Cubital tunnel or not?

Pain in the elbow and shoulder is not really typical of ulnar neuropathy at the elbow so with reportedly normal NCS on three occasions I would be starting to think of other possibilities but it's impossible to completely exclude the diagnosis. An alternative way of checking the nerve at the elbow is with ultrasound scanning and that sometimes shows an abnormality when the nerve conduction studies do not but the expertise to do the ultrasound is not widely available. Dr J Bland

thanks for your comment. you seem an expert on the subject. do  you also suffer from this problem?

If I'm 'expert' in anything it's carpal tunnel syndrome, but the expertise mainly lies in just having systematically collected the data on every patient I have ever seen with it over 25 years or so and being able to analyse that with the computer. I do test quite a lot of people with ulnar nerve problems at the the elbow - the cubital tunnel is only one of several possible sites of compromise of the ulnar nerve at or near the elbow and it often impossible to distinguish them without detailed testing so I prefer the broader term 'UNE - Ulnar Neuropathy at the Elbow'. Oddly I do indeed get intermittent ulnar nerve symptoms myself, mostly during the night when I occasionally wake having spent too long with my left hand under my head and loss of sensation in a clearly ulnar nerve distribution - but I never get elbow (or any other) pain. Although my website is mainly concerned with carpal tunnel syndrome it does have a page devoted to UNE because the two problems are quite often confused with each other. Dr J Bland

It's philosophically impossible to prove a negative, so no it is not possible to wholly exclude the possibility, but it is pretty unlikely with your combination of symptoms and test results. Dr J Bland

Clinical neurophysiologist - I'm easily looked up on the web. Dr J Bland

We don't really know because there is no fully accurate reference standard to compare against. Less work has been done on UNE than CTS so the best I can offer is an educated guess based on statistics. Our conception of what is 'normal' is based on the range found in 95% of the asymptomatic population so the specificity (the proportion of normal individuals with a normal test results) should, by definition, be 95%. However 'normal ranges' are often derived from sets of subjects that are not really representative of patient populations so this has to be taken with a pinch of salt and 80-90% is probably more realistic in patient populations. How good a test is at excluding something really depends on the sensitivity (the proportion of subjects with a condition who have a positive test result). This is even harder to know accurately in the absence of an accurate reference but for UNE it may well be as low as 50-70% I think. I suspect that there are quite a few people (myself included in the left arm) who get intermittent ulnar nerve compression related to posture sufficient to cause symptoms, but do not do any measurable damage to the nerve - ie it recovers fully between insults - so that there is nothing to find on testing. A Bayesian approach to this helps considerably in clinical practice. Bayes theorem allows you to start from a prior estimate of how likely a condition is and then modify your estimate based on a test result, rather than just relying on the result of a test as an absolute arbiter - the prior probability can be as simple as the population prevalence or can be a clinical assessment of the symptoms. A further step beyond this diagnostically is not to take a test results as a binary YES/NO answer but to take account of how abnormal it is - in CTS measurements 5 standard deviations above the healthy population mean are more likely to indicate CTS than measurements 3 standard deviations above the mean, and measurements 2 standard deviations below the mean, indicating very 'good' nerve function probably make CTS more unlikely than a measurement close to the mean - but in very few diagnostic situations is such a sophisticated mathematical approach actually taken. JB

thank you for detailed comments and explanations. no doctor found the cause and I am miserable

and there is delay in my projects.

Is it not possible to make a video chat with you for a possible diagnosis?

Unfortunately I'm not likely to be able to arrive at a diagnosis by video consultation. My particular expertise lies in the use of 'hands on' tools - EMG/NCS and ultrasound imaging to supplement what I am able to glean from history and examination so I really need to see people in person - and like every other doctor I often don't know the answer anyway, even with the aid of all the extra tools. Sorry. JB