The diagnosis of an ulnar neuropathy at the elbow (UNE) is based on clinical signs and symptoms, electrodiagnostic and sonographic studies.1^–3 Patients with a motor conduction block (mCB) usually have muscle weakness.4 Surgery is often preferred in patients with UNE and muscle weakness,4 but results without surgery may be more favourable than previously thought as our prior study has suggested.5 The current study was intended to evaluate the long term outcome of patients with a severe mCB of ⩾50% in comparison with patients without an mCB.

PATIENTS AND METHODS

Patients

Between May 1998 and November 2005, 244 patients were diagnosed with UNE at our hospital. The clinical and electrophysiological findings of the patients with UNE were obtained using a previously described protocol.6 The compound muscle action potentials (CMAPs) of the abductor digiti minimi (ADM) or first dorsal interosseous (FDI) were recorded in a belly tendon montage using surface electrodes. For this analysis, the most severe abnormalities at the elbow of one of these derivations were used. mCB was defined as percentage decline in CMAP amplitude above the sulcus in comparison with CMAP amplitude below the sulcus.

Two groups of patients were selected: patients with UNE and an electrophysiologically severe mCB of ⩾50% (50% group) and a control group consisting of patients with UNE with an mCB of ⩽20% (20% group). These two groups were matched for surgery, severity of muscle weakness and age. The study was approved by the local medical ethics committee, and informed consent was obtained from each subject prior to the investigation.

Clinical evaluation

Each patient was screened for current symptoms of UNE.6 The type and date of an eventual operation was noted.

Clinical outcome measures

At follow-up, each patient was asked to indicate whether the initial complaints were absent (remission), improved, stable or progressive.

In order to compare muscle strength at the time of diagnosis to muscle strength at follow-up, we used a Medical Research Council (MRC) sum score. This score was calculated by adding the individual MRC grades of four ulnar muscles, ADM, FDI, flexor carpi ulnaris and flexor digitorum profundus (maximum score 20). Clinical examination was performed by one of the investigators, who was unaware of the initial clinical and electrophysiological data. Patients were classified as being in remission, improved, stable or progression. Remission was defined as a normal neurological examination and no complaints at follow-up; improved was defined as an improvement in the MRC sum score of 2 or more points compared with the first examination; stable was defined as an MRC sum score difference between the two examinations of less than 2 points; and progression was defined as an MRC sum score reduction of 2 or more points. The analysis was based on the number of patients and not on the number of affected arms.

For the analysis of the patient’s judgment and clinical examination, outcome was dichotomised: a favourable outcome was defined as remission or improvement, and a poor outcome was defined as either a stable or progressive classification.

Statistical analysis

For comparison of independent ordinal variables, the Mann–Whitney test was used, and for paired samples, the Wilcoxon signed rank test was used. For the comparison of proportions, the χ² test was applied. The level of significance in all analyses was set at 0.05.

RESULTS

Patient characteristics

Of the 244 patients with UNE, 38 (16%) had an mCB of ⩾50%. Six patients were lost to follow-up: two could not be contacted, one was not able to come to the hospital, two had incomplete data at the time of diagnosis and one patient was excluded because of an underlying polyneuropathy.

An mCB of ⩽20% was present in 136 of the 244 patients (56%); 32 patients were selected to be representatives of the control group.

The general and electrodiagnostic characteristics for both groups are listed in table 1. Thirty-eight men and 26 women were re-examined after a median of 25 months (interquartile range (IQR) 14 to 38). The left side was more often involved than the right side in the 50% group (p = 0.0008) (table 1). Twenty-three patients (72%) in the 50% group and 22 (64%) patients in the 20% group underwent surgery (p = 1.00). In one patient in the 50% group, surgery was planned but spontaneous recovery occurred prior to operation. Of those who underwent surgery, neurolysis was performed in 21 (91%) patients in the 50% group and in 15 (68%) patients in the 20% group. Only two (9%) patients of the 50% group and seven (22%) patients of the 20% underwent a transposition (p = 0.12).

Electrodiagnostic studies

Detailed information about the degree of mCB, CMAP at the wrist, below and above the sulcus, sensory nerve action potential at digit V after stimulation at the wrist and presence of spontaneous muscle activity on needle EMG are listed in table 1. Median CMAP of 2.2 mV (IQR 0.7–3.1) above the sulcus was lower in the 50% group compared with the median CMAP of 8.7 mV(IQR 3.1–10.0) in the 20% group (p<0.0001). Also, spontaneous muscle activity in the ADM on needle EMG was more often present at the time of diagnosis in the 50% group (p = 0.04).

Follow-up

Clinical outcome

Symptoms and signs

The symptoms and signs of UNE of the two groups at baseline and at follow-up are given in table 2. At diagnosis, significantly more patients in the 50% group had paresis of the ADM (p = 0.008), FDI (p = 0.003) and flexor digitorum profundus muscle (p = 0.04). Both groups showed a significant decline in the number of patients with paraesthesias and numbness at follow-up. Weakness/clumsiness, dorsal branch sensory loss and severity of paresis of the ulnar nerve innervated muscles (except the flexor carpi ulnaris) decreased significantly in the 50% patient group only.

At follow-up, less atrophy of the ADM and FDI was present in the 50% group (p = 0.03), an interesting but not easily interpretable finding. At follow-up, more patients in the 20% group had presence of sensory signs (p = 0.06), paresis of FDI (p = 0.03) and atrophy of the ADM (p = 0.005) and FDI (p = 0.06).

The MRC sum score of the patients in the 50% group improved significantly. The median MRC sum score at diagnosis was 16 (IQR 14.5–17) and 20 at follow-up (IQR 18–20) (p<0.0001). In the 20% group, the median MRC sum score was 18 (IQR 16–20) at baseline and only 18.5 (IQR 17–20) at follow-up (p = 0.03).

Patient’s and clinical judgment

Table 3 shows the patient’s judgment and clinical outcome at follow-up. Patients in the 50% group more often demonstrated a good outcome according to the patient’s judgment and also after clinical assessment. Surgically treated patients had a significantly better outcome in the 50% group (p = 0.05 and p = 0.0005). A good outcome after conservative treatment was present in approximately 80% of the patients in the 50% group.

DISCUSSION

In a prospective cohort of patients with UNE, we determined whether there was a difference in the long term outcome of patients with a severe mCB of ⩾50% compared with patients without an mCB. Although we matched the two groups for age, surgery and presence of muscle weakness, we were not able to obtain a perfect match. This underlines the clinical differences between patients with different electrophysiological patterns. We were able to deal with the possible effect of surgery because the groups were well matched on this basis. However, patients with an mCB of ⩾50% were slightly younger, had a shorter onset and more often had muscle weakness. The clinical differences are apparently so strong that we were not able to match 32 out of 136 patients in the 20% group with the same clinical signs and symptoms as in the 50% group.

The patient’s judgment as well as the clinical examination showed a significantly better outcome in the 50% group.

The more favourable outcome in the 50% group was probably a result of the demyelinating nature of the ulnar nerve damage. Incidental external compression at the superficial location of the ulnar nerve, especially above the elbow, may lead to an mCB, which can be the underlying pathophysiological mechanism in this patient group.7 The left side was significantly more often involved than the right side in patients with an mCB. It is possible that patients lean more often on the left elbow, which makes the left ulnar nerve more vulnerable to compression.

A few studies dealing with the prognostic role of electrodiagnostic testing did not reveal its relevance to the prognosis of UNE.8 9 However, these studies did not take mCB into account or did not have a large number of patients with a severe mCB.

Some plastic surgeons advocate surgery on patients with UNE without a confirmative electrodiagnostic assessment.10 Our present study clearly shows the clinical, electrophysiological and prognostic heterogeneity of UNE. To give patients with UNE optimal information about the nature and prognosis of their nerve lesion, we emphasise assessment on the basis of clinical as well as electrodiagnostic findings.

The prognosis was good in about 80% of the patients in the 50% group treated conservatively, albeit on the basis of a small sample size. This suggests that surgery in patients with an mCB of ⩾50% may not always be required. We propose that this hypothesis be evaluated in a randomised study.