The 'triad' - CTS, Trigger Finger and Dupuytren's contracture
This group of three conditions are often seen together in the same patient, either at the same time or one after the other, they are:
Trigger digit (Stenosing Tenosynovitis) – in this disorder thickening and irregularity of the tendons which connect the forearm muscle to the fingers causes them to get stuck in the pulleys which hold the tendons against the bones at each joint. This can cause pain and locking of the fingers in a flexed position. When they are forcibly straightened with the other hand they ‘click’ straight again – hence the name ‘trigger’ digit.
Dupuytren’s Contracture – this condition is caused by thickening of fibrous tissue in the palm called the palmar fascia. As it thickens it contracts and this too can cause the fingers to be pulled into a flexed position. This tends to affect the little and especially the ring finger more.
Carpal Tunnel Syndrome – if you are reading this site you know all about CTS!
All three are probably commoner in diabetes and if you read the rest of this page you will see that diabetes is a recurring theme throughout. The evidence for diabetes as one obvious common predisposing factor which might account for the co-occurrence of these disorders is compelling but they also seem to occur together quite often even in the non-diabetic population.
There have now been several studies demonstrating the association between trigger digit and CTS. As early as 1966 George Phalen found 33 cases of trigger digit in 166 patients having carpal tunnel surgery and Hombal in 1970 found trigger digits in 29 of 132 non-rheumatoid hands with CTS.
Looking at it the other way around, in a recent study of 211 patients with trigger digits, but without other relevant disease such as rheumatoid arthritis, 91 (43%) also had CTS (Kumar 2009). It also seems to be the case that, the more fingers are affected by trigger digit the more likely you are to develop CTS as well (Wessel 2012). These authors looked for diagnoses of CTS in the period 2 years before and three years after surgery for trigger digit. In 160 patients with triggering of a single digit 26(16%) had CTS but in 140 patients with triggering of multiple digits 58 (41%) had CTS. The increased risk of CTS appears to increase fairly linearly with the number of trigger digits in the CTS hand:
- 0 digit = 13% CTS
- 1 digits = 20% CTS
- 2 digits = 36% CTS
- 3 digits = 44% CTS
- 4 digits = 80% CTS
These authors also carried out a multivariate statistical analysis on their data and not only demonstrated these associations between trigger digits and CTS but also found diabetes to be an independent risk factor for CTS in this population. There was a marked tendency for the two problems to affect the same hand with only 3 of 84 patients having trigger digit in one hand and CTS only in the other hand. There was however no clear tendency for CTS to either precede or follow the trigger digit.
Other studies have found similar results:
- 101/875 (11.5%) CTS cases required trigger finger surgery within 3 years (Harada 2005)
- 61/265 (23%) CTS hands developed trigger finger within 12 months of CTS diagnosis, not all surgically treated (Hayashi 2005)
- 11% of female patients with CTS also had trigger digits (Loong 1977)
- In a group of 101 patients referred to a hand clinic for either CTS or trigger digit, 67 were mainly referred with CTS, 41 with trigger digit but 66(61%) of the patient group actually had both disorders. A very high percentage of patients referred to this clinic were diabetic (57 = 53%) but the incidence of diabetes was the same in the trigger finger only, CTS only and both diseases groups (Rottgers 2009)
It has been suggested that patients who develop one trigger digit have an increased risk of developing more subsequently (Wessel 2012 referencing Goshtasby 2010 and Kumar 2009) and this would seem inherently obvious, but is hard to pin down with hard figures.
Some of these authors have given thought to whether CTS (or its treatment) might cause trigger finger or vice versa. It is certainly plausible that trigger digits might lead to CTS – an inflammatory process of the tendons and their sheaths occurring in the carpal tunnel as well as at the finger pulleys might increase the carpal tunnel pressure and swelling of the hand resulting from disuse might also contribute to CTS. One might also imagine that carpal tunnel surgery might disturb the environment and function of the tendons enough to result in trigger digit. A new study in the Journal of Hand Surgery (Lee 2014) explores this idea by looking at the extent to which 'bowstringing' of the tendons occurs after carpal tunnel surgery in patients who do, and do not, develop trigger digits after surgery. These authors did find that more palmar excusrion of the flexor superficialis tendons after surgery was associated with the development of post-operative trigger digit but unfortunately they did not report in the paper whether their ultrasound measurment of tendon excursion was different in the two groups before surgery and technically what they have demonstrated is an association, not causation. It remains possible that the presence of trigger digit itself may in some way be causing the increased bowstringing or that a third factor leads to both increased bowstringing after surgery and the development of trigger digit. It should also be noted that blaming the increased risk of trigger digit on the surgery fails to explain the increased incidence of trigger digits in CTS patients BEFORE surgery.
One group looked for risk factors for the development of trigger finger after carpal tunnel surgery in 792 hands, 6.3% of which subsequently developed trigger digits (Goshtasby 2010). They concluded that the presence of osteoarthritis and undergoing endoscopic rather than open surgery increased the risk. The study by Wessel however suggests that there is no predictable order to the occurrence of CTS and trigger digit and it seems unlikely that one disorder directly causes the other. It is much more probable that whatever personal factors predispose a patient to develop CTS also predispose to trigger digits. Sometimes there is an obvious linking cause - both conditions have been observed together in patients with mucopolysaccharidoses (MacDougal 1977, van Heest 1998) as well as in diabetes – but in idiopathic CTS it is likely that the same genetic predisposition underlies CTS and the associated conditions rather than an identifiable disease.
Another curious study from Israel looked at the nerve conduction studies in a group of 62 non-diabetic patients with trigger digits who had no symptoms of CTS on direct enquiry and compared them with 13 healthy controls (Garti 2001). 39 of the trigger digit patients (63%) had prolonged distal motor latencies though the criterion for prolonged DML was very strict (>3.6 msec abnormal)
In patients referred to the Canterbury carpal tunnel clinic 6% of 4899 patients without nerve conduction evidence of CTS reported a previous history of trigger digit while 12% of 9128 patients with confirmed CTS did so – double the incidence and a fairly incontrovertible (p < 0.00001) finding, even when based on patient self-report of the diagnosis of trigger digit.
The association between CTS and Dupuytren’s Contracture is less well established. In Phalen’s review of 654 CTS cases 8 patients had co-existent Dupuytren’s. Out of 904 patients treated for CTS at the University of Louisville between 1962 and 1972, 42 (4.6%) were recorded as having Dupuytren’s disease in the ipsilateral hand (Nissenbaum 1980). Most of these patients seem to have developed Dupuytren’s before CTS but in 5 cases the Dupuytren’s appeared after surgery for the CTS. These authors felt that the outcome of combined surgery was rather poor in their cases, with 13 of 15 patients who had surgery performed for both disorders at the same time experiencing post operative complications , and that surgery for both conditions should probably not be performed at the same time. They prefer to treat the Dupuytren’s surgically and the CTS with steroid injection.
The most interesting study looked at 120 diabetic patients and 120 non-diabetic controls (Chammas 1995). The occurrence of Dupuytren’s, Limited mobility of the fingers (LJM), CTS and flexor tenosynovitis in these patients was as follows:
|Type I DM||60||21 (35%)||21 (35%)||16 (25%)||14 (23%)|
|Type II DM||60||17 (28%)||17 (28%)||9 (15%)||10 (16%)|
|Controls||120||10 (8%)||14 (12%)||5 (4%)||2 (2%)|
When they looked at cases with multiple problems, of 21 patients with both Type 1 diabetes and Dupuytren’s, 9 had CTS (43%) and 12 had limitation of finger movement.
As one might expect there was also a link between diabetic peripheral neuropathy and CTS.
|Incidence (%)||Type I||Type II|
This was a well conducted study with CTS diagnosis confirmed by nerve conduction studies. It is notable that the CTS incidence in the control subjects was exactly what one might expect in the population from other studies and the higher incidence seen in patients with diabetes and Dupuytren’s probably indicates a genuinely increased risk. I have not yet found a study which looks at the incidence of CTS specifically in non-diabetic Dupuytren’s patients.
The relationship between diabetes, generalized neuropathy and CTS is complex. No-one doubts that diabetes causes nerve problems of many different types. The commonest form results in loss of sensation in the feet first – what is known as a ‘length-dependent’ neuropathy – ie it affects the longest nerve fibres first and then gradually spreads to shorter ones as it becomes more severe. Studies such as that of Chammas suggest that the incidence of CTS is also higher in diabetic patients than controls though that study did not have enough patients to fully factor in all the confounding influences which might have produced their findings. At present therefore is seems very probable that diabetes predisposes to CTS but it has not been conclusively proven.
A part of the problem lies in the difficulty in accurately defining the diagnoses of CTS and diabetes. Diabetes is at least defined by fairly concrete, if arguable, criteria. If your blood glucose measurement, under defined conditions, is above a certain level then you officially have diabetes. However, a modestly elevated blood glucose does not produce any symptoms and the clinical state of hyperglycaemia (having a raised blood sugar) would never have been recognised as a ‘disease’ before the blood test became available. Diabetes the disease, historically, was defined by the symptoms of very high blood glucose, which leads to increased urine output, increased drinking, metabolic derangement, coma and death. Asymptomatic hyperglycaemia is common in Western populations – there are many people walking the streets, apparently well, who have the biochemical state of diabetes but no symptoms. When you recruit apparently healthy control subjects for medical experiments, some of them will technically have diabetes without knowing it, and you will not discover this unless you carefully check their blood sugars.
Carpal tunnel syndrome shares many of the same diagnostic characteristics. In CTS the disease is still defined by the symptoms because we do not have a test which is equivalent to the blood glucose measurement that allows us to exactly divide the population into CTS and non-CTS. The closest approach we have to such a test is the nerve conduction velocity measurement but there is less international agreement about what should be considered ‘abnormal’ median nerve conduction at the wrist. Analagous to the situation of ‘asymptomatic hyperglycaemia’ is ‘median mononeuropathy at the wrist’ on nerve conduction studies – some people without any hand symptoms have slow median nerve conduction at the wrist. Some doctors would argue that, just as patients with a high blood glucose but no symptoms are diabetic but don’t know it yet, patients with slow median nerve conduction at the wrist have CTS but don’t know it yet. Other doctors hold to the view that you only have CTS when you develop the symptoms of CTS – both views have validity but the existence of these laboratory based ‘disease’ states without symptoms complicates all assessments of whether one disease is associated with the other, especially when both clinical diseases are very common and the laboratory abnormalities are even commoner.
Two further issues complicate things still further. Firstly diabetic neuropathy is clearly a nerve problem, but not all diabetics suffer from it. As with defining the presence or absence of CTS, we do not have an absolutely reliable way of saying whether any given diabetic patient has neuropathy or not. The nerve conduction studies are much less reliable for diagnosing diabetic neuropathy than they are for CTS. It is quite possible that diabetes might predispose to CTS only through the intermediary of causing diabetic neuropathy and that the median nerve at the wrist might be the first nerve to be affected by some forms of diabetic neuropathy. When studying the association between the diseases we should therefore ideally distinguish between diabetic patients with neuropathy and those without… but we have no reliable way of doing this! One could even argue, in the extreme, that a patient with asymptomatic hyperglycaemia and NCS evidence of median mononeuropathy at the wrist, but no symptoms of anything, actually has the disease – ‘early diabetic neuropathy’.
Secondly, as we actually define CTS by it’s symptoms, there is the possibility that having diabetes, diabetic neuropathy, or even just asymptomatic hyperglycaemia, might itself modify the symptoms of CTS and thus interfere with our ability to decide whether someone has CTS or not. Add to this the fact that diabetes seems to predispose to several other hand problems, the symptoms of which can be mixed with those of CTS, and we have a recipe for diagnostic difficulty.
Revision date - 23rd August 2014