Cross-Sectional Comparison of Quality of Life and Illness Intrusiveness in Patients Who Are Treated with Nocturnal Home Hemodialysis versus Peritoneal Dialysis

Concise Methods

This protocol was approved by the Research Ethics Board of the Toronto General Hospital–University Health Network. All patients who had ESRD and were undergoing NHD or PD were deemed eligible for this study when they remained on either NHD or PD for a minimum of 3 mo. All patients were required to have a working knowledge of English. None had any acute illness or was hospitalized. Written informed consent was obtained from each patient. Surveys were mailed to our patient cohorts during February to June 2006. At 3 wk, a follow-up telephone call was placed to ascertain whether patients had received their survey. A repeat mailing was sent after our telephone reminder. Survey results were included in the study when they were returned within 4 mo of mailing. All dialysis-related blood tests were obtained via standardized protocols, and their results were stored in an electronic database. All responses were anonymous.

Results

The study population consisted of 93 patients with ESRD (36 NHD and 57 PD). Of the eligible patients with ESRD, 69% (72% NHD and 67% PD) responded. Their baseline characteristics are summarized in Table 1.

Overall, the majority of NHD and PD patients were male (67 [NHD] versus 55% [PD]; P = 0.28) and had similar comorbid status as assessed by the modified Charlson Index (1.14 ± 0.25 [NHD] versus 1.82 ± 0.33 [PD]; P = 0.14). NHD patients were younger (49 ± 2 versus 61 ± 2 yr; P < 0.001), although they had a similar renal replacement vintage as PD patients (10.8 ± 1.7 [NHD] versus 7.6 ± 1.0 yr [PD]; P = 0.1). There was a trend toward increased frequency of renal transplants in NHD patients (31 [NHD] versus 14% [PD]; P = 0.08). NHD patients also tended to be more educated than PD patients. There were no differences in the frequency of patients who lived alone (25 [NHD] versus 18% [PD]; P = 0.41).

Compared with NHD, plasma urea and creatinine concentrations were higher in PD patients. NHD patients had higher hemoglobin concentration than the PD population (124 ± 2 [NHD] versus 117 ± 2 g/L [NHD]; P = 0.026). Plasma albumin and calcium were higher in the NHD cohort in comparison with the PD patients (39 ± 2 [NHD] versus 37 ± 2 g/L [PD]; P < 0.001) and (2.41 ± 0.03 [NHD] versus 2.27 ± 0.30 mmol/L [PD]; P = 0.002), respectively. Plasma phosphate was lower (1.11 ± 0.06 [NHD] versus 1.63 ± 0.07 mmol/L [PD]; P < 0.001) in the NHD patients (Table 2).

Overall, the KDCS was similar between the NHD and PD groups (75.1 ± 3.5 [NHD] versus 68.2 ± 3.9 [PD]; P = 0.20). Physical component summary (55.0 ± 2.3 [NHD] versus 52.3 ± 1.8 [PD]; P = 0.35) and mental component summary (61.6 ± 4.7 [NHD] versus 60.0 ± 3.5 [PD]; P = 0.77) did not show any difference between the two study cohorts.

Among the subcategories of the KDCS, the perceived amount of social support was less in the NHD group compared with the PD patients (66.0 ± 5.2 [NHD] versus 79.0 ± 2.8 [PD]; P = 0.027). There was also a trend toward less burden of kidney disease, an emotional reaction to the effect of kidney disease, in the PD population (36.0 ± 4.4 [NHD] versus 47.0 ± 3.8 [PD]; P = 0.092). There was a trend toward better sexual function in the NHD group (82.0 ± 5.4 [NHD] versus 62.0 ± 9.0 [PD]; P = 0.07). The remainder of the KCDS components yielded little difference between the two study populations and are summarized in Table 3.

Similarly, there was no difference in the BDI scores between the two patient groups (11 ± 1.7 [NHD] versus 12 ± 1.4 [PD]; P = 0.52). Total illness intrusiveness score was similar between the two patient groups (17.0 ± 1.3 [NHD] versus 15.0 ± 0.9 [PD]; P = 0.40). Of the various subcategories that composed the total illness intrusiveness score, all domains were similar between the two groups (Table 4).

For ascertainment of the influence of potential confounders on perceived level of social support, age, gender, level of education, history of renal transplant, comorbid illness index, BDI, and plasma albumin levels were applied to multivariate regression modeling. After adjustment for these variables, the perceived level of social support remained lower in the NHD group versus the PD patients (64.9 ± 5.2 [NHD] versus 76.0 ± 2.8 [PD]; P = 0.04). We did not discern additional differences among our patients by stratifying according to residual renal function (data not shown).

Discussion

Patients’ perceived level of wellness is arguably the most important health outcome for those who undergo renal replacement therapy. In addition, there is an emerging body of literature that associates poor QOL with nutritional status, dialysis adherence, hospitalizations, and survival of patients with ESRD (20). Although QOL is an important aspect of the assessment of a patient with ESRD, the balance between intrusion and efficacy of dialysis therapy must be weighed carefully. This study was the first to compare QOL, depressive symptoms, and illness intrusiveness of both kidney disease and treatment modalities in patients who were undergoing NHD versus PD. Our results suggest that well patients who undergo home dialysis (NHD and PD) have similar overall QOL, illness intrusiveness scores, and BDI values. Less social support was perceived by NHD patients relative to the PD cohort. Our data do not support the notion that augmented solute clearance alone determines the perceived health-related QOL of an individual. For patients who have the option to dialyze at home, this study suggests that NHD is not perceived as a more intrusive treatment while also demonstrating that patients who are on PD have similar subjective symptomatic control of their kidney disease.

Relationships among QOL, morbidity, and dialysis dosage was first ascertained by the National Cooperative Dialysis Study (21); however, subsequent randomized, controlled trials failed to show any dramatic increase in QOL associated with an augmentation in solute clearance. Indeed, the Hemodialysis (HEMO) study did not yield any “clinically meaningful benefits” in QOL of either the dosage or the flux intervention arm, despite a clear separation in fractional urea and β2-microglobulin clearances (22). Similarly, the investigators from the Adequacy of Peritoneal Dialysis in Mexico study found no long-term beneficial effect in QOL by increasing peritoneal small solute clearance in their study population (23). In contrast, enthusiastic reports of improved QOL with intensive hemodialysis have been observed using before-and-after designs (24,25). In the context of the results of this study, how does one reconcile the lack of increased benefits in QOL for patients who undergo NHD compared with PD? First, the potential beneficial effect of NHD versus conventional in-center hemodialysis was not the principal focus of the study and could not be dismissed. Second, we used a cross-sectional study design to ascertain at one point in time the QOL in home dialysis patients; we did not compare the sustainability of the reported QOL among PD and NHD patients. To our knowledge, QOL tends to diminish over time in patients with ESRD (26,27), especially in PD patients with a decline in residual renal function (15). It is possible that NHD patients are able to maintain their QOL scores better than PD patients or vice versa. Finally, the ultimate derivation of QOL perceived by an individual is complex, so it is reasonable to suggest that augmentation of uremic solute alone should not be considered the principal determinant of QOL in the ESRD population.

Given that the human cost of chronic disease management must include an appropriate balance among illness intrusion, therapeutic efficacy, and QOL, it is important to acknowledge the similarity in illness intrusive score between PD and NHD patients. It has been widely supposed that NHD would be a more intrusive therapy compared with the simplicity of PD. It is therefore interesting to note that our results do not support this assumption. Rather, our data were consistent with previous literature indicating that there was little difference among patients who underwent dialysis at home (28). In contrast, conventional in-center hemodialysis patients have uniformly the worst illness intrusiveness score (28). Taken together with our data, it is tempting to propose that the quantification of illness intrusiveness in ESRD cannot be simply be a reflection of the complexity of the technology but rather the sum of patients’ coping strategies (29), social support (30,31), and illness behavior (32) in addition to the technical challenges of the medical intervention. In light of this hypothesis, it is important to comment that NHD patients exhibited a lesser degree of perceived social support in comparison with the PD cohort, perhaps reflecting the demand of intensive home hemodialysis. The determinants of perceived intrusion and control over one's illness requires further exploration, especially given the present momentum in allocating patients with ESRD toward home-based dialysis regimens.

Although perceived QOL is an important determinant of well-being of patients with ESRD, the actual therapeutic efficacy of intensive hemodialysis should not be minimized. In this respect, we were able to discern multiple biochemical differences between NHD and PD patients. The clinical benefits of higher albumin and hemoglobin levels in conjunction with normalization of phosphate are beyond the scope of this study; however, on the basis of published literature, each of these indicators has been robustly validated to provide survival benefits in dialysis patients (33–35). The clinical advantages of intensive hemodialysis continue to be of great interest among various investigating groups, which will require ultimate substantiation by randomized, controlled trials (36).

Conclusions

Choosing a dialysis modality is a difficult decision to make for most patients with ESRD. Our study showed that among well NHD and PD patients, similar QOL, depressive symptoms, and illness intrusiveness scores were observed. This study also highlights two important points: The complexity in determining subjectively derived parameters in a chronic disease population and that the increase in solute clearance alone is insufficient in determining QOL. Our study is limited by its cross-sectional observational nature. We also did not account for the impact of duration of home therapy alone on QOL and illness intrusiveness. To our knowledge, these are the only available data comparing two home dialysis modalities in the present era. Additional research is required to delineate the interactions among complex medical interventions (dialysis modality) and patients’ illness behavior, coping strategies, and social support so as to individualize best the delivery of optimal ESRD care in the home setting (37,38).

Disclosures

None.