Posted: 11/27/2011; Nat Rev Nephrol. 2011;7(11):611-612. © 2011 Nature Publishing Group
Abstract and Introduction
Abstract
Frequent nocturnal hemodialysis has emerged as a promising option for patients, with some studies demonstrating that it improves clinical outcomes. However, the negative results of a recent randomized trial have cast some doubt on whether nocturnal home hemodialysis has any advantages over conventional hemodialysis.Introduction
The Frequent Hemodialysis Network (FHN) has recently published the results of their trial comparing frequent nocturnal hemodialysis with conventional hemodialysis.[1] This trial has been reported as 'negative' as the investigators were unable to report a positive benefit of nocturnal hemodialysis (six sessions per week) over conventional hemodialysis (three times per week) according to their prespecified composite end point of death, left ventricular hypertrophy and physical well-being. These results are disappointing as enthusiasm for nocturnal hemodialysis schedules has been growing in several centers around the world. Marked anecdotal improvements in patient well-being have been reported, and most clinicians are struck by the apparent benefits when patients are switched to this modality.[2] However, very little data from controlled trials have been published to support these observations.The investigators of the FHN trial initially set out to enroll 250 patients, although recruitment difficulties meant that only 87 patients were randomly assigned to conventional or nocturnal hemodialysis. At 12 months, left ventricular mass data were available for only 76 patients. Only three deaths occurred during the follow-up period. Patients receiving nocturnal hemodialysis had an adjusted mean reduction in left ventricular mass of 10.9 g (95% CI −23.7 to +1.8 g; P = 0.095) with no appreciable difference in the physical health composite score (both groups demonstrated improvement over the 12 months). Although the change in left ventricular mass was not significant, the authors reflect that changes in left ventricular mass of a similar size have been associated with improvements in mortality of up to 50% in other studies. Significant improvements were seen in two clinically important prespecified outcomes—hypertension and hyperphosphatemia—among patients receiving nocturnal dialysis, which are factors that could be expected to impact on mortality in the long term.
There is little doubt that the trial was underpowered. By enrolling 90 patients there was an 80% chance of detecting a significant 19.6 g reduction in left ventricular mass and a 7.4 point improvement in the physical health composite score. For a 10 g difference in left ventricular mass, 275 patients would have been required to obtain 80% power. It is worth noting that the left ventricular mass effect size was not markedly dissimilar to that in the FHN frequent (short daily) trial,[3] or the nocturnal hemodialysis trial carried out by Culleton and colleagues, which is the only other randomized, controlled trial to compare nocturnal hemodialysis with conventional hemodialysis.[4] Both of these trials reached significance for the outcome of change in left ventricular mass. More patients or a longer follow-up might have resulted in a positive outcome for the FHN trial, although this possibility remains speculation.
The power of the study is not the only issue to consider. The amount of dialysis delivered is also somewhat problematic. Firstly, the conventional arm received about 45 min more dialysis per session than the US average (256 min compared with the usual 213 min),[5] whereas the nocturnal group received 30 h per week (as a session duration minimum of 6 h was stipulated), which is less than the more commonly used 8 h sessions for this modality (providing >40 h per week). Secondly, only 72.7% of patients completed the required number of dialysis sessions in the nocturnal arm. These factors meant that there was less difference in delivered dialysis hours than might have been expected between the conventional and nocturnal arms, with less likelihood for differential effects to become apparent. A high level of noncompliance with the regimen also raises the possibility of noncompliance with other aspects of the treatment, such as volume control, phosphate control and medication usage. Finally, higher weekly average weight gains were seen in the nocturnal group than in the conventional group, which is a commonly reported phenomenon explained by improved well-being and better appetite. These volume changes may have impacted on volume loading, left ventricular dilatation and subsequently left ventricular mass.[6]
Despite the weight gains, the investigators reported improved blood pressure control, which has been widely reported previously for nocturnal hemodialysis cohorts. The ideal blood pressure for patients on hemodialysis is not well defined. Whether normal blood pressure targets used in the general population are associated with the best outcomes in patients on hemodialysis is disputed. Observational, but short-term, data suggest the best survival outcomes are associated with systolic blood pressure readings of 140–160 mmHg.[7] However, longer-term data are more suggestive that blood pressure targets in the general population are more appropriate for patients on dialysis. Phosphate control is significantly associated with mortality in large observational studies of patients on hemodialysis, again with little data from randomized, controlled studies to confirm this observation.[8] Most groups reporting nocturnal hemodialysis patient cohorts have commented on the marked improvement in phosphate control with this modality, with most patients not requiring phosphate binders and some even requiring phosphate supplementation. However, perhaps as seen in the FHN nocturnal trial, translating this potential benefit into improved survival in a randomized, controlled trial setting has not yet been achieved.
Can we say that nocturnal hemodialysis at least does no harm? The main concern from this FHN trial was the trend to more vascular access complications in the nocturnal arm. Again, this result did not reach statistical significance, but was it clinically significant? Importantly, 45% of patients used catheters to dialyze at home, a statistic that most home dialysis units would find unusually high. Furthermore, 44% of the vascular access events occurred in patients with catheters, which reduces the number of events that units might expect to see in more typical patients with permanent access. The use of buttonholing was nearly twice as common in the nocturnal group, which is a practice identified by some as being associated with a significantly increased rate of access infection and complications.[9] Avoidance of buttonholing in the home setting might therefore be appropriate.
Finally, how much benefit is simply a result of having dialysis at home? This point was raised by the investigators, particularly in relation to the physical health composite score. Patients who receive home hemodialysis tend to be more enthusiastic about participating in their own care and are empowered to do so and are a self-selected group. Although large observational studies have repeatedly demonstrated improved survival for patients on home hemodialysis, eliminating the biases associated with these analyses is difficult.[10] Interestingly, a recent ANZDATA analysis also showed no benefit of extended-hours dialysis (at home) compared with conventional home hemodialysis.[10]
Although the FHN trial by Rocco et al. is technically a negative trial, it is unlikely to deter nocturnal hemodialysis enthusiasts from their path. The trial may be dismissed as underpowered, which is possibly the correct interpretation, but we should be encouraged that home hemodialysis is a safe and effective treatment. Looking to the longer term, we should also be encouraged by the improved blood pressure and phosphate control. Whether we will ever see a large enough randomized, controlled trial of nocturnal versus conventional hemodialysis to answer the question of mortality as an outcome is doubtful.
More specifics on this article can be found here:
Medscape Article Details
No comments:
Post a Comment