The trend of hospitalization in frail home hemodialysis patients versus otherwise well in-center hemodialysis patients: Exploratory insights

INTRODUCTION

End-stage renal disease (ESRD) continues to pose a significant global health challenge, with millions of patients reliant on renal replacement therapies (RRT) such as hemodialysis (HD) for survival.^[1,2] Among the RRT modalities, HD is the predominant choice, delivered either as in-center HD (ICHD) or home HD (HHD), with modality selection often influenced by patient characteristics, healthcare infrastructure, and local practices.^[3,4] Despite ongoing improvements in dialysis technology and patient management, hospitalization rates remain alarmingly high in this population, largely due to the complexity of comorbidities and the physiologic impact of advanced kidney disease.^[5,6]

Hospitalization is a critical endpoint in HD patients, as it is closely linked to increased morbidity, mortality, and healthcare costs, as well as reduced quality of life.^[7] Recent studies confirm that individuals on maintenance HD are hospitalized more frequently and for longer durations than the general population, predominantly due to cardiovascular events, infections, and vascular access complications.^[8,9] Readmission and recurrent admissions are also common, imposing further strain on both patients and healthcare systems.^[10]

Dialysis modality has emerged as a significant determinant of both hospitalization risk and patient-centered outcomes. Recent evidence suggests that HHD can offer advantages over conventional ICHD, including greater scheduling flexibility, improved biochemical control, and enhanced patient autonomy.^[3,4,11] International studies have reported that HHD is associated with lower rates of hospitalization and improved survival, especially among well-selected, healthier, and more independent patients.^[11,12] However, these studies often exclude or underrepresent frail, highly comorbid individuals, limiting the generalizability of their findings.^[13]

Frailty is characterized by diminished physiological reserve and increased vulnerability to stressors. It affects a substantial proportion of the ESRD population and is a recognized predictor of adverse outcomes, including hospitalizations and mortality.^[14] Notably, frailty is associated with functional decline and decreased ability to live independently, making modality selection and care planning particularly challenging.^[14,15] In our center, HHD is specifically reserved for patients with the highest frailty scores, significant comorbid burden, and mobility limitations, contrasting with the selection criteria for HHD in most international studies, where patients tend to be less frail and more functionally independent.^[13,16]

This discrepancy raises an important clinical question: Does the observed reduction in hospitalization rates with HHD in international, healthier cohorts also extend to a more vulnerable, frail population when compared to demographically similar but less frail and more mobile ICHD patients?

To fill this knowledge gap, our exploratory study compared hospitalization trends, frequency, and etiology in a tertiary care setting between frail HHD patients and less frail, ambulatory ICHD patients. This comparison is necessary to ascertain whether the potential advantages of HHD apply to frail ESRD patients. This information will help guide clinical judgment, resource allocation, and future research priorities for this particularly vulnerable population.

MATERIAL AND METHODS

RESULTS

A total of 42 patients were included in the analysis, which were evenly split between HHD and ICHD modalities (n = 21 each). The mean age of the cohort was 66.1 years (SD = 13.4), with a range from 18.0 to 93.0 years. The sex distribution revealed a slight female predominance, with 54.8% of patients being female.

Table 1 shows the clinical and demographic characteristics of the two groups: ICHD and HHD. No statistically significant differences were found in the demographic or clinical characteristics between the two dialysis groups, including the vascular access used between the two groups. However, the HHD group had a significant association of patients with a CFS score of 6 (p = 0.0048), and assisted mobility (p < 0.001).

Figure 1 visualizes the relationship between HD modality, number of admissions after treatment, and gender. We observed 59 in-hospital admissions in the ICHD group, with all patients (n = 21) admitted to the hospital once or more, and 52.4% (n = 11) of patients admitted 3 or more times over a follow-up duration of 1 year. In the HHD group (n = 21), we observed 29 in-hospital admission encounters, with 13 patients in the group being admitted once or more than once; 9.5% (n = 2) of the patients were admitted three or more times, whereas 8 patients (38%) did not experience an in-hospital admission encounter.

Close

Figure 1: Frequency of hospitalization in association with gender and dialysis modality. HHD: Home hemodialysis , OHD: Outpatient hemodialysis, HD: Hemodialysis

Export to PPT

The most common reason for admission in the ICHD group was HD access malfunction (n = 18), whereas in HHD, it was infection (n = 12). Table 2 presents the comparative analysis of the frequency and etiology of admission encounters between the patient groups undergoing ICHD and HHD.

There was a statistically significant difference in the number of hospital admissions between the two patient groups (ICHD; n = 59 vs. HHD; n = 29) (p = 0.001). Among patients with no hospital admissions, all of them were found to be HHD (n = 8 out of total n = 21) compared with ICHD (n = 0), whereas those with 1–2 admissions had a roughly equal distribution between the two modalities (HHD: 52.4%, ICHD: 47.6%). Conversely, patients with three or more admissions were predominantly in the ICHD group (52.4%) rather than the HHD group (9.5%).

HD access malfunction was more common in the ICHD group (p = 0.015), with 90.5% of HHD patients having no arteriovenous fistula (AVF) malfunction compared with 52.4% of ICHD patients. Similarly, acute coronary syndrome (ACS) was encountered more frequently in the ICHD group (p = 0.048), with no admissions due to ACS in the HHD group.

In addition, although not statistically significant, trends were observed in other admission etiologies. For example, peripheral vascular disease (PVD) was more common among ICHD patients than among HHD patients (28.6% vs. 19.0%, p = 0.719), although this difference did not reach statistical significance. Figure 1 visualizes the relationships among HD modality, number of admissions, and gender.

Looking specifically at the data, 100% of cases of no admission cases were observed in the HHD group. Among males, 45% of patients had 1–2 readmissions in the ICHD group and 55% in the HHD group, whereas in the female group, this distribution was even (50% ICHD and 50% HHD). All 5 males with ≥3 readmissions were included in the ICHD group, whereas in the female group, this ratio was 1:3 (25% HHD and 75% ICHD). Associations between HD modality and number of admissions were found to be significant in both males and females (p = 0.02 and p = 0.03, respectively).

Table 3 describes the results of binary logistic regression analysis exploring admission etiologies with a possible association with HHD versus ICHD. Among the examined etiologies at admission, catheter-related bloodstream infection (CLABSI) as a diagnosis was more frequently encountered in ICHD than in HHD (estimate: −1.5507, p = 0.297).

Patients experiencing AVF malfunction presented higher log odds of belonging to ICHD than to HHD, with an estimate of 1.4842 (p = 0.250). This implies that patients with AVF malfunction are 4.41 times more likely to have ICHD than HHD compared with those without AVF malfunction. Although this association did not reach statistical significance, the substantial odds ratio suggests a potential influence of modality preference on AVF malfunction. For other clinical factors, such as chest infection and PVD, the log odds of belonging to ICHD over HHD were not significantly different (p = 0.214 and p = 0.603, respectively). Specifically, the odds ratio associated with chest infection was 0.1627, indicating a decrease in the odds of ICHD. However, none of these associations reached statistical significance.

Furthermore, factors such as chest pain for workup, volume overload, gastrointestinal complaints, hypotension, number of admissions, age, and sex were not significantly associated with the modality of dialysis.

DISCUSSION

In this study cohort, we noted significant differences in the frequency, etiology, and pattern of all-cause in-hospital admission between the two study groups, i.e., ICHD and HHD, over a follow-up period of 1 year, which can impact and reflect the quality of life, survival benefit, and cost burden. We noted that the patients who were receiving HHD had a high CFS score (plus 6), required mobility aids, and had a high burden of clinical comorbidities compared with those receiving ICHD (p ≤ 0.001). Another difference in our study is the frequency of HHD 3-4 sessions with a dialysis dose of 12-16 h per week compared with most international studies with HHD 5 sessions per week with a duration of 4-6 h per session.

Despite the above-mentioned HHD-related limitations and higher burden of frailty and clinical comorbidities, we noticed better outcomes in the HHD group over a follow-up period of 1 year. This is an unexpected paradox that is of significance in our study settings regarding better patient-associated outcomes in the HHD group. Although our HHD population is frail and has a lower weekly dialysis dose than other studies do, they are admitted less frequently, have fewer recurrent admissions, fewer cardiovascular events, and have fewer CLABSI and vascular access malfunctions than ICHD patients. In several international studies, including Bremer et al.^[18], a higher mean hospitalization rate for ICHD composed the majority of total admissions in comparison to HHD, peritoneal dialysis, and frequent HD. Moreover, those undergoing frequent home HD (5–6 sessions/week) had the lowest rates of hospital admissions.^[19] These findings are in keeping with our study findings, but there is a significant difference with our HHD populations regarding higher frailty scores, burden of comorbidities, assisted mobility, and lower HHD prescription and dose.

We noticed a lower rate of cardiovascular events needing inhospital care in the HHD group as compared to the ICHD group, which is a significant clinical outcome observed over a 1-year follow-up. The improvement noted with HHD is likely due to the prolonged duration of HD sessions with gentle ultrafiltration distributed over extended hours, thus avoiding significant hemodynamic changes and shorter inter-dialytic periods.^[20] There is a modest survival benefit demonstrated with HHD in comparison to ICHD across multiple studies.^[21,22]

We noted fewer incidences of CLABSI and AVF malfunction in patients receiving HHD than in the ICHD group. This contrasts with findings in international studies that AVF malfunction likely results from increased HD frequency rather than the modality of HD.^[23,24] The better outcomes in our study in the HHD population can be due to personalized care given during HHD by supervised nurses, while in international studies concerning HHD, there is self-cannulation/connection of AVF or Permcath with a machine by patients or their caregivers. This can make them more prone to vascular access complications, including infections and AVF malfunction, as compared to our study. As noted previously, personalized nursing can reduce the incidence of AVF complications and can address issues pertaining to AVF.^[25]

CONCLUSION

In this pilot study, frail HHD patients – despite a higher comorbidity burden and lower dialysis dose compared to previous cohorts – demonstrated significantly fewer and less recurrent hospitalizations, as well as reduced access-related and cardiovascular complications, than their otherwise healthier ICHD counterparts. This paradoxical finding underscores the potential benefits of HHD even in highly vulnerable populations. These exploratory results highlight the need for larger, prospective multicenter studies to further evaluate hospitalization patterns, healthcare utilization, and cost-effectiveness across dialysis modalities, thereby reforming future clinical practice and policy.