Reevaluating the risk of DVT in people with ALS

Providers caring for people with amyotrophic lateral sclerosis (ALS) are all too familiar with the clinical scenario of a patient with substantial leg weakness reporting new leg edema. The provider is left to weigh the danger of deep vein thrombosis (DVT) against the far more common and essentially benign edema of venous stasis.¹ Furthermore, the incidence of sudden death in people with ALS has been estimated at over 6%, raising the possibility that fatal pulmonary embolism is a notable source of mortality.² For such a potentially important issue, the lack of data has been both surprising and frustrating.

In this issue of Neurology®, Gladman et al.³ present a well-described prospective observational study of a cohort of consecutive patients seen at the University of Toronto multidisciplinary ALS clinic. The investigators report an annual incidence of DVT of 11.2%, nearly 4 times that previously reported. In the subset of patients with leg-onset disease or leg weakness that impaired mobility, the incidence was more than 3 times higher, over 35%. DVT was associated with decreased leg function on the Revised ALS Functional Rating Scale (ALSFRS-R), decreased leg strength on examination, and reduced leg mobility by the lower extremity activity score.

For years, a few imperfect reports have provided the primary evidence guiding clinicians' thinking regarding the incidence of DVT in people with ALS, providing estimates of 2.7%–3.3% annually.^4,5 While most of these events appear to be in patients with leg weakness, these studies were retrospective and limited in their clinical characterization.

In the present prospective study, the investigators employed symptom review, physical examination, and Doppler ultrasound every 6 months for a year. Thus, whereas previous studies evaluated only symptomatic DVT, these higher estimates of DVT include asymptomatic DVT. In so doing, the authors contribute perhaps the most comprehensive set of evaluations available and lend credence to the clinical suspicion that DVT is common among people with ALS.

The authors include a sophisticated discussion that appropriately contextualizes the results, avoids overstating the evidence, and notes that the proximate cause of the DVTs was not accounted for by the study; for example, falls and dehydration were not well-documented. Even so, the results must be evaluated in the appropriate context to allow clinicians to appreciate the nuances of the study and direct further study in this area.

The amount of censored data is large—approximately 40% of the cohort died or was lost to follow-up over the course of the study. Loss to follow-up is a major challenge for all ALS observational studies, and the authors acknowledge this shortcoming, but it can be difficult to completely correct statistically for such data censoring. Ultimately, the number of observed events was small—just 4 DVTs were identified. This casts some degree of uncertainty over the accuracy of the risk estimates, because with so few observed events, random variation could lead to overestimation of DVT incidence.

Of interest, the average disease duration since symptom onset was 40 months at enrollment. Given that 50% of people with ALS die within 36 months of symptom onset and 90% die within 60 months,⁶ this has a few potential implications: (1) this population had very advanced disease or (2) this population had abnormally slowly progressive disease. In reviewing the baseline vital capacity, it appears that the pulmonary function of patients in the study was quite good at baseline (>75% of expected). In reviewing the ALSFRS-R, the function was somewhat lower at 30 (out of 48). The relatively long disease duration, relatively high pulmonary function, and middling ALSFRS-R score, combined with the data that a somewhat higher than expected 40% of participants had leg-onset disease, suggests that the participants tended to have slowly progressive disease affecting the limbs, primarily the legs. Given that in this report and previously published studies DVT was highly correlated with leg function, this might have biased the results toward a higher overall incidence of DVT in all patients with ALS, which could affect the generalizability of the results. At the same time, the data from the study should generalize well to the subpopulation of people with ALS who have leg weakness.

Overall, this is a very thorough and well-conducted study that should alert the ALS clinical and research community that DVT may be a major source of morbidity in people with ALS and is worth further study. Future studies should confirm the present findings and aim to delineate an appropriate role for secondary or primary screening and prevention of DVT in people with ALS. Such explorations will need to account not only for the high risk of DVT in people with ALS but also for the risks of anticoagulation given the high risk of falls in this population.

• © 2014 American Academy of Neurology