The study you link to makes no attempt to control for recent reduction of salt intake because of an increased risk of cardiovascular disease. To put it plainly

The study you link to makes no attempt to control for recent reduction of salt intake because of an increased risk of cardiovascular disease. To put it plainly, people who consume abnormally low amounts of salt don't do it out of choice - nobody enjoys eating their food without salt. They do it because the doctor tells them that it's either that, or a heart attack. It makes sense of course that cardiovascular events will be more prevalent in this group than any other group. Studies like the one you link to control for pre-existing cardiovascular disease, but having CV disease doesn't mean you cut down on salt. The same methodological flaw has previously caused studies to report that very low alcohol consumption is better than moderate alcohol consumption for the heart, for example. I have a link somewhere if you want. Btw, the study you linked to used the data from two studies comparing two anti-hypertensive drugs to each other or to placebo, so we know that basically everyone in the study already had some kind of underlying disease. Additionally, the study data included a one-time measurement, which significantly reduces its statistical power, despite its large population (you gotta take multiple measurements to minimise sampling error).