A 12-year Follow-up in Sporadic Inclusion Body Myositis

An End Stage With Major Disabilities

Fieke M. Cox; Maarten J. Titulaer; Jacob K. Sont; Axel R. Wintzen; Jan J. G. M. Verschuuren; Umesh A. Badrising


Brain. 2011;134(11):3167-3175. 

In This Article


Patient Characteristics

The original, complete cohort comprised 64 Dutch patients with sporadic IBM (43 males). Classification of patients in the various diagnostic categories is given in Supplementary Table 1. Patients, if grouped according to having definite or probable sporadic IBM, did not differ in age at onset, age and duration of symptoms at first visit, functional grading scales and muscle testing scores at first visit. Therefore, they are presented as a single group. At follow-up, 46 patients had died, one patient was lost to follow-up and 17 patients were still alive (12 males). Of these, one male declined to participate in this study and one female was unable to give informed consent due to concomitant disease. This left 15 surviving patients who were evaluated for the second time. The median time to follow-up was 12 years (interquartile range 11–13). Baseline characteristics of the cohorts of deceased or surviving patients and of the complete cohort are summarized in Table 1. Not unexpectedly, the mean age of the patients in the surviving cohort was significantly lower at baseline, with better scores on strength and functional scales. The age at onset was lower in the surviving patients compared with the deceased cohort.

Muscle Strength

Mean decline in muscle strength by manual muscle testing was 3.5 ± 1.6% per year and 28.8 ± 11.9% over a 10-year time period (P <0.0005). The mean Medical Research Council score at baseline for the surviving 15 patients was 285 ± 19 points, at follow-up this was 184 ± 52 points. The correlation between these two scores was 0.39.

No correlation was found between age or duration of symptoms at baseline and the rate of decline on manual muscle testing. No associations were found between the rate of weakness progression and gender, presenting symptom, HLA-B8, -DR3 or -DR53.

As shown in Fig. 1, the rate of decline on manual muscle testing was similar for most patients, except for three (Patients 3, 4 and 10). One patient deteriorated more rapidly on manual muscle testing (9.5% per year); this was, however, partly due to a null score for his left leg because of an amputation for vascular obstruction. He was excluded from the strength analysis in order to correct for this rapid decline due to the amputation, although the rate of decline in his three remaining extremities was still above average (7.9% per year on manual muscle testing). Two other patients showed a moderate decline compared with the others (0.5 and 0.9% per year), without obvious reasons for it.

Figure 1.

Sum scores of manual muscle testing (MMT) and quantitative muscle testing (QMT) and functional grading scales (Barthel index, Rivermead mobility index, Brooke's functional grading scale) for the surviving patients at baseline and follow-up. All figures show a decline in strength and function in time. QMT was not performed in one patient at baseline and in 2 patients at follow-up.

For quantitative muscle testing, the mean rates of strength decline were 5.4% ± 3.5 per year and 39.4% ± 21.8 per 10 years (P <0.0005). Mean score at baseline for the surviving 15 patients was 2996 ± 913 N, at follow-up this was 1473 ± 753 N. The correlation between these two scores was 0.74. No factors could be identified that modulated the quantitative muscle testing. At follow-up, quantitative muscle testing was not performed in two patients. One refused to undergo measurements with the hand-held myometer and in the other case there was a technical problem with the myometer. A second investigation was planned, but the patient became seriously ill leading to prolonged hospital admission.

The manual and quantitative muscle testing scores on both visits showed good correlation (r = 0.73 and 0.92, respectively, P <0.001).

The pattern of selective muscle involvement was still present after a mean disease duration of 20 years (Fig. 2a). Finger flexors, quadriceps muscles and all the muscles of the lower leg were most severely affected, especially compared with the relatively spared neck muscles, shoulder abductors and hip abductors. In the hand, the opponens pollicis and adductor pollicis muscles were still relatively spared, allowing patients to maintain some grip function, despite severely weakened flexors and extensors of the other fingers. Asymmetry of muscle weakness was present in all patients at follow-up.

Figure 2.

(a) Severity of muscle weakness according to Medical Research Council (MRC) scores of distinct muscle groups for the surviving patients. The median disease duration was 20 years. The finger flexors, quadriceps and all the muscles of the lower legs are affected the most, compared to relatively spared neck muscles, shoulder abductors and hip abductors. In the hand, the opponens pollicis and adductor pollicis muscles are relatively spared as well. (b) Median strength decline according to MRC scores of distinct muscle groups for the surviving patients at a median disease duration of 20 years. The largest decline in strength is seen in the finger flexors, iliopsoas and quadriceps muscles and all muscles of the lower legs.

Median decline on manual muscle testing for the quadriceps over the follow-up period was 2.5 points on the Medical Research Council scale (interquartile range 2–4) compared with a median decline of the hipabductors of 1 point (interquartile range 0–2). The deep finger flexors declined with a median of 3.5 points (interquartile range 1–4), versus a median decline of 1 point (interquartile range 0–1) of the opponens pollicis (Fig. 2b).

In general, sporadic IBM progresses to be a more distal myopathy (Supplementary Fig. 1). The lower leg is weakened most severely, followed by the forearm and the upper leg, the upper arm and lastly the neck muscles. The rate of progression showed the same pattern.

Functional Status

At their initial visit, the functional status of the patients was considered to be good, as shown by almost maximal scores on all three functional grading scales [Barthel index: 20 (19–20), Rivermead mobility index: 13 (12–14), Brooke's functional grading scale: 4 (4–6)]. However, at the second visit, all patients scored significantly worse on all functional grading scales as compared with baseline [Barthel index: 11 (6–16), Rivermead mobility index: 4 (1–9), Brooke's functional grading scale: 13 (9–14); P = 0.001; Fig. 1]. These scores show that 40% of patients are completely or severely dependent (Barthel index <10), and 20% of patients are moderately dependent (Barthel index 10–15). No differences in the degree of deterioration were found related to age, duration of symptoms, gender, HLA-B8, -DR3 or -DR53 positivity. To illustrate the impact of this decline in functional grading scale scores, two patients with a decline comparable to the mean decline found in all patients are discussed in the Supplementary Material.

At baseline, a good correlation was only found between the Rivermead mobility index and manual and quantitative muscle testing and between the Barthel index and quantitative muscle testing. At follow-up, all functional grading scales correlated well with the sum scores of the manual and quantitative muscle testing.

Dysphagia was present in 12 (80%) of the surviving patients as determined by the use of the questionnaire. Three of them did not have dysphagia at the first visit and swallowing function had deteriorated in four patients. In 20% of patients, dysphagia was obstruction related (e.g. food becoming stuck in throat, repetitive swallowing), in 7% aspiration related (choking) and mixed in 53%. Only one patient had undergone a cricopharyngeal myotomy resulting in a good, but temporary effect (5 years) on the discomforting obstructive symptoms.

At baseline, after a median disease duration of 11 years, 63 patients were living at home and one patient lived in a nursing home. In the surviving cohort of 15 patients, after a mean disease duration of 20 years, three patients were living in a nursing home and 12 at home with adaptations (stair lift, no thresholds, stand-up chairs). Nearly all patients who were still living at home required considerable help with daily activities from their partners or other caregivers.

At baseline, 47 patients (73%) used a device to assist mobility, including nine (14%) who used a wheelchair. At follow-up, all 15 surviving patients used a wheelchair to some extent. The mean time from the first symptom to using a walking stick was 11 ± 5 years and the mean time to the first use of a wheelchair was 16 ± 4 years. Seven patients (47%) were completely wheelchair-bound. The median time to complete wheelchair dependency was 24 years. None of the patients was able to climb stairs any longer.

Life Expectancy

Forty-six of the 64 patients died during follow-up. The median age at death was 81 years (80 years for men, 84 years for women). In the Netherlands, life expectancy adjusted for gender and age at onset is 79 years (77 years for men, 83 years for women; Statistics Netherlands, 2010; Fig. 3), and so life expectancy was not shortened in our group of patients with sporadic IBM.

Figure 3.

Kaplan-Meier curve showing a comparable survival between sIBM patients and an age- and sex-matched Dutch general population. The curve for the general Dutch population is adjusted for life expectancy for each individual sIBM patient based on the age of onset and gender.

The only factor associated with life expectancy in sporadic IBM was gender, which is in accordance with the general Dutch population life expectancy, with women living longer than men. The presenting symptoms (quadriceps or non-quadriceps), HLA-B8, -DR3 or -DR53 positivity were not associated with a different life expectancy.

The causes of death in our patient group are summarized in Table 2 and compared with those in the general Dutch population, in the age category of 80–85 years, in 2004, which includes the median year of death in our patient group (Statistics Netherlands, 2010).

When compared with an age-matched general Dutch population, the cause of death in sporadic IBM patients was significantly more often a disorder of the respiratory system, specifically pneumonia; this being the cause of death in 13 patients. In five of these, it was related to aspiration.

Cachexia, defined as severe wasting with loss of weight and muscle mass, was also a significantly more common cause of death in patients with sporadic IBM than in the general Dutch population, whereas cancer was less frequent. Patients tended to die less often of cardiovascular diseases; however, after a Bonferroni correction, statistical significance could not be substantiated.

Frequencies of causes of death in the general Dutch population in 2004 in the age category ranging from 70 to 90 years are comparable to the frequencies found in the age category of 80–85 years.

Unfortunately, in eight patients the cause of death could not be further clarified

Euthanasia and Continuous Deep Sedation

In six patients (13%), end-of-life care interventions were applied; three patients (6.5%) requested euthanasia because of unbearable suffering and severe loss of quality of life due to extensive weakness. The ages of death were 68, 76 and 84 years, respectively. Continuous deep sedation was used in three cases (6.5%), two of whom had severe disabling swallowing dysfunction and were dehydrated and cachectic. As they were completely bedridden, they chose not to feed themselves by artificial means and requested continuous deep sedation. The third patient had developed pneumonia with respiratory insufficiency, and chose not to undergo further treatment for this infection; continuous deep sedation was applied. All three patients died within a day of the sedation being initiated. The ages were 73, 79 and 79 years, respectively. All six patients were completely bedridden at that time.