Predictors of Health-Related Quality of Life in Stroke Survivors After Inpatient rehabilitation

Background. Disordered and decreased quality of life is the most important consequence of stroke for stroke survivors. The aim of the study was to determine the predictors of the health-related quality of life (HRQOL) in stroke survivors six months after discharge from inpatient rehabilitation. Methods. We conducted prospective cohort study which involved 136 (48.5% males and 51.5% females) survivors. We examined seven potential predictors of HRQOL: age, sex, stroke type, stroke side, functional status (Barthel Index-BI and modified Rankin Scale-mRS), cognition (Mini-Mental State Examination; MMSE) and stroke severity (National Institutes of Health Stroke Scale; NIHSS). HRQOL was assessed by Stroke Impact Scale (SIS) 3.0. Using Pearson's correlation and multiple logistic regression analysis we described the relationships between mRS, NIHSS, BI, MMSE and HRQOL. Results. Baseline mRS and NIHSS scores negatively correlated with seven SIS domains except with strength. Baseline BI scores positively correlated with seven SIS domains except mobility and baseline MMSE scores positive correlated with memory, ADL, hand function and participation role and negatively correlated with emotion, communication and mobility domains. Decrease of both mRS and NIHSS scores during the observed period positively correlated with increase of all SIS domains. Ischaemic stroke positively correlated with emotion and communication and stroke in brainstem negatively correlated with communication domain. Memory domain positively correlated with female sex and with stroke in the right hemisphere. Age wasn't significantly associated with any SIS domain. Conclusion. We conclude that major factors in predicting the improvement of hand and were increase of BI and decrease of mRS and NIHSS scores. Female sex, stroke in right hemisphere and increase of BI and MMSE scores predicted better memory. Baseline mRS and NIHSS scores were predictors for improvement of hand function and increase of mRS predicted decrease of hand function and participation role.

strength, physical functioning, mobility, hand function, ADL and participation role were increase of BI and decrease of mRS and NIHSS scores. Female sex, stroke in right hemisphere and increase of BI and MMSE scores predicted better memory.
Baseline mRS and NIHSS scores were predictors for improvement of hand function and increase of mRS predicted decrease of hand function and participation role.
Key words: stroke, survivors, health-related quality of life, inpatient rehabilitation Background Stroke is one of most devastating neurological diseases and one of the leading causes of significant disability [1]. The World Health Organization (WHO) defines stroke as a focal or global neurological impairment of sudden onset, lasting more than 24 hours or leading to death of presumed vascular origin [2].
Stroke is the second cause of death worldwide and 87% of recorded deaths occur in low-and middle-income countries [3]. Stroke mortality declined in high-income countries like the United States of America (USA) [4,5] and today stroke is on the fourth place as cause of death in the USA [6,7].
More than one third, out of 15 million people who suffered from stroke every year worldwide, will die and about 30% of stroke survivors will be significantly disabled [8,9]. Both physical and mental disabilities are common after stroke but the cognitive impairments are very often unrecognized despite of their high prevalence which ranged from 20% to 80% [10]. While physical disabilities tend to improve during the time cognitive impairments are progressively worsening [10,11].
Impaired and decreased quality of life is the most important consequence of stroke for stroke survivors (12). Health-Related Quality of Life [HRQOL] is a broad, multidimensional concept referring to those aspects of people's lives that reasonably relate to their health [13].
Age, sex, socioeconomic status, stroke severity, motor impairments and depression are some of factors associated with poor HRQOL in stroke survivors [14][15][16]. Static and dynamic balance ability, cognition function, motor function, neurological deficit, stroke side and stroke type may influence on the total HRQOL or on some domains such as strength, memory, hand function, emotions, activity of daily living (ADL) [17,18].
Assessment of the HRQOL is important measure of stroke treatment and rehabilitation efficacy [19]. It is assessed that only 10% of survivors make a full recovery, about 25% of all survivors recover with minor impairments and about 40% all survivors continue to live with moderate disabilities, while 15 to 30% live with severe disabilities and are full dependent [20]. About 25% of survivors need additional treatment and rehabilitation in specialized institutions [21].
The aim of the study was to determine the potential predictors of the HRQOL in stroke survivors six months after discharge from inpatient rehabilitation.

Methods
We carried out prospective cohort study which involved at first 216 survivors. Out of all of them 11 dropped out, 9 died, 60 didn't go to inpatient rehabilitation. Finally, cohort included 136 stroke survivors aged 40-79 years of age from the teritory of the Southeastern Serbia. All stroke survivors had inpatient rehabilitation in tertiary health institutions in the period of 30 days at Clinic for Physical Medicine and Rehabilitation, of the Clinical Centre Niš, Serbia, followed by prolonged rehabilitation in spa conditions for 30 days under medical control.
In Serbia, most stroke survivors with functional deficit after acute treatment in intensive care units (ICU) are offered 14 to 30 days of inpatient rehabilitation.
Rehabilitation begins in the ICU when patient state is stable and after discharged majority of survivors from the territory of the Southeastern Serbia continuate rehabilitation at the Clinic for Physical Medicine and Rehabilitation, of the Clinical Centre Niš.
The inclusion criteria were: the first-ever stroke, the age from 40 to 79 years, possible communication with all participants and the written consent of all participants to participate in the study.

Functional status
Functional status was assessed with Barthel Index (BI) and by the modified Rankin scale (mRS).
BI measures the patient's performance in 10 activities of Activities of Daily Living (ADL), a group of questions that is related to self-care (feeding, grooming, bathing, dressing, bowel and bladder care, and toilet use) and a group related to mobility (ambulation, transfers, and stair climbing). The maximal score for BI is 100 indicating that the patient is fully independent in physical functioning. The lowest score is 0, representing a totally dependent bedridden state [22].
The modified Rankin scale (mRS) [23] is regarded as a global stroke scale scoring patients' daily functioning from 0 (no symptoms) to 5 (severe disability, bedridden, in need of constant care) or 6 (death). Higher points indicate a severe disability. MMSE is considered as a useful measure for the assessment of cognitive decline in stroke survivors [24]. This scale examines orientation, memorizing, attention and calculation, delayed memory, speaking, reading, writing and drawing. The maximum score is 30, and a score of less than 24 points indicates the cognitive impairments.

The National Institutes of Health Stroke Scale
Neurological impairments were assessed by the National Institutes of Health Stroke Scale (NIHSS). This scale is consisted of 15 items for evaluation of size and location of stroke in the acute phase and as measure of neurological handicap. Total scores range from 0 to 42 and higher scores indicate greater stroke severity [25].

The Stroke Impact Scale
The Stroke Impact Scale (SIS) is a stroke-specific and self-reported questionnaire which consists of 59 items measuring eight domains (strength, hand function, activities of daily living/instrumental activities of daily living, mobility, communication, emotion, memory and thinking, and participation role). Each domain of SIS has a range of 0-100 and higher scores indicated better HRQOL [26].

The statistical analysis
All the calculations were done into the SPSS software package version 18.0 and S-

Results
The total of 136 (48.5% males and 51.5% females) stroke survivors participated in the study. The average age of participants was 63.72 ± 8.73.   Baseline average scores of hand function, mobility, strength, ADL and participation role domains were the most decreased at the admission to inpatient rehabilitation.
The highest average scores were found in communication, memory and emotion domains. Significant increase of all eight domains was determined six months after discharge from inpatient rehabilitation. Table 3 presents Pearson's correlation coefficients of the examined possible predictors and the scores of SIS domains.  Table 3.   p < 0.001) was associated with significantly decreased of physical functioning.
The most important predictors for improvement of hand function were baseline mRS and NIHSS scores and increase of the mRS during the study.
The most important factor predicting decreased of participation role was increase of mRS (95% CI: -35.012 to -24.133; p < 0.001).

Discussion
We found that stroke survivors were significantly disabled with decreased of all eight domains of HRQOL when they were admitted to the inpatient rehabilitation.
Hand function, physical functioning, mobility and ADL domains were the most affected.
According to the presented results, baseline mRS and NIHSS scores were significantly negatively correlated with HRQOL. Decreasing of mRS and NIHSS scores during the six months were positively correlated with the recovery of all affected domains, especially with the recovery of physical functioning, hand function and participation role.
Recovery of physical function in our study had great impact on the improvement of all other domains of HRQOL such as emotion, communication, ADL and participation role. Numminen et al. [27] found that despite of good physical recovery stroke survivors reported some impairments in HRQOL six months after the stroke.
In our study improvement of hand function significantly influenced on the overall HRQOL of survivors. In the study of Carod-Artal et al. [19], NIHSS, BI and mRS were the main predictors for the recovery of strength domain, the MMSE was the main predictor for the memory and communication domains, BI and mRS were predictors for the ADL and hand function domains. Our results are partially in agreement with these findings.
In our study, factors that significantly correlated with lower HRQOL scores were greater stroke severity, physical disability and hand dysfunction. Our findings are in agreement with the findings of Numminen et al. [27], Alguren et al. [28], Pinkey et al. [29] and Yeoh et al. [30].
Stroke severity, dependence in ADL, degree of handicap, and length of hospitalization were important factors associated with worsen quality of life in stroke survivors after thromolitic therapy [27]. Findings of Alguren et al. [28] showed the predictability of baseline NIHSS score for HRQOL in the chronic phase.
Greater stroke severity measured by NIHSS at admission was predictor of worsen HRQOL [29]. Our results are in agreement with results of these studies.
Study of Rønning and Stavem [33] showed that neither age, sex, comorbidity, nor baseline disability was an important determinant of change in HRQOL from one to six months following acute stroke. Only baseline scores in SF-36 questionnaire were associated with changes in HRQOL from one month to six months after stroke.
In our study age was not associated with any SIS domain six months after discharge from the inpatient rehabilitation. One more study didn't find any association between age and sex and HRQOL in stroke survivors [21]. Pinkey et al. [29] found insignificant association between older age and worsen HRQOL in stroke survivors.
Gurcay et al. [16] found that age and functional status were the most important independent factors which were associated with HRQOL.
We found that female sex with stroke in right hemisphere and increase of BI scores during the six months was predictor for improvement of memory domain. In our study right hemisphere stroke predicted lower emotion domain in survivors six months after discharged from inpatient rehabilitation. Chen et al. [21] showed similar findings. Yeoh et al. [30] showed that stroke side only had impact on the motoric functions, but not on the emotional domain of HRQOL.
Different results presented Hopman et al. [34]. They found that the left hemisphere stroke predicted worse emotional functioning even six months of follow-up. Study of de Haan et al. [35] showed that locations of stroke and stroke types did not affect emotional domain of quality of life.
In our study communication domain was significantly negative correlated with stroke in the brainstem. The brainstem stroke can result in loss of motor function and with appearance of other symptoms such as ataxia, double vision, vertigo, and dizziness, difficulty swallowing, speech deficits, numbness, and even paralysis of one side of the body or both.
Niemi et al. [31] found that patients with either a right or left hemisphere stroke had more frequently evidenced deterioration of quality of life than patients with no brainstem lesions. One study showed worse quality of life in patients with left hemispheric stroke [32].
Survivors in our study with ischaemic stroke had significantly higher scores in memory, communication and hand domains compared with those with hemorrhage.
Recovery of hand function wasn't significant until six months after the inpatient rehabilitation and recovery of this domain significantly increases the overall HRQOL of stroke survivors in our study. Morris et al. found that impairment of hand function and ADL independence predicted perceived physical activity [17]. According to the results of Nichols-Larsenet al. [36] worsen HRQOL in the physical domain was associated with age, more comorbidities and reduced hand function.

Conclusion
According to presented results, major factors in predicting the improvement of strength, physical functioning, mobility, hand function, ADL and participation role were increase of BI and decrease of mRS and NIHSS scores. Female sex, stroke in right hemisphere and increase of BI and MMSE scores predicted better memory.
Baseline mRS and NIHSS scores were predictors for improvement of hand function and increase of the mRS predicted decrease of hand function and participation role. We require from all participants the written consent to participate in the study. It was one of condition to involve them to the study.
Consent for publication: "Not applicable"