CRANIAL NERVE INJURIES LIFE AFTER TREATMENT-QUALITY OF LIFE

TO THE EDITOR I wish to comment on the article' by Maniglia and Han. I have reviewed the data which indicate a 13.5% damage rate during carotid endarterectomy and con- sulted with our colleagues in vascular and neurosur- gery who perform similar operations. Our combined opinion is that this is a disproportionately high per- centage of injury and certainly well beyond the range of our experience. Certainly the involved surgeons should reconsider their techniques to minimize neural damage.

Long-term follow-up as suggested by the authors may have a modicum of value although in our review the periodic observation of the cranial nerve deficits is not particularly advantageous to the patient. Whether the problem can be minimized by a specialized team as suggested in the article or by proper technique re- mains a moot point as well.

Louis W. Welsh, MD John J. Welsh, MD Jenkintown, Pennsylvania

1. Maniglia AJ, Han DP. Cranial nerve injuries following ca- rotid endarterectomy: An analysis of 336 procedures. Head Neck 1991;13:121-124.

REPLY: Drs. Louis and John Welsh are under the impression that the 13.5% cranial nerve injuries following carotid endarterectomy is a disproportionally high incidence of complication which is well beyond the range of ex- perience based on the impression of their vascular and neurosurgical colleagues. Our article reviews the lit- erature very thoroughly and indicates an 8% to 16% incidence of cranial nerve injuries, reported in differ- ent series (Ransom et al, Matsumoto et al, Swann et al, Dehn et al, etc.).

Hertzer, Feldman, Biven, and Tucker a t the Cleve- land Clinic Foundation, in 1980, reviewed 240 pa- tients, prospectively, and reported an 8% incidence of vocal cord injuries, 5% hypoglossal nerve dysfunction, and 3% facial nerve weakness (16%). Drs. Welsh and their vascular and neurosurgical colleagues should do a thorough review of their own series analyzing the rate of injury of cranial nerves without the exclusion of cases with mild weakness or paresis. They might very well be surprised to find, afier a thorough study, that the percentage of cranial nerve injury following carotid endarterectomy in their series is probably very similar to what has been reported in the literature. Then, instead of anecdotal information, they will have more factual data. I will be looking forward to their report.

TO THE EDITOR: Although cancers of the head and neck region are rel- atively uncommon, the treatment associated with these tumors may produce side effects of sufficient magnitude to affect adversely the patient's quality of life. For example, radiotherapy may result in soft tis- sue necrosis, pain on swallowing, or dysphagia, while surgery may produce some disfigurement, nerve dam- age, dysphagia, altered taste perception, or the forma- tion of fistulae.

Two recent reviews have indicated that the psy- chosocial sequelae of the treatment associated with head and neck cancers are relatively unexplored. ' v 2

Problems identified included speech and eating diffi- culties, social and marital problems, and increased levels of anxiety and depression. However, there were several methodologic problems associated with many of the studies that can be summarized as follows:

1. The majority were retrospective, based on small samples which, in general, were not stratified by stage of disease and did not include control groups. Furthermore, the retrospective nature of some of the studies lead to an over-representation of patients with positive treatment outcomes.'

2. The measures of quality of life tended to be rather crude and largely focused on functional parame- ters. None of the studies used instruments used in studies of other cancer patients (e.g., the Rotter- dam Symptom Checklist, the Spitzer QL index, the Functional Living Index for Cancer, the EORTC quality of life questionnaire). Detailed in- formation about the scales appropriate for use with cancer patients is provided e l~ewhere .~ -~

3. None of the studies investigated the extent to which moderating variables, such as social sup- port, psychological state and adjustment to the disease and treatment, might influence outcome.

4. Little consideration was given to the quality of life of the caretakers of head and neck cancer pa- tients. The relevance of this can be appreciated from the results of studies that have investigated the quality of life of both cancer patients and their next of kin.6*7

The importance of measuring quality of life can be seen from the results from studies of patients with breast cancer. As a consequence of such work, much is known about the psychosocial morbidity following mastectomy: the differences in psychological func- tioning between patients undergoing conservative sur- gery and those undergoing mastectomy: and the ex- tent to which coping styles influence outcome.'o," The application of these results can be seen in the develop- ment of intervention," and communication skills pro- grams.13

Additionally, quality of life data can provide im- portant information to be used in decision making es-

Anthony J. Maniglia, MD Department of Otolaryngology- Head and Neck Surgery Case Western Reserve University Cleveland, Ohio

554 Letters to the Editor HEAD & NECK NovernbedDecember 1991

pecially where two forms of treatments are known to be equivalent in terms of survival, or where the aims of treatment are palliative rather than curative. Deci- sions about treatment inevitably involve trade-offs be- tween length of survival and quality of life, and there may be differences between individuals regarding the extent to which they would wish to trade quality of life for length of survival. If quality of life information is well documented for the posttreatment period, this will help provide doctors and patients with more infor- mation to help maximize lifestyle and trade disability from treatment for potential survival.

Further work needs to be undertaken using con- trolled trials to measure the quality of life in head and neck cancer patients. Instruments that have been shown to be reliable and valid in studies with other cancer patients should be used to assess the effect of diagnosis and treatment on psychological, social, occu- pational, and sexual functioning. Where possible, questionnaires should be completed by the patients themselves; this has been shown to be a more reliable method than relying upon quality of life assessments made by doctors and nurses.14

It is essential that an operational definition of quality of life be stated at the outset of a trial and the objectives of collecting quality of life data made clear. This will increase the likelihood of selecting the most appropriate questionnaires, determine the period of time over which the data should be collected, and con- sequently ensure that the best use is made of the addi- tional resources necessary to collect and analyze such data. Generally, quality of life assessments will need to be made prior to treatment, during treatment, a t the end of treatment, and at least once during the first year following treatment because the majority of re- currences occur within 1 year.

For those unfamiliar with quality of life and/or psychosocial research, there are practical consider- ations to bear in mind vis-a-vis the recruitment of data collection staff. For example, being asked to com- plete the same questionnaires on a number of occa- sions may seem tedious to patients and the data col- lection staff will need to be prepared to explain the rationale for collecting information that may appear to patients to be unrelated to their disease. Further- more, staff may be asked the meaning of some of the items or response categories on the questionnaires, and it is essential that they are able to provide expla- nations without biasing responses. It will be appreci- ated, therefore, that the staff involved should be fa- miliar with research techniques and, if not, training will need to be provided to help recruit patients into quality of life trials, ensure that the questionnaires used are completed correctly, and keep missing data to a minimum.

One final issue regarding treatment relates to the role of economic evaluation. Although cancers of the head and neck region are relatively uncommon, some of them are among the most preventable of tumors,

being most prevalent in individuals who have a his- tory of heavy drinking and/or smoking.15 Thus ques- tions might be asked about the opportunity costs of providing relatively expensive treatment to patients with advanced disease compared to financing health education programs, or ensuring that general practi- tioners are alerted to the need to screen individuals at most risk of developing such cancers. Such evalua- tions would be particularly relevant for patients with head and neck cancers because they often are initially seen with advanced disease, and also because they may continue to smoke or drink heavily during and/or after treatment which increases the risk of recurrence.

Methods of economic evaluation that can be used in combination with quality of life assessment include cost-effectiveness analysis and cost-utility analy- sis.16,17 By combining information relating to quality of life and costs, comparisons can be made of the rela- tive costs of achieving different outcomes.

In summary, the evaluation of the quality of life of patients following treatment for cancers of the head and neck region could provide useful information which would (1) aid decision making where different modes of treatment result in similar clinical outcomes; (2) provide a focus for rehabilitation programs; and (3) help identify those individuals unlikely to cope with the consequences of diagnosis and treatment in order that they may receive additional support. Together with economic evaluations, such data would also pro- vide more comprehensive information on the costs and consequences of the uses of health care resources than that currently provided by mortality statistics.

Jenny Morris, PhD Centre for Health Economics University of York Heslington, York, England

1.

2.

3.

4.

5.

6.

7.

8.

9.

Pruyn J, Jong P, Bosman L, et al. Psychosocial aspects of head and neck cancer-a review of the literature. Clin Otolaryngol 1986;11:469-474. Morris J. The quality of life of head and neck cancer pa- tients: a review of the literature. Discussion Paper 72, Centre for Health Economics, York, UK, 1990. Morris J. The measurement of quality of life: how to choose an appropriate test. Cancer Topics, 1991;

Maguire P, Selby P. Assessing quality of life in cancer patients. Br J Cancer 1989;60:437-440. Aaronson A, Bullinger M, Ahmedzai S. A modular ap- proach to quality of life assessment in cancer clinical tri- als. Recent Results in Cancer Research 1988;111:231- 249. Cassileth B, Lusk E, Strouse T, et al. A psychological analysis of cancer patients and their next-of-kin. Cancer

Morris J, Royle G. Offering patients a choice of surgery for early breast cancer: a reduction in anxiety and de- pression in patients and their husbands. SOC Science Med 1988;26:583- 585. Morris T, Greer S, White P. Psychological and social ad- justment to mastectomy: A 2 year follow up study. Can- cer 1977;40:2381-2387. Bartelink H, van Dam F, van Dongen J. Psychological ef- fects of breast conserving therapy in comparison with

8~75-76.

1985;55:72-76.

Letters to the Editor HEAD & NECK NovembeVDecember 1991 555

10.

11.

12.

13.

14.

15.

16.

17.

radical mastectomy. Znt J Radiation Oncol Biol Phys

Greer S, Morris T, Pettingale K. Psychological response to breast cancer: effect on outcome. Lancet 1979;2:785- 787. Temoshok L. Personality, coping style, emotion and can- cer: towards an integrative model. Cancer Surveys

Spiegel D, Bloom J, Kraemer H, Gottheil E. Effect of psy- chosocial treatment on survival of patients with meta- static breast cancer. Lancet 19891:888-891. Maguire P, Faulkner A. Improve the counselling skills of doctors and nurses in cancer care. Br Med Journal

Slevin M, Plant H, Lynch D, Drinkwater J, Gregory W. Who should measure quality of life, the doctor or the pa- tient? Br J Cancer 1988;57:109-112. Boyle P, Madarlane G, Maisonneuve P, Zheng T, Scully C, Tedesco B. Epidemiology of mouth cancer in 1989: a review. J Royal Soc Medicine 1990;83:724-730. Drummond M, Stoddart G, Torrance G. Methods for eco- nomic evaluation of health care programmes. Oxford: Oxford University Press, 1987. Feeny D, Labelle R, Torrance G. Integrating economic evaluations and quality of life assessments. In Spilker B (ed). Quality of life assessments in clinical trials. New York Raven Press, 1990:71-83.

1985;11:381-385.

1987;6545-567.

1988;297:847-849.

REPLY Dr. Weymuller correctly notes that a multivariate analysis of our data reflects a significantly increased risk of recurrent disease in the neck (p = 0.0001) and death with cancer present ( p = 0.0121) if clinically positive neck nodes are treated with irradiation alone compared with irradiation followed by planned neck dissection.’ Irradiation alone results in a similar prob- ability of neck control compared with irradiation and neck dissection for early-stage neck disease, but yields inferior results for advanced neck d i ~ e a s e . ~ . ~

In an unpublished analysis of 459 patients with 493 clinically positive heminecks, Ellis noted the fol- lowing 5-year neck control rates for irradiation alone compared with irradiation plus neck dissection: N1, 86% vs. 93% ( p = 0.28); N2A, 79% vs. 68% 0, = 0.60); N2B, 70% vs. 91% ( p < 0.01); and N3A, 33% vs. 69% (p < 0.01). Some patients who were scheduled to un- dergo a neck dissection following irradiation did not do so if they were not good medical candidates for sur- gery and the nodeb) regressed completely during the course of radiotherapy, biasing the comparison of re- sults against the combined treatment group. There was also a tendency to treat more advanced presenta- tions within each substage with irradiation plus neck

NECK-NODE BIOPSY BEFORE RADIOTHERAPY

TO THE EDITOR: The article by Ellis et al.’ adds supportive information to the evolving concept that neck-node biopsy does not compromise local recurrence rate or survival if subse-

dissection (e.g., for N2B with two 1-cm nodes, irradia- tion alone; for multiple 3- to 5-cm nodes, irradiation plus neck dissection). These factors give even more strength to the argument that neck dissection after ir- radiation significantly improves the rates of neck con- trol compared with irradiation alone.

quent treatment includes definitive radiotherapy. I am writing to point out that another piece of informa- tion contained in this study was not emphasized by the authors. In Table 7 the probability of recurrence in the neck being more likely after treatment with ra- diotherapy alone has a p value of 0.0001. Similarly, the probability of death with disease present after treatment of the neck with radiotherapy alone is 0.012.

The authors also comment that “the possible ex- planation as to why the biopsy group had a strong trend to improved neck control and significantly higher cause specific survival rate-may be that most of the N1 and N2A heminecks in the biopsied group were converted to NX heminecks with a resulting dis- ease control rate of loo%.”

Viewed from a slightly different perspective, this data indicate that the combination of surgery and ra- diation for neck disease is the most favorable form of management.

Ernest A. Weymuller, Jr, YD Department of Otolaryngology University of Washington Seattle

1. Ellis ER, Mendenhall WM, Rao PV, et al. Incisional or ex- cisional neck-node biopsy before definitive radiotherapy, alone or followed by neck dissection. Head Neck 1991;13:177-183.

Our current policy for managing patients with clinically positive neck nodes in whom the primary is to be treated with irradiation alone is to use radio- therapy alone for the neck if there are one or two pos- itive nodes 1 2 cm in diameter located within the high-dose volume that have regressed completely by the end of radiotherapy. Additionally, patients with lymphoepithelioma may often be treated with irradia- tion alone because the probability of disease control is higher for this histology than for other subsets of squamous cell c a r ~ i n o m a . ~ We prefer to add a neck dissection 4 to 6 weeks after irradiation for patients with more advanced neck d i ~ e a s e . ~

Although the probability of neck disease control with irradiation alone increases if the nodes regress completely during treatment, the probability of sal- vaging a patient after failed irradiation for clinically positive nodal disease is r e m ~ t e . ~ In a series of 139 pa- tients treated with irradiation alone a t the University of Florida for clinically positive neck nodes, only 2 (6%) of 35 patients who developed a neck recurrence were salvaged.6 Similarly, Bernier and Bataini re- ported that only 1 (1%) of 102 patients who developed an isolated neck recurrence after irradiation alone for a positive neck node was successfully salvaged.’ Therefore, the decision to add the neck dissection is made before initiation of irradiation and is rarely

556 Letters to the Editor HEAD & NECK NovembedDecember 1991