Kidney function as an underestimated factor for reduced health related quality of life in patients with Fabry disease

Since June 2001, an increasing number of patients with genetically proven Fabry disease from all over the country are overseen at our hospital. In regular visits, physical examination and detailed evaluation of medical history and medications are performed by a trained nurse and a physician. Furthermore, patients are enrolled in an international observational cohort study if they provided written informed consent [10]. The local Ethics Committee of the University of Würzburg approved the study. For the current analyses, patients who had their first visit until August 2009 were included, if they were older than 16 years with available data on HRQoL within three months before or after the first visit. We analyzed data from the first visit at our center, regardless of time since the diagnosis of Fabry disease or whether or not they were receiving ERT at the time of enrollment.

HRQoL was assessed by the Medical Outcomes Study SF-36 questionnaire [11] at baseline and was repeated after six months [2]. The SF-36 is a validated instrument to describe HRQoL in a broad spectrum of diseases and is widely used in clinical practice and clinical research [12, 13]. Patients respond to questions about activities of daily life and physical and social well being, resulting in two summary scales (physical component summary [PCS] and mental component summary [MCS]) and eight scales describing various physical and mental aspects (physical and social functioning, physical and emotional role, bodily pain, general and mental health, vitality). Each dimension is displayed on a scale ranging from score 0 to 100, representing impaired vs. improved health [14].

Glomerular filtration rate (eGFR) was estimated according to the simplified Modification of Diet in Renal Disease formula [15]. Stages of CKD were classified as eGFR ≥ 60 ml/min/1.73 m², eGFR < 60 ml/min/1.73 m² and renal replacement therapy (RRT), i.e., dialysis or kidney transplantation (KTx). History of arrhythmia, angina pectoris, significant cardiac procedure without coronary stent implantation, or transient ischemic attack (TIA) was considered as minor vascular event, whereas history of myocardial infarction, coronary stent implantation, coronary artery bypass, stroke or implantable cardioverter-defibrillator (ICD) were described as major vascular event. Pain was assessed by using the visual analog scale (VAS), ranging from no pain (score 0) to maximal pain (score 10). Patient’s pain therapy (i.e. frequent use of analgetics) was not altered.

Analyses were performed using SAS 9.1 (SAS Institute, Cary, NC). Characteristics of study participants and the different scales of HRQoL were compared across CKD stages using ANOVA, Kruskal-Wallis test, χ²- test, and Fisher’s exact test, as appropriate. Factors associated with the various dimensions of HRQoL were examined using linear regression analysis. To assure a linear relationship with the dependent variable, the logarithmic form was used for “time since Fabry diagnosis”. In multivariate analyses, the following variables were included in the model; age, gender, kidney function, vascular event, pain and pain-therapy. We investigated the relationship between pain and HRQoL according to their interaction with pain-therapy. Notably, the variable “enzyme replacement therapy, initiated within three months following the first visit” was not considered as a risk factor for HRQoL since impaired physical and mental status of the patient represents the cause of initiating ERT; i.e., the variable can be seen as a measure of “sickness” and does not describe the effect of ERT.

The robustness of the results was further evaluated by (1) including only four variables in the multivariate model, to minimize the risk of observing potential associations by chance due to a large number of candidate variables (“overfitting”). Thus, the least significant variable was ignored; (2) including patients in whom data of the SF-36 was available within nine months prior or following the first visit (three additional patients); (3) considering kidney transplant recipients (n = 2) not in the category of RRT, but according to eGFR within the respective CKD category; (4) not differentiating between minor and major CV events, but including CV events as a single category.

Ninety-six patients older than 16 years of fifty-two families were treated at our center between June 2001 and August 2009. The mean age of the cohort was 40 yrs, 47% were male, and the diagnosis of Fabry disease was made a median time of nine months prior to the first visit. All patients were naïve to ERT. Characteristics of study participants stratified by kidney function are displayed in Table 1. Most of the patients, in particular most women, had preserved kidney function, whereas patients with impaired kidney function were more likely to be male. All patients receiving RRT were of male gender (n = 9), of whom two underwent kidney transplantation and seven received dialysis. Minor vascular events were also more common in patients with CKD, as was a history of major vascular events. Furthermore, patients with impaired kidney function, in particular those receiving RRT, were more likely to be initiated on ERT within three months following the study visit. Written informed consent was obtained for participation and publication of clinical and individual data.

Data on HRQoL were missing in six patients, and in three additional patients the SF-36 was not completed within three months following the first visit, thus leaving eighty-six patients for further analyses (Table 2). Of note, all patients with missing HRQoL had preserved kidney function. Chronic kidney disease was generally associated with reduced HRQoL. More specifically, lower scores on dimensions of mental health, as represented by social functioning, role emotional, mental health, and the MCS, were observed in patients dependent on RRT. In contrast, physical aspects, such as physical functioning, role physical, bodily pain, and the PCS, were affected even in patients with moderately impaired kidney function, i.e., eGFR < 60 ml/min.

Similar to the described findings, impaired kidney function, particularly the need of RRT, was associated with markedly reduced HRQoL across all scales in univariate linear regression analysis (Tables 3a and b). Furthermore, male gender, pain and pain-therapy were significantly and meaningfully associated with lower scores on the SF-36 scales, except general and mental health. Age and a history of vascular events, neither minor nor major events, were not found to be related to HRQoL. This is shown in detail in Additional file 1. In multivariate modeling, impaired kidney function and pain emerged as the main factors associated with lower scores on the SF 36, in particular on dimensions of physical aspects of life, such as physical functioning, role physical, bodily pain, general health and the PCS, while controlling for gender, vascular event and pain-therapy. In contrast, only the need for RRT was found to be associated with reduced HRQoL in mental/social parameter dimensions, when adjusting for the described variables. By including the respective interaction term, we tested whether the relationship of pain and HRQoL was altered by pain-therapy. Being on pain-therapy, but still suffering from pain was related to even lower scores on the role physical-scale (p = 0.001), while this finding approached significance in bodily pain (p = 0.06), role emotional (p = 0.08) and the PCS (p = 0.07). The variables included in the multivariate models explained a median of 35% (min.-max; 10-52%) of the variability in the respective SF-36 scale.

We tested the robustness of our results in a first step by including only four explanatory variables in the multivariate linear regression model. Excluding vascular event, which represented the least significant variable, did not result in significant changes of our findings, neither on the magnitude of the direction, nor on the significance levels or the respective variables. Similarly, by including three additional patients in whom the SF-36 was completed within nine months prior or following the study visit, no significant changes in the estimates of the variables in the multivariate model were found; however, the 95% CI became narrower, resulting in pain-therapy being significant on the p < 0.05 level in the physical functioning scale. Moreover, when we analyzed the two KTx-patients in the respective eGFR category, i.e., excluding them from the RRT category, we did not detect major differences in the findings, except in the variable describing kidney function. Generally, the associations of RRT with the respective scales strengthened including narrower 95% CI in physical aspects, such as physical functioning, role physical and emotional, and the PCS. No changes were observed in mental parameters. Details are shown in Additional file 1. Finally, treating CV-events in a single category did also not alter the results: neither became the variable CV-event significant, nor did the associations of the remaining determinants in the model substantially change. Additional file 1 shows this in more detail.

We are aware of several limitations of our findings. As in any observational study, no relationship can be determined sufficiently, i.e., the direction of an association in cross-sectional analyses cannot be explored. It would also be desirable to investigate the changes in HRQoL in FD patients and how these may be associated with changes in renal function, optimized pain therapy, the occurrence of vascular events, but also with the potential effects of ERT. Moreover, the differentiation of pain severity according to receiving or not receiving therapy for neurogenic pain is diffuse, because it may depend on differences in perception of pain but also on differences in patient care. Besides that, an influence of pain peaks, which are not detected by the use of the VAS at one single point of time, on HRQoL is not excluded.