View complete transcript
[00:00:44] Dr. Jussi Sipilä: Thank you for the invitation, and it's nice to hear that you're interested in our research.
[00:00:50] Dr. Eduardo Fernandez: Congratulations for the article, I think it's very interesting. And I like the opening sentence, One dies with Parkinson's disease, not from it. It is [00:01:00] something that we have probably heard many times, but it is not entirely true, is it?
[00:01:05] Dr. Jussi Sipilä: No, unfortunately, it doesn't seem to be, at least, although I have to stress out from the beginning that we just can't be really sure yet. There are conflicting results on this, still.
[00:01:17] Dr. Eduardo Fernandez: We will discuss a bit more about the literature, but as you said, there is multiple studies evaluating mortality in Parkinson's disease. So what motivated you to carry out this study?
[00:01:28] Dr. Jussi Sipilä: Well I like the opening sentence to pick that up from a publication in the MD journal. And while reflecting on it, I thought this is something I've been saying to my patients also quite frequently. And I'm not quite sure if it's true. I've seen many patients who have quite dismal disease causes and die early.
And on the other hand we just did a paper on, the prognosis of PD patients in [00:02:00] southwestern Finland, in which we did not have any, any general population controls. And we were left quite unsure if those would have shown us something that we didn't see since we saw that the prognosis had become better in men, but was that just because maybe men have become more healthy in the general population or something like that, we weren't sure.
[00:02:24] Dr. Eduardo Fernandez: Before we dive into the results of the study, as you discussed in the article for epidemiological studies is very important. On one hand, the sample size, but also the quality of the data and I would like you to just expand a bit more on where do you obtain the data and the quality of them.
[00:02:44] Dr. Jussi Sipilä: Yes, I agree. It's crucial to have valid data for epidemiology, and this is the main problem with epidemiological studies. Balancing on the sample size and the data it's never optimal, I think, or at [00:03:00] least not frequently, and I must say I'm a bit jealous to the colleagues who study diseases such as MS in the Nordics, who have great disease specific registries, something we would also need, but they're hard to come by.
So we ascertain our patients from the medication reimbursement registries of the National Health and Social Security Agency. And every patient with PD or Parkinsonism that might be helped by levodopa or other medications is entitled to this reimbursement, and it can be obtained by one certificate signed by the, treating neurologist and explaining how he or she has come to the diagnosis what's the base of it and then sends it to the institution where an expert doctor checks it, and if it checks out, the reimbursement is granted and it's for life. And I don't [00:04:00] think I can recall any patients who haven't had it granted or haven't seeked for it. So, you can say that every patient with Parkinson's when that's who's been diagnosed by a neurologist and doesn't have things like vascular Parkinsonism or medication induced Parkinsonism will get the entitlement.
On the other hand, we are not quite sure if it's 100 percent accurate and there are preliminary data suggesting it's positive predictive value might be around 80%, so quite close to something we get usually in clinical practice. And we tried to make it more accurate in these data by obtaining also the death certificate data and the causes of death for the patients and excluding all patients who had died of a typical parkinsonism or similar disorders that might be mistaken for PD. Then we obtained also [00:05:00] population controls from statistics, Finland for those people, we only had the data on their date of birth, date of death, and sex. So no clinical data for them, but we must check that they don't have the reimbursement. So they wouldn't be Parkinson's and patients.
Then we also have data from the National Registry from Healthcare which well, actually in this case, it was the main source of the PD data. We just used the SSIF data to check if It correlated, and we also use the health care data to obtain data on comorbidities to calculate the Charleston comorbidity index and such to see what factors would affect the prognosis.
And we also had data from the National Cancer Registry, which we used for that. So that's our data sources.
[00:05:52] Dr. Eduardo Fernandez: So correct me if I'm wrong, but so you managed to gather a cohort of PD patients covering the whole of [00:06:00] Finland with the diagnosis confirmed by a neurologist as with an accuracy as as much as we can achieve in clinical practice, but also with some filters to remove any other secondary causes of Parkinsonism.
And you also combine that with a population of healthy controls, again, that was not bias that was community based.
Exactly.
Excellent. So let's get into the results that I think they are quite interesting. So, overall, it seems like there is an increase in mortality rates for people with Parkinson's disease.
But one of the most surprising data, at least to me, was that this increased mortality is present very early in the disease course.
[00:06:41] Dr. Jussi Sipilä: Yes, we were a bit surprised by that too. On the previous data, it was something we didn't expect and thought that maybe we might see some increased mortality. Some perhaps 2, 3, 4 years from diagnosis and actually we, when we first [00:07:00] look at the results I mistook it as such because I had this preconception but fortunately, our senior author who is a cardiologist and a statistical magician corrected me.
And yeah it was present from one year on and kept increasing for the first 12 years, after which it plateaued at 29%. Excess mortality compared to controls.
[00:07:23] Dr. Eduardo Fernandez: And so this increased mortality is also higher in people with sort of young onset Parkinson's disease. Is that correct?
[00:07:30] Dr. Jussi Sipilä: Yeah, it seems that early onset patients had the highest hazard ratio for excess death in this data, unfortunately.
[00:07:39] Dr. Eduardo Fernandez: Are these findings something that we have found before in the literature, or is this something like new?
[00:07:45] Dr. Jussi Sipilä: Well, There are conflicting reports on this. There are reports that have even recently found no difference in mortality between patients and controls. And on the other hand, there are reports [00:08:00] for example, from Israel recently and also from the UK, which have reported excess mortality. But the hazard ratio differs between these reports. So the Israeli results were quite close to ours, whereas the excess mortality in the UK paper was only slightly elevated. And on the other hand, recent data from Estonia and China reported no difference.
There's also a bit of a problem with the definition of early onset disease since some epidemiological studies have age cutoffs in place and, may report only persons from age 40 years onwards, or even 50 years onwards, you can't really see any EOPD patients. So it's bit difficult to compare the results between studies.
There are reports, at least from Korea, that EOPD patients have a really increased excess mortality. And also there was a [00:09:00] paper from Norway too.
[00:09:01] Dr. Eduardo Fernandez: One of the sort of promising results from your study is that the survival or the mortality has decreased over the years of follow up from the study, particularly men, as you mentioned before. What do you think this is due to, is this that we are better at managing Parkinson's disease or there is any other factors related to general health that have improved mortality?
What's your interpretation of these results?
[00:09:25] Dr. Jussi Sipilä: And we don't really know. This is something that has baffled us since our last paper with the southwestern Finland's patients. And we really thought it just improved prognosis in men in general. It has come closer to the life expectancy of women in Finland, although we are still far behind, but in this paper, we also saw that the excess mortality in PD patients has declined in men and was actually in the most recent time period in this paper, it was almost exactly the same between men and women, during the first seven years after [00:10:00] diagnosis. On the other hand, I'm not quite sure if we took data for the next five years, say, and repeated this, if we would see the same trend, because in the middle time period, in this paper, it had actually gone a bit up from the baseline. So this might be something we see in epidemiological studies that it fluctuates somewhat.
So I'm not quite hundred percent sure that it's really happening and it's a robust trend. We need more data.
[00:10:31] Dr. Eduardo Fernandez: Okay, as you mentioned with the studies in the literature, there is a lot of variations in the results and most of the times, this can be partly explained by Differences in the methodology, the data, the populations and so on. But we know that Parkinson's disease is a very heterogeneous condition, clinically, pathophysiologically. So do you think about the different results in the population? Do you think it's mainly to do with the methodology [00:11:00] of the studies? Or do you think that also variations in them? Pathophysiology and survival and prognosis in Parkinson's disease can play a role?
[00:11:08] Dr. Jussi Sipilä: I think it's both. We know, for example, that the prevalence of GBA1 mutations differs by population and that's associated with a more malignant disease usually, and on the other hand, we don't really have all LRRK patients here in Finland and so on and so on. So there's so many different Parkinson's diseases that I think it's really major source of differences between these results in different countries.
[00:11:40] Dr. Eduardo Fernandez: I hope we will be able to just differentiate a bit of these Parkinson's subtypes and then try to elucidate a bit better the prognosis between them. Just the final thing, as you said, it's very difficult to make conclusions on comparing studies due to the differences in methodology, but what are your overall conclusions? Or ask in a different [00:12:00] way. What should we say to our patients in clinic when they ask about the mortality?
[00:12:04] Dr. Jussi Sipilä: Yeah, my take home for my clinical practice was that unfortunately you can't really say that you die with Parkinson's disease, not from it, at least in general. There clearly are patients as we saw it plateaus after 12 years. So there are patients beyond that who live over 12 years and their prognosis appears to be quite the same.
As those of the general population, but during the first 12 years in our data, we saw that there are patients who have a more malignant disease and they die early from it. The underlying cause of death was actually during the first 7 years, it was much more commonly PD in those patients compared to those who died later. So it appears it was really the parkinsonism and PD that was the main problem with them.
So I usually say that there is a possibility that this may be lethal. And, we [00:13:00] unfortunately at the outset, we don't know who this concerns. So time will tell, but I also usually tell my patients that these are the best years usually you have because you're are closest to normal. So try to enjoy these and not to think too much about the future at this moment.
[00:13:19] Dr. Eduardo Fernandez: Let's see the optimistic side of things. Yeah. Well, thank you very much for your time today and congratulations for your article.
[00:13:26] Dr. Jussi Sipilä: Thank you. It was a nice conversation.
[00:13:29] Dr. Eduardo Fernandez: Thank you very much to the listeners and I would encourage them to read the full article published in Movement Disorders Clinical Practice. Bye bye for now. [00:14:00]