A prospective study of quality of life in ALS patients treated with noninvasive ventilation

Patients and methods.

QL was measured using the SF-36 questionnaire.3 When completed this provides scores for eight QL domains. Four are functional status measures: “Physical Functioning,” “Role Physical,” “Role Emotional,” and “Social Functioning”; three are measures of well being: “Bodily Pain,” “Vitality,” and “Mental Health”; and the last—“General Health”—is a measure of overall health. Final scores range from 0 to 100%, higher scores indicating a more positive QL status.

Generalized disability was assessed using the Amyotrophic Lateral Sclerosis Functional Rating Scale (ALSFRS),4 and sleepiness was assessed using the Epworth Sleepiness Scale.5

Results.

During the study period, all consecutive patients offered NIPPV consented to treatment and agreed to enter the QL study. The NIPPV cohort comprised 16 patients (one woman), the control group 11 patients (two women). The age distributions for the two groups were comparable: the mean age in the NIPPV group was 61.3 (SD, 6.8) years, and in the control group 61.2 (7.6) years. At baseline, three patients in the NIPPV group had significant bulbar involvement compared to none in the control group.

Only one patient (from the NIPPV group) withdrew from the study, moving too far from the hospital to complete assessments. All other patients completed QL assessments until within 3 months of death (mean time from last QL assessment to death, 76 [47] days). During the course of the study, 11 NIPPV patients died, all of whom used ventilation until death (mean survival on NIPPV, 257 [87] days). Five NIPPV patients were still alive at the time of writing (mean survival to date, 371 [129] days). By visit 4, two patients were using the ventilator for periods during the day, and by visit 5 for at least 4 hours in the day, increasing to over 20 hours during periods of chest infection. By visit 6, one patient was using NIPPV for over 16 hours a day. No control patient had died at the time of writing.

There were no significant differences between the time from baseline to subsequent visits for the NIPPV and control groups. Mean (SD) time from baseline until visit 1 was 37 (7) days for the NIPPV and 39 (9) days for the control group; until visit 2, 125 (29) and 136 (19) days; until visit 3, 194 (32) and 187 (35) days; and until visit 4, 286 (27) and 263 (37) days. Visits 5 and 6 (NIPPV group only) took place 351 (23) and 455 (36) days after baseline.

At baseline and visit 4, there was no significant difference between the ALSFRS scores of the two groups. Scores for both groups were low, indicating considerable levels of generalized disability: mean baseline ALSFRS score was 26 (SD 2) in the NIPPV group and 27 (6) in the control group.6 The scores fell in both groups at similar rates, indicating disease progression. Results of VC and ELBG Pco₂ measurements are given in table 1. The control group had no evidence of respiratory muscle weakness or hypoventilation, at baseline and visit 4. Baseline ELBG results were abnormal in the NIPPV group, suggestive of hypoventilation, after ventilation these values normalized. There was a significant difference between the Epworth Sleepiness Scores of the two groups at baseline: the NIPPV group had a mean (SD) score of 9.25 (5) and the control group 5 (3). After ventilation, mean Epworth Sleepiness Scores fell to 4 (3) (p < 0.0001).

Table 2 shows the mean “Vitality” domain SF-36 scores for the NIPPV and control groups. The number of patients contributing to the score at each visit is shown, as is the mean score for a normal age-matched population. The control group score was not significantly different from normal, and did not change significantly during the study. Baseline scores in the NIPPV group were significantly lower than normal, but improved significantly at visit 1, with a mean change of 19%.

In the remaining domains of the SF-36 questionnaire there was no significant difference in the baseline scores between the NIPPV and control group. “Bodily Pain” and “Mental Health” scores (well-being domains) for both groups were as high as in the normal population, at baseline and at subsequent visits.

Both groups had significantly reduced baseline scores compared with normal subjects in the functional status and overall health measure (the “Physical Functioning,” “Role Emotional,” “Social Functioning,” “Role Physical,” and “General Health”) domains, which, in view of the patients’ generalized disability, was to be expected. Scores in these domains fell progressively, consistent with disease progression, and the rates of decline were comparable between the groups.

Discussion.

Using the SF-36 questionnaire, we have shown that the QL of patients with ALS with sleep-disordered breathing improves significantly with NIPPV, as measured by the “Vitality” domain. This improvement was maintained despite increasing generalized disability and progressive respiratory muscle weakness. Patients on NIPPV completed questionnaires to within 3 months of death and in the majority, the final scores were higher than baseline scores. Improved QL scores were recorded at 15 months after ventilation in the longest surviving members of the cohort. In the SF-36 domains that reflect well being, patients with ALS on NIPPV had scores as high as a normal age- and sex-matched population. As expected in a disease causing generalized muscle weakness, scores in domains reflecting overall functional status were low, but comparison with a group of patients with ALS not requiring ventilation shows that use of noninvasive ventilation does not of itself cause reduced QL scores. Although these findings suggest that noninvasive ventilation is beneficial in ALS, there are a number of issues that need to be considered in the interpretation of the results.

First, there was a predominance of male subjects in both groups. This does not reflect the authors’ patient population, nor indicate a gender bias in the provision of ventilation. All consecutively ventilated patients during the period of the study were recruited, and the gender imbalance arose by chance. We do not believe that this would invalidate the results. Second, randomization of patients may have improved the study design. NIPPV is an established treatment for ventilatory failure, and as most patients were referred with daytime hypercapnia we did not think it ethical to withhold treatment from such patients. A control group consisting of patients with ALS with comparable generalized disability but without evidence of sleep-disordered breathing was studied. Opponents of the use of NIPPV in ALS point to the progressive nature of the disease, suggesting that any improvement in QL obtained from treatment of sleep-disordered breathing will be negated by the increasing disability afforded by prolonged survival. Many domains in the SF-36 are functional measures. As we expected scores in these domains to fall as disability increased, without a control group this may have been interpreted as indicating that ventilation reduced QL. Scores fell equally in the control subjects, making this unlikely. A fall in QL of the control subjects could be caused by the development of untreated sleep-disordered breathing. One study found significant sleep disruption in patients with ALS with a mean VC of 60.7%, whereas sleep was normal in patients with mean VC 86% of predicted.6 Evidence of sleep-disordered breathing was found in three of 18 asymptomatic patients with ALS,7 but all three affected patients had significant bulbar involvement and experienced obstructive events. We conclude that sleep disturbance is unlikely when respiratory muscle strength is relatively well preserved in patients without significant bulbar involvement, as was the case in our control group.

We conclude that treatment of sleep-disordered breathing with noninvasive ventilation enhances QL as measured by the “Vitality” domain of the SF-36 questionnaire, with improvement maintained until death despite increasing disability, and we support the use of this technique in patients with ALS.