I assume this is a flexible PVC. I also assume that you're using a lab extruder and pilot set up for this tubing trial. Even if it's production set up, it appears that your tubing is poorly fused or under fused, and/or poorly mixed w/in the extrusion process used to manufacture it.
Please grant that it is difficult to make assessments based on the above photo and limited information about your formulation and your processing equipment, die set up, and extrusion conditions. Given those limitations, I think you have either a processing or material related problem. Here are my suggestions and opinions, for your consideration.
The tube appears to be filled w/ gels. I'm guessing that the outer surface is fairly well fused, but the inner surface is not. The inner layer appears to be a melt that is carrying unfused PVC particles along from the transition zone of the screw to the die. This could be compound (too slow a fusion curve) or machine (low shear/mixing and poor heat profile) related and is usually associated w/ the screw design.
Clear compounds are sometime difficult to process because they do not contain fillers and other additives as melt viscosity builders, thus the clear melt's low viscosity does not develop adequate shear during processing. What melts in the transition zone carrys un-melted particles of PVC resin along, but these particles experience little shear or mixing w/in subsequent extruder zones. Thus, the futility of adding heat to correct the problem. Adding heat just invits degradation and the extra heat lowers viscosity even further. Complicating mixing and shear.
One thing to try would be to grind up the existing tubing and try to run a second pass, using this regrind as your feed. I suggest this because your tube appears to be under fused or poorly fused. Adding heat to your first pass (powder feed to melt to tubing) may not achieve anything other than degradation. But, intuitively more heat and shear will improve fusion and tubing appearance. The practital way to do this is to change the screw profile, not just raise the heater zones. But, if your second pass, using the regrind from the first pass is improved, then you know you must increase shear and mixing during transition (screw design) or reformulate the compound to match your present extrusion equipment.
I'm a little puzzled that the gels/rough surface is/are concentrated on the inner surface of the tubing. But, i think this is the results of your screw and die design. The compound could certainly be optomized for better fusion and easier processing through reformulation. This would requiire some detail about the existing clear formulation and some knowledge about your other tubing formulations that can be successfully run on your existing equipment. Which I do not have.
I hope these comments are helpful,