (44%), leg restlessness (41%), insomnia (31%), and excessive sleepiness (28%) were the most prevalent complaints.⁴ Another survey showed that 61% of the cancer patients had significant sleep deficits, but there was no difference in sleep complaints between the cancer patients and patients with medical conditions other than cancer.⁵ Almost half of the group had a poor sleep efficiency (defined as the percentage of time spent asleep) below 85%. Those patients receiving radiation or chemotherapy tended to have more sleep disturbances than those not receiving treatment. In addition, sleep problems predicted deficits in quality of life.

Prevalence rates of insomnia symptoms have ranged from 30% to 50%.^6,7 Only two studies have attempted to distinguish between having subclinical and clinical levels of insomnia in cancer patients. These studies found prevalence rates of insomnia symptoms of 48% in breast cancer survivors and 32% in prostate cancer survivors.^8,9 With insomnia syndrome defined as sleep-onset latency or wake after sleep onset greater than 30 minutes, at least 3 nights per week, associated with a sleep efficiency lower than 85% and significant daytime impairments or marked distress, the prevalences were 19% and 18%, respectively.^8,9 In 95% of the cases, the insomnia syndrome was chronic (i.e., duration > 6 months).

Not much is known about how sleep disturbances vary as a function of cancer sites. In one study, the prevalence of sleep disturbances was greater in breast cancer patients,⁴ but in another study the prevalence was greater in ovarian cancer patients.¹⁰ Moreover, as most studies have been conducted several months or even years after patients completed their cancer treatment, the extent to which insomnia symptoms are exacerbated by cancer treatments is unclear.

Large-scale epidemiologic and longitudinal studies are needed to better depict sleep difficulties in cancer patients, characterize the natural course of sleep disturbances throughout cancer care in terms of incidence and remission, and enable comparison across cancer sites and other cancer characteristics. One ongoing longitudinal study conducted among 998 patients with mixed cancer sites has found that the overall prevalence rates of the insomnia syndrome and of insomnia symptoms (including those with an insomnia syndrome) at baseline (T1—i.e., before or right after the surgery) were 27.2% and 56.4%, and they decreased to 21.3% and 39.5% 2 months later (T2), respectively.¹¹ The prevalence rates of insomnia symptoms were highest in patients with breast cancer and gynecologic cancer and lowest in patients treated for prostate cancer. The incidence of insomnia from T1 to T2 was 19%, and the remission rate was 32.0%. Together, these data indicate that insomnia is already prevalent at the time of cancer diagnosis and surgery. Future analyses will determine the course of insomnia symptoms as adjuvant treatments (e.g., chemotherapy, radiation therapy, hormone therapy) are introduced.

Sleep disturbances are also very common in patients with advanced cancer. In a study of patients with metastatic breast cancer, 63% reported sleeping difficulties. Difficulty falling asleep was associated with both depression and pain, whereas increased awakening during the night was associated just with depression.¹² In another study of 100 palliative care patients attending a pain and symptom control clinic, 72% reported sleep disturbances (63% reported difficulty staying asleep; 40% reported difficulty falling asleep).¹³ Difficulty falling asleep was mostly associated with fatigue and anxiety, whereas early awakening was more strongly associated with fatigue. In a prospective study, 25.9% of terminally ill cancer patients reported sleep disturbances at admission to a palliative care unit.¹⁴ Another study found that patients with advanced lung cancer reported poorer sleep and more daytime sleepiness than healthy controls, and that sleep disturbances of lung cancer patients were characterized by breathing difficulties, cough, nocturia, and frequent awakenings, all of which may be suggestive of sleep-disordered breathing.¹⁵ Finally, a study suggested that poor sleep quality and use of sleep medications were, along with hopelessness and depression, the best predictors of desire for hastened death in 102 terminally ill patients attending a palliative care unit,¹⁶ thus emphasizing the importance of offering appropriate sleep management to these patients.

A few objective studies of cancer patients, using either polysomnography or actigraphy, have been conducted to characterize the sleep disturbances. When the sleep of patients with breast or lung cancer, patients with insomnia, and volunteers with no sleep problems were compared, the insomnia patients had the shortest total sleep time on polysomnography, but the lung cancer patients had the longest sleep onset latency, the lowest sleep efficiency, and the greatest wake time during the night.¹⁷ There were no differences in stress levels or emotional state between the cancer patients and the volunteers. There also was no difference in reported total sleep time between the cancer patients and the volunteers. Interestingly, unlike the insomnia patients, the cancer patients did not underestimate total sleep time or overestimate wake time during the night.

In the same study, although there was no difference in the amount of sleep-disordered breathing, the cancer patients had a higher prevalence of periodic limb movements in sleep (PLMS) than insomnia patients or healthy volunteers.¹⁷ However, more recently, in two small-scale studies (17 and 33 patients), an elevated prevalence of obstructive sleep apnea (OSA), ranging from 12% to 91.7%, was found in patients with head and neck cancer.¹⁸ Prospective studies are warranted to investigate to what extent OSA is caused by the cancer itself or by the cancer treatment. Sleep-disordered breathing also appears to be frequent in patients with brain tumors, with tumor removal resulting in a significant decrease in the apnea–hypopnea index.¹⁹ An ongoing research study indicates that the prevalence of OSA in women with breast cancer who have completed chemotherapy was 48%, and the prevalence of PLMS was 36%.²⁰ These high prevalence rates of PLMS and OSA may help explain some of the sleep disturbance found in this population.

Actigraphy, a noninvasive, continuous, ambulatory measure of circadian rest–activity rhythms, has also been used to characterize the sleep and rhythms of patients with cancer.^21–24 Studies comparing cancer patients to healthy controls have consistently shown less contrast between daytime and nighttime activity in cancer patients, a pattern indicative of circadian disruption.^21,25 In a study of 85 women with breast cancer, 72-hour actigraphy before the start of chemotherapy demonstrated a mean total sleep time of 6 hours, with only 76% of the night spent asleep. On average, the women napped for about 1 hour a day.²³

Fatigue

Fatigue is one of the most frequent and disturbing complaints of patients with cancer^3,26: more than 75% of patients who undergo chemotherapy or radiation therapy report feeling weak and tired. Cancer-related fatigue has been defined as a “persistent, subjective sense of tiredness related to cancer and cancer treatment that interferes with usual functioning.”²⁷ It is believed to be distinct from general fatigue, as it is unrelated to exertion level and is not relieved by rest or sleep. It has been reported that 76% of patients receiving chemotherapy report fatigue at least a few days each month,²⁸ interfering with daily life, reducing quality of life,^3,29 and being one of the key reasons for discontinuing treatment.

An increasing number of studies in the past few years have followed cancer patients over time. Overall, these studies suggest that fatigue is highly prevalent before as well as during and after treatment.^26,29 In one study, 66% of the women reported at least some fatigue before treatment and 84% reported fatigue during treatment.²³ Additionally, the percentage of women reporting extreme fatigue doubled from approximately 5% before treatment, to approximately 10% during treatment. Several studies have suggested that fatigue can continue for months, and even years, after the completion of therapy. A recent systematic review of the literature identified 10 longitudinal studies on cancer-related fatigue and concluded that cancer-related fatigue may persist for up to 5 years after completion of adjuvant treatments.³⁰

Pathogenesis of Sleep Disruption

Patients with cancer may complain of insomnia, hypersomnia, or both, but the pathogenesis of this sleep disruption can be quite varied. Chemotherapy, radiation therapy, and hormone therapy may all contribute to the problem, but studies looking at their different effects on sleep patterns are lacking. In addition, commonly administered analgesics such as opioids, and antiemetic medications such as corticosteroids, are also known to disrupt sleep.³¹ The estrogen deficiency induced by chemotherapy and hormone therapy, the abrupt cessation of hormone replacement therapy at cancer diagnosis, or an ovary removal may each trigger or exacerbate preexisting hot flashes. A study using objective measurements of both sleep and hot flashes in breast cancer survivors showed that nocturnal hot flashes were associated with more wake time and more stage changes to lighter sleep.³²

The amount of insomnia in cancer patients can be as high as the amount found in depressed patients, so clinicians should not overlook the possibility that poor sleep in cancer patients may indicate some psychological distress. However, there is evidence that, although insomnia and psychological distress are interrelated, there are still a significant proportion of patients who have only insomnia. In one sample of newly diagnosed breast cancer patients, insomnia was the most frequent symptom, reported by 88% of the patients, and was correlated with high levels of distress and anxiety.³³ However, contrary to the belief that disturbed sleep before treatment is attributable to the increased stress and anxiety resulting from a recent diagnosis of a life-threatening illness, insomnia and fatigue were rated high even in those patients who rated themselves low on anxiety. Similarly, another study revealed that 46% of prostate cancer survivors with an insomnia syndrome did not have clinical levels of anxiety or depressive symptoms.⁹

Pain has often been thought to be the cause of sleep disruption, not only in patients with cancer but also in patients with a multitude of other medical conditions.² It is not yet known whether the pain contributes to poor sleep or whether the pain medications contribute to poor sleep, or both. One hypothesis is that pain may be the initial cause of the frequent awakenings, but psychological distress prevents the patient from falling back to sleep.³⁴ A second hypothesis is that while sleep leads to recovery and repair of tissue and may offer a temporary cessation of the psychological awareness of pain, poor sleep leads to difficulty managing pain.³⁵ In this way, a cycle of pain and poor sleep may become self-perpetuating. In a study examining the relationship between pain and sleep disruption, patients with breast cancer, lung cancer, insomnia (with no cancer) and normal controls were questioned. Although those with breast cancer reported pain before bedtime, neither their poor sleep nor that of the patients with lung cancer was associated with reports of pain.¹⁷ Another study conducted in patients with advanced cancer showed significant correlations between pain and poor sleep quality.³⁶ Moreover, those patients with poor quality of life had the most disturbed sleep and the highest levels of pain.