Disparity in Health-Related Quality of Life Between Kidney Transplant Recipients and Hemodialysis Patients, and Related Implications for Physical Rehabilitation: A Cross-Sectional Study

Background: Health related quality of life (HRQOL) is an important indicator of medical treatment and is a strong predictor of disability and mortality. The literature has shown mixed evidence about whether kidney transplantation improves HRQOL compared with other renal replacement modalities. The aim of this study was to compare the HRQOL in kidney transplant recipients (KTRs) and hemodialysis (HD) patients. Methods: A cross-sectional study of 100 KTRs and 272 HD patients from two central kidney units in West Bank, Palestine. The HRQOL was assessed using the Short Form-36 Health Survey. Multivariable linear regression was used to estimate differences in mean HRQOL scores between KTRs and HD patients. Results: After accounting for sociodemographic variables, KTRs had higher clinically important differences than HD patients in HRQOL mean scores in role-physical, bodily pain, general health, vitality, social functioning, role-emotional, mental health, physical and mental component summaries ranging from 15.5 points for social functioning (95% CI 10.1, 20.7) to 34.3 points for general health (95% CI 28.7, 39.9). However, unexpectedly, KTRs had signicantly lower HRQOL physical functioning than HD patients by 24.5 points (95% CI 18.7, 29.9). Conclusions: Kidney transplantation is associated with important improvements in most aspects of HRQOL but also with noticeably poor physical functioning. HRQOL among KTRs should be routinely measured and closely monitored in clinical settings. KTRs should be encouraged to participate in individually tailored physical exercise programs. Identication and elimination of barriers to physical functioning may improve HRQOL and prevent premature mortality among KTRs.


Background
Chronic kidney disease is a growing global burden in terms of mortality, disability, and nancial costs to the community (1,2). Recent estimates suggest that around 13% of the general population worldwide have some form of chronic kidney disease (3). According to the systematic analysis of the Global Burden of Disease Study 2016, around 1.2 million deaths and 35 million disability-adjusted life years in 2016 were attributed to chronic kidney disease, which represents an increase in mortality and disability by 98% and 62% since 1990, respectively (4). A signi cant proportion of patients with chronic kidney disease progress to end stage renal disease (5,6). Renal dialysis and kidney transplantation are the only two types of treatment currently available for patients with end stage renal disease (7). Globally, around 2.6 million patients received dialysis in 2010, and this number is projected to raise to 4.9 million by 2030 (8). Recent estimates show that around 90,000 kidney transplants were performed globally in 2016, which re ects an increase of 18% since 2015 (9). The literature suggests that kidney transplantation improves survival rates among patients on renal replacement therapy (10,11). However, kidney transplantation requires long-term immunosuppressive therapy, which is associated with signi cant side effects that affect patient's health-related quality of life (HRQOL) negatively, such as recurrent infection, metabolic disorder, renal toxicity, fatigue, and poor self-perception of physical appearance (12)(13)(14).
HRQOL is a signi cant independent predictor of hospitalization and mortality in patients on renal replacement therapy (15,16). HRQOL remains is an important health outcome measure in kidney transplant recipients and in dialysis patients, which help clinicians and patients make rational decisions about the optimal choice of treatment modality. The current literature shows mixed ndings on whether kidney transplantation is associated with clinically important improvement in HRQOL in kidney transplant recipients as compared with kidney dialysis patients (17,18). In addition, clinically relevant HRQOL could be in uenced by disease severity and comorbidities, socioeconomic status, cultural background, religion, environment, and other various factors (19). Research examining HRQOL among kidney transplant recipients (KTRs) and hemodialysis (HD) patients in the Middle East is sparse. The aim of this study was to compare the HRQOL in KTRs and HD patients in West Bank, Palestine.

Study design
We conducted a cross-sectional study among KTRs and HD patients.

Setting and patient recruitment
The study was conducted in two of the twelfth kidney dialysis units in West Bank between May and August 2017. The rst unit is located at Palestine Medical Complex in Ramallah city and the second unit is located at An-Najah National University Hospital in Nablus city. The total number of kidney dialysis patients in these two units (n = 451) represented about 37% of all kidney dialysis patients (n = 1216) in West Bank in 2017 (20). HD patients were eligible to participate in the study if they were aged 18 years or over and were on HD for at least three months at recruitment. KTRs were eligible to participate in the study if they were aged 18 years or older and had functioning kidney transplant for at least on year at recruitment. All eligible patients were invited to participate in the study while attending their scheduled HD sessions or follow-up appointments at nephrology clinics during the study period. One of the researchers (RJ) met all eligible patients and invited them to participate in the study after explaining the purpose of the study and the nature of participation. All patients who signed written informed consents were then interviewed by the same researcher and completed a self-reported questionnaire in Arabic. The questionnaire collected data on sociodemographic variables and HRQOL.

Assessment of HRQOL
We assessed participants' perceived HRQOL using the 36-Item Short Form Health Survey (SF-36) (21).
The SF-36 self-report survey is a widely used instrument to assess the perceived HRQOL, which has been validated in several languages including Arabic (22,23)and in patients with different health conditions including HD and kidney transplantation (24)(25)(26). The SF-36 assesses different domains of HRQOL using eight subscales (physical functioning, role-physical, bodily pain, general health, vitality, social functioning, role-emotional, and mental health) and two component summary measures are derived from the eight subscales (physical component summary (PCS) and mental component summary (MCS)). The PCS represents a summary of physical functioning, role-physical, bodily pain, and general health, where the MCS represents a summary of vitality, social functioning, role-emotional and mental health subscales. The number of questions pertaining to each subscale range from two for bodily pain and social functioning to ten for physical functioning, and the number of responses to each question ranges from two options (yes, no) to six-point Likert scale (none, very mild, mild, moderate, severe, and very severe). Each response option is numerically coded and then converted into a score of 0 to 100. A mean score for each subscale and the two component summary measures is computed, with higher mean scores indicating better perceived HRQOL (21).

Statistical analysis
We used frequencies with percentages to summarize the sociodemographic variables. We used chisquare tests to examine for any statistically signi cant differences between the KTRs and HD patients in sociodemographic characteristics. The renal replacement therapy type (Kidney transplantation and hemodialysis) and sociodemographic characteristics were modeled as predictors of mean scores of SF-36 subscales, PCS, and MCS. We used multivariable linear regression to assess the relationship between sociodemographic variables with PCS and MCS scores among KTRs and HD patients separately.
Similarly, we used multivariable linear regression to assess differences in HRQOL scores between KTRs and HD patients after adjusting for sociodemographic characteristics. The HRQOL scores for all subdomains, PCS and MCS were not normally distributed. Therefore, we used bootstrap sampling and estimation method, with 1000 repetitions, in all regression analyses. All inferential statistical tests were two-sided. A P-value less than 0.05 was considered statistically signi cant. All analyses were performed using IBM SPSS Statistics computer program (version 26.0, IBM Corp).

Results
Three hundred and seventy-two patients participated in the study (100 KTRs (26.9%) and 272 HD patients (73.1%)). The sociodemographic characteristics of participants are summarized in Table 1. The sociodemographic characteristics differed signi cantly between the two groups for all variables except for residential area (city, village, or refugee camp). The KTRs were younger than HD patients; 36% of KTRs aged 18-29 years as compared to 6.3% HD patients. About 75% and 58.1% of KTRs and HD patients were men, respectively. Additionally, higher proportion of KTRs (72%) obtained secondary school education or higher than HD patients (39.3%).
Unadjusted differences in perceived HRQOL scores between KTRs and HD patients As shown in Fig. 1, KTRs had signi cantly higher HRQOL scores in all SF-36 subscales (except physical functioning), PCS, and MCS than the HD patients. The higher HRQOL scores among KTRs than HD patients ranged from 22.9 points for social functioning to 46.8 points for role-physical. Surprisingly, KTRs had signi cantly lower mean physical functioning score than HD patients by 41.8 points. Distributions of physical functioning scores among KTRs and HD are shown in Fig. 2. Table 1 Sociodemographic characteristics of kidney transplant recipients and hemodialysis patients Sociodemographic factors associated with perceived HRQOL scores among KTRs and HD patients The detailed associations between sociodemographic variables and HRQOL scores at SF-36 subscale level are not the focus of the current study. Therefore, we report here only the adjusted associations between sociodemographic variables with PCS and MCS scores in KTRs and HD patients (detailed data on associations between the sociodemographic variables and HRQOL scores at SF-36 subscale level are available from the authors upon reasonable request). As shown in Table 2  In addition, HD patients with higher than secondary school education level had increased MCS score by 11.6 points than HD patients with primary education level (95% CI 1.9, 20.7). We found no statistically signi cant differences in PCS or MCS scores according to gender or residential area among either KTRs or HD patients (  (Table 3).

Discussion
The aim of this study was to compare the perceived HRQOL in KTRs and HD patients. The current study has shown that KTRs were more likely to be younger, men, and with higher education level than HD patients. Younger KRTs had signi cantly lower MCS score than older KTRs. Conversely, younger HD patients had signi cantly higher MCS score than older HD patients. There were no signi cant differences in PCS scores by age group in either KTRs or HD patients. After controlling for sociodemographic variables, we found clinically signi cant differences in perceived HRQOL between KTRs and HD patients. KTRs had signi cantly higher perceived HRQOL scores in role-physical, bodily pain, general health, vitality, social functioning, role-emotional, mental health, physical component summary, and mental component summary than HD patients. These observed differences ranged between 15.5 points in social functioning and 34.3 points in general health. However, contrary to our expectations, KTRs had signi cantly lower perceived HRQOL in physical functioning than HD patients by 25 points. The magnitude of observed differences in HRQOL scores between KTRs and HD patients in the current study are considered clinically important differences based on the cut-off points for the minimal clinically important difference (3-5 points) for any health condition (27), 6-11 points for patients with stage ve chronic kidney disease (28), and 15 points for patients with heart disease (29). Our observed differences between KTRs and HD patients in age, gender, and education level are consistent with those of other studies and suggest that KRTs are more likely to be younger, men, and have higher education level (30,31). One explanation of the observed higher proportion of younger individuals among KTRs than HD patients is that younger patients with end stage kidney disease are less likely to have comorbid conditions than older patients, which makes younger patients better candidates for kidney transplantation (32). Our nding of high proportion of men than women among KTRs and HD patients could be explained by the higher prevalence of non-communicable disease risk factors and the faster progression of end stage kidney disease in men more than women (33,34). A plausible explanation for our nding of higher education level (a proxy measure for socioeconomic) among KTRs than HD patients is that socioeconomic status advantaged patients may have higher expectations for health and preference for kidney transplantation (35). Additionally, individuals with higher socioeconomic status are more likely to have lower prevalence of comorbid conditions and other contraindications for kidney transplantation, such as smoking and obesity (36). This explanation also applies to our nding of higher physical and mental health summary scores among KTRs and HD patients. The lower mental health summary score among younger KTRs found in the current study could be explained by their potential higher levels of worry about allograft survival, employment, and other economic factors (37,38).
The ndings of our study are consistent with the ndings of previous studies suggesting that kidney transplantation is associated with better physical and mental HRQOL scores than other renal replacement therapies (25,39,40). In contrast to these ndings, however, a systematic review with meta-regression, found no clinically meaningful differences between KTRs and HD patients in four dimensions of HRQOL (role physical, bodily pain, vitality, and mental health) after adjusting for age and diabetes. However, the ndings of meta-regression analyses should be interpreted with caution because any observed differences aggregated across studies may not necessarily apply to individual patients within single studies (ecological fallacy). The explanations for the contradictory ndings between our study and the aforementioned systematic review are largely unclear. One plausible explanation is that long-term use of immunosuppression medications by KTRs is commonly associated with important side effects, such as signi cant weight gain, recurrent infection, diabetes, hypertension, cardiovascular disease, and poor selfperception of physical appearance (13,14). For example, a cohort study of nondiabetic KTRs (n = 25,837) found that 17.7% and 12.3% of KTRs who were prescribed steroid-containing and steroid-free immunosuppressive regimens developed new onset of diabetes within three years post transplantation, respectively (41). This suggests that initial improvements in some aspects of HRQOL among KTRs may decline over time. This plausible explanation is supported by the ndings of previous cohort studies examining HRQOL among KTRs (17,37,42). For instance, a ve-year prospective cohort study of 110 KTRs and HD patients found no clinically important differences between KTRs and HD patients in ve dimensions of HRQOL (physical functioning, bodily pain, role-emotional, mental health, and MCS scores) by the end of follow up (42).
The present study found that KTRs had lower perceived HRQOL in physical functioning by 25 points than HD patients, which is a novel nding that, to our knowledge, has not previously been described. The exact mechanisms underlying this nding are not very clear. This is because physical functioning is determined by several factors including physical activity, physical tness level, and other known barriers to physical activity faced by KTRs, such as fear of physical activity and allograft rejection, immunosuppression side effects, new-onset of medical comorbidities, obesity, socioeconomic factors, and inadequate self-care strategies and clinician guidance on physical exercise (43,44). One potential explanation for our nding is that determinants and barriers of physical functioning might be more prevalent, or have greater negative impact, among KTRs in our sample as compared to other KTRs in other regions, which needs to be further explored. For example, one recent study showed that about 46% of KTRs in Palestine were unemployed (30), whereas only 6% of German KTRs were reported to be unemployed (40). Another potential explanation is that perceived HRQOL measures do not account for individual variations in expectations and actual experiences of healthcare outcomes, which may be driven by diverse factors, such as culture, spirituality, socioeconomic status, personality, and other sociodemographic factors (45). This explanation suggests that KTRs may under report their HRQOL if their high expectations for health and recovery are not met completely. However, we think that this explanation is not plausible because we would expect KTRs to under report their HRQOL in other subscales of physical and mental health subscales as well, which we did not observe. Therefore, we think that the observed lower HRQOL in physical functioning score among KTRs compared with HD patients is highly likely to be explained by the aforementioned barriers to physical functioning among KTRs and inadequate vocational and physical rehabilitation after kidney transplantation (40,44). For example, a qualitative study among KTRs showed that about 80% and 40% of KTRs reported having a sedentary lifestyle and receiving little clinical guidance on physical exercise after transplantation, respectively (44).
The present study included a relatively large number of KTRs and HD patients from two kidney units representing 37% of patients on renal replacement therapy in the West Bank (20). The sociodemographic characteristics of participants in our study are very similar to those of other studies from Palestine and other regions (30,31). In addition, we accounted for key predictors of HRQOL using multivariable regression analysis. Additionally, the Palestinian population is highly homogeneous in terms of ethnicity, culture, spirituality, and physical environment. This enhances the validity of our study, and therefore our ndings are highly likely to be generalizable to KTRs and HD patients in Palestine. However, our ndings may not generalize to other populations and regions with different healthcare system, culture, religious beliefs, ethnicity, and other socioeconomic factors in uencing HRQOL.
Several limitations need to be noted regarding the present study. First, we used a cross-sectional design to address the aim of the study, and therefore, the temporality and direction of observed differences in perceived HRQOL scores between KTRs and HD patients, presumably attributed to successful kidney transplantation, cannot be established with great con dence. Nonetheless, our ndings, taken together, are largely consistent with the ndings of other prospective cohort studies (17,46). Second, we collected no information on some factors which adversely affect the HRQOL among patients with end stage renal disease, such as comorbid conditions and biochemical variables, as creatinine (47). However, these factors are more prevalent among HD patients, thus lack of adjustment for those variables is not expected to attenuate the observed clinically important differences in HRQOL values between KRTs and HD patients. Third, self-reported perceived HRQOL is considered a subjective indicator and may be in uenced by individual expectations of health and recovery, which may underestimate or overestimate actual healthcare outcomes.
Despite the marked improvements in various aspects of HRQOL attributed to successful kidney transplantation, healthcare professionals managing patients on renal replacement therapy should be aware that KTRs have low perceived HRQOL in physical functioning, which should be addressed and monitored closely. There is some evidence from a systematic review that poor physical capacity among KTRs is associated with poor HRQOL and increased risk of hospital admission rates, length of hospital stay, and mortality (48). For instance, the ndings of a cohort study of 10,875 KTRs suggest that a modest improvement in physical functioning may decrease mortality rate by 11% among KRTs with low levels of physical functioning (49). In addition, a recent systematic review found that exercise therapy for KTRs is associated with signi cant improvements in HRQOL and exercise tolerance, and may improve renal allograft function (estimated glomerular ltration rate) (48). More recently, the Japanese Society of Renal Rehabilitation has published a guideline recommending exercise therapy for KTRs (50 Availability of data and materials: The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request.