Cervical spine cancer

Although there has been a growing and considerable amount of literature concerning the shoulder and resultant disability thereof following various forms of neck dissection for cancer of the head and neck, relatively little attention has been paid to the status of the cervical and upper thoracic spine after such procedures. This may well be due to the decreased incidence of trapezius weakness and atrophy that has been seen concomitant to the increased frequency of modified and selective neck dissection procedures that are employed.

However, it has become apparent to clinicians involved in cancer rehabilitation that there is still a relatively high incidence of shoulder pain, weakness, and dysfunction even in patients undergoing a modified or selective node dissection procedure. For this reason, attention has shifted to the shoulder in head and neck cancer. Since many patients who have a modified or selective neck dissection as treatment for head and neck cancer will still have some resultant, frequently transient, weakness in the trapezius due to nerve apraxia, traction during surgery, or unnecessary or rough manipulation of the accessory nerve, it is still necessary and prudent to consider the cervical and upper thoracic spine in performing a therapeutic evaluation in these patients.

LITERATURE REVIEW

Talmi et al1 examined pain in the neck after neck dissection. Although reports of disability after neck dissection have been directed toward shoulder dysfunction and pain, little had been said concerning pain in the operative site. The authors found that chronic pain localized to the operative site was uncommon, but that shoulder pain and disability were frequent after a radical neck dissection but uncommon after a modified neck dissection and the incidence was comparable to that seen in the literature. Cheng, Hao, and Lin2 performed a prospective study of an objective comparison of shoulder dysfunction after3 neck dissection techniques: selective, modified, and radical. Although shoulder strength was somehow decreased in the selective dissection group, EMG findings of the accessory nerve were normal in this group. Patients who underwent selective nerve dissection experienced the least damage to cranial nerve XI, and the least shoulder disability and pain following this procedure. Despite short comings in their work, it is notable that it addresses the new neck procedures, which are becoming a mainstay in head and neck surgical approaches. The fact that EMG and torque muscle evaluations were used instead of simple manual muscle techniques is worthy of mention and adds some solidification to the results.

Kuntz and Weymuller3 looked into the quality of life evaluation of patients with the 3 procedures: radical neck dissection (RND), modified radical neck dissection (MRND), and selective neck dissection (SND). Since the difference in survival between treatment arms of the 3 procedures may well be comparable, treatment options may indeed become quality of life driven. They found the MRND and RND group had worse shoulder function at 6 and 12 months compared to pretreatment scores. Also, although the MRND group reported more shoulder disability at 6 months postsurgery, by 12 months there was no difference, implying that although the accessory nerve might be slightly damaged during a MRND, the ensuing disability was relatively temporary. Pinsolle et al4 studied 124 patients, and found an incidence of severe or significant shoulder disability in 7%, 34%, and 51% of patients undergoing an RND, MRND, and SND (supraomohyoid neck dissection in their paper) respectively. However, they pointed out that there was a large variation in the degree of functional disability and pain in patients with similar neck dissections, and that a functional and sometime transient disability can be expected in any neck dissection procedure where the accessory nerve is placed in traction, ie, dissected but not transected.

Nori and his colleaguess used intraoperative electro- neurography to assist in determining the presence of motor input in the spinal accessory nerve from the cervical plexus (C2, C3, C4) and found that although there is motor contribution from C2, C3, and C4 to the trapezius muscle is neither consistent or significant. They hypothesized that patients with good functional outcomes following MRND procedures may have good contributions from C2, C3, C4 or its branches, while those with less favorable outcomes did not. Additional studies along this line, comparing the electroneurography of the trapezius nerve among the different procedures, are in order.

London, London, and Kay6 in their study reminded surgeons that the accessory nerve can be damaged even in cervical lymph node biopsy or surgical procedures other than a standard neck dissection, and that awareness of iatrogenic neural injury and its consequences, could avoid delays and treatment. Patten and Hillel7 outlined the "11th nerve syndrome" and contended that numerous findings of shoulder disability that are not attributable to accessory nerve palsy but are well described by adhesive capsulitis of the glenohumeral joint. They proposed that adhesive capsulitis accounts for the persistence and variability of shoulder symptoms after neck dissection that cannot be attributed to trapezius muscle function.

Hillel et al8 looked both subjectively and objectively at a series of patients undergoing a RND, and found great variability in disability following the procedure; peak torque for the affected side was 0% to 85% of the normal side, and 8 of 11 patients had pain free antigravity range of motion of 100 deg or less in forward flexion/ abduction. The fact that shoulder pain limited the motion is significant and lends support to the findings of London et al that adhesive capsulitis might account for shoulder pain and limitation of range of motion that is observed. Fialka and Vinzenz9 performed EMG on the trapezius after RND and demonstrated damage mainly in the descending part of the trapezius while in the majority of patients the ascending part of the trapezius was only slightly damaged or normal. It is of importance that these authors reported on the efficacy of physical therapy in cases of irreversible shoulder disability seen after a RND.

Remmler et al10 in a prospective longitudinal study looked at the shoulder function of 103 neck dissections, a mixture of RND and MRND. They were among the early investigators that found that the MRND was followed by a significant, but temporary and reversible period of shoulder function. Just one year earlier, Sobol et al11 reported severely abnormal electromyographic readings in RND patients, less abnormal electromyographic readings in the MRND patients, and the latter improved at 1 year. Correlational analysis revealed that the physical parameters correlated well with the electromyographic findings, whereas each patient's perception of disability did not. This is of interest because it has been observed by many other investigators that objective testing frequently did not mirror the patient's function or perception of function. Leipzeg et al12 were one of the first group of researchers to report the difference in disability between RND patients and MRND patients. Those patients in whom the nerve, muscle, and vein were preserved had less dysfunction (30%) than those with MRND (50%) and classical RND (60%). There was less associated pain with nerve sparing procedures, and curiously, 40% of patients who underwent the RND had minimal disability.

In an attempt to define the impact of RND and MRND on patient's permanent disability, Schuller et al13 interviewed via survey 243 patients, and showed no advantage of one surgery over the other in returning patients to their pretreatment employment status. This of course is not a reliable measure of disability status in patients who have had a neck dissection procedure, but merely a statement of the percentage of patient's returning to work. Radiation therapy, however, did not contribute to the patient's "permanent disability."