Association between poor clinical prognosis and sleep duration among breast cancer patients 1

ABSTRACT Objective: to investigate the association between clinical progression and the quality and duration of sleep in women with breast cancer. Method: longitudinal study, with 114 participants, conducted in a hospital in Brazil. The instruments used were: questionnaire for sociodemographic and clinical characterization, Pittsburgh Sleep Quality Index; Beck Depression Inventory and Herth Hope Scale. Data were analyzed through descriptive statistics and survival analyses (outcome: poor clinical progression), using the Kaplan-Meier curve, Log-rank test and Cox proportional model. Results: a higher probability of poor clinical progression was verified in women with sleep durations of less than six hours or nine hours and over (p=.0173). Conclusion: the results suggest the importance of further studies that seek to verify whether the quantitative management of sleep disorders would have an impact on the progression of breast cancer. Women should be encouraged to report sleep problems to nurses.


Introduction
The clinical outcome of breast cancer and the survival of the women affected depend on some extensively studied prognostic factors, such as: the presence of lymph node metastases, the size and histological type of the tumor, its nuclear grade, the estrogen (ER) and progesterone (PR) hormone receptor status, the presence of the Human Epidermal growth factor Receptor 2 (HER-2) oncogene and the tumor cell proliferation (1) .
The probability of survival for five years for patients with early stage I cancer was estimated to be 97%, while for women with the more advanced stage (stage IV) this was 57%, according to a study of 252 women with breast cancer at a university hospital (2) . It also demonstrated that the probabilities of survival for ten years for the initial and advanced stage, were 97% and 0%, respectively (2) .
Recently, the literature has shown an association between the prognosis and clinical progression of breast cancer and sleep of poor quality or of insufficient or excessive amounts for the needs of the individual (3) .
A study that aimed to verify an association between breast cancer symptoms and poor quality of sleep concluded that 65% of the women suffered from sleep problems (4) . This study is relevant considering that the literature suggests that poor sleep quality, in addition to having a negative impact on the daily routine of the affected patient, may be associated with poor clinical evolution of the cancer. The sleep characteristics that contribute to poor quality include; fragmented sleep and, consequently, its low efficiency; prolonged latency and early awakening; and a non-restorative sensation of sleep (3) ; Thus, the set of results can be interpreted and referred to as poor quality sleep.
In a recent study, in 2016, with 1011 people, between 35 and 45% presented problems of either duration or quality of the sleep and the authors concluded that in Australia sleep problems are an endemic health problem that needs interventions at policy level (4) . Thus, it was hypothesized that the poor quality of sleep, as well as inadequate duration, could constitute aggravating factors in the clinical progression of women with breast cancer.
According to population and evidence based studies found in the literature, the excessive or insufficient duration of sleep may be related to an increased risk for breast cancer, which highlights the requirement for more studies on this topic (5)(6) . Sleep restriction produces a stress reaction in the body, being best characterized immunologically as an increase in the leukocyte and neutrophil count and C-reactive protein serum levels.
Authors have shown that a single night of sleep restriction may be sufficient to cause this increase and that to return to the base count requires an recovery time of eight to ten hours of sleep (7) .
Stress has been associated with the progression of cancer, particularly breast cancer (8) . It should be highlighted that stress increases the synthesis of glucocorticoids, which alters the immune response, as well as cell proliferation and apoptosis in various tissues, and may be one of the mechanisms by which sleep deprivation is associated with increased incidence of breast cancer (9) .
Mutilating treatment, which is often necessary, can cause women to undergo alterations in their selfimage, functional loss and mental, emotional and social changes, frequently leading to depression (10) . Authors have verified that poor quality of sleep in patients with cancer is associated with the presence of depression, anxiety, pain and impairment of the sense of well-being, factors that also affect the quality of life of people with cancer (11) . In contrast, in several studies, it is suggested that hope can be an effective strategy to help patients cope with the difficulties and achieve their goals, especially regarding cancer patients (11) .
Globally, breast cancer is the second most common type of cancer and the most common in women. It is the leading cause of cancer death in women worldwide, with an estimated 520,000 deaths in 2012 (12) . It can be considered an epidemic and, in Brazil, its high incidence and mortality due to late diagnosis has been verified (12) .
These data demonstrate the relevance of breast cancer related studies.
Thus, the aim of this study was to investigate the association between clinical progression and quality and duration of sleep in women with breast cancer.  women were included in the study. All the women who met the inclusion criteria were included in this study. The following criteria were used for the selection: Inclusion criteria: women 18 years of age or over; diagnosed with breast cancer, T qq N qq M 0 of any stage (T qq = any tumor extent; N qq = any regional lymph node invasion, M 0 = clinically no distant metastasis) (13) ; who agreed to participate in the study. Exclusion criteria: Karnofsky Scale score lower than 70 (capable of caring for oneself, incapable of normal activity or work); inadequate medical (such as mucositis, pain, nausea, shortness of breath, vomiting) or emotional (such as crying, apathy, aggressiveness) conditions to respond to an interview.

Methods
The women who left the study for reasons unrelated to the expected outcome were considered dropouts.
Data collection: the researcher collected the data, with the patients being recruited prospectively. Regarding the prognostic factors for control, the following were considered: the presence of lymph node metastases, the stage, I, II or III, and size of the tumor, the estrogen (ER) and progesterone (PR) hormone receptor status, and the presence of the Human Epidermal growth factor Receptor 2 (HER-2) oncogene (2) .
Instruments used:

1) Sociodemographic and Clinical Characterization
Questionnaire -evaluated by a committee of judges, specialists in oncology and sleep, applied at the beginning of the study for the sociodemographic and clinical characterization of the participants. The clinical data were obtained from the medical records.

2) Pittsburgh Sleep Quality Index -PSQI-BR, validated
in Brazil (14) . The scale subjectively assesses sleep quality and disturbances for the previous month. It contains 19 questions grouped into seven components: subjective sleep quality, latency, duration, efficiency, sleep disturbances, use of sleep medication and daytime dysfunction. The total score ranges from 0-21 points, with scores higher than five indicating poor quality of sleep. The higher the value obtained, the worse the evaluation of sleep quality, and the overall score of five points constitutes the cut-off point that distinguishes between subjects with poor sleep and sleep disorders (above five points) and those with good quality sleep (five points or less). The full PSQI was used for investigation of the association with the overall score.

3) Beck Depression Inventory -BDI, validated in
Brazil (15) . The original scale consists of 21 items, including symptoms and attitudes, with their intensity varying from zero to three. The items refer to sadness, pessimism, sense of failure, lack of satisfaction, guilt, feeling of punishment, self-deprecation, selfaccusations, suicidal ideation, crying spells, irritability, social withdrawal, indecisiveness, distortion of body image, inhibition to work, sleep disorders, fatigue, loss of appetite, weight loss, somatic concern, and decreased libido. The recommended cutoff points are: less than ten points, no or minimal depression; 10 to 18, slight to moderate depression; 19 to 29, moderate to severe depression; 30 to 63 points, severe depression. In this study, subjects in the first category were grouped as 'without depression' and those in the other three as 'with depression'.
4) The Herth Hope Scale -HHS is a self-report scale validated for use in Brazil, with adequate psychometric properties (16) . It is designed to facilitate the assessment of hope at various intervals from which variations in its

Results
The population was characterized as having a mean  Table 1.        The majority of these studies refer to cardiovascular diseases and metabolic syndrome (17) . There are a more significant number of studies that attempt to associate the quantitative and qualitative changes of sleep with the risk of developing breast cancer, however, they do not investigate the possible consequences of sleep disorders on the progression of the disease (2) . Thus, there is a gap, which this study sought to address.
In a recent consensus of the American Academy of Sleep Medicine, specialists highlighted that sleep duration of less than seven hours is associated with, among other health problems, an increased risk of mortality and damage to the immune system (17) .
However, according to the authors, there is insufficient evidence to show that sleeping longer than nine hours is associated with health problems (17) . pain, and mortality (17) . The adverse outcomes related to insufficient sleep are stress, pain, neurocognitive dysfunction, psychiatric symptoms, and mortality (18) .
Specialists have also emphasized the importance of the subjective evaluation of this parameter, as well as the relevance of differences among individuals, stating that people who realize that they are sleeping very little, or too much, should be instructed to seek a health professional (19) .
It was possible to show that people who had grade-3 tumors, or the fastest growing, actually had fewer hours of sleep compared to others. New studies linking sleep apnea and other disruptions to cancer add to evidence that poor sleep can be deadly (20) .
The majority of the women in this study (56.1%) presented poor quality sleep, a slightly lower percentage to that found in another study, in which 61% of the women with breast cancer presented poor quality sleep at the start of the treatment (21) . Regarding the mean sleep duration of 6.5 hours found in the present study, this was lower than that found by other authors, of 7.0 hours at the start of the treatment in women with breast cancer (22) . indicating that the management of these symptoms in these women is necessary and should be a priority (23) .
The literature shows that breast cancer is commonly associated with various symptoms, such as depression, pain, fatigue and poor sleep quality (9,24) .
In this study, depression, although present in the majority, presented no influence on the clinical progression of the disease. A recent metaanalysis demonstrated the efficacy of interventions for psychosocial outcomes following breast cancer surgery, with cognitive behavioral therapy promoting improvements in anxiety, depression, and quality of life (24) .
The hope scores, for the women with poor clinical progression, were slightly above the mean of the possible variation, showing that studies directed toward the development of strategies are necessary to encourage hope in these women. Hope is an underinvestigated variable and has the potential to become an effective tool for use in the quotidian of health professionals (25) , however, in this study, it presented no effect on the disease progression.
Similarly, there was no significant relationship between the prognostic factors of the cancer and clinical progression by the end of the follow-up period. Other authors, in analysis of estimated survival over five years, found the following to be factors associated with the risk of death: tumor size, lymph node involvement, number of lymph nodes removed and estrogen and HER-2 receptor tumor markers (2) . This was, however, a retrospective study, using a longer follow-up period than the present study, as is the case for the majority of www.eerp.usp.br/rlae 7 Mansano-Schlosser TC, Ceolim MF.
the studies that take this approach, which are generally cohort studies with a five to ten year follow-up period (2) .