Physiology&Behavior,Vol. 50, pp. 1103-1109. ©Pergamon Press plc. 1991. Printed in the U.S.A.

0031-9384/91 $3.00 + .00

A Descriptive Study of Learned Food Aversions in Radiotherapy Patients R I C H A R D D. M A T r E S 1

Monell Chemical Senses Center, Philadelphia, PA W A L T E R J. C U R R A N , JR.

Fox Chase Cancer Center, Philadelphia, PA WILLIAM POWLIS AND RICHARD WHITrlNGTON

Hospital of the University of Pennsylvania R e c e i v e d 2 M a y 1991 MATTES, R. D., W. J. CURRAN, JR., W. POWLIS AND R. WHrITINGTON. A descriptive study of learnedfood aversions in radiotherapypatients. PHYSIOL BEHAV 50(6) 1103-1109, 1991.--While animal studies have documented the efficacy of ionizing radiation as a conditioning stimulus for food aversion learning, little is known about the incidence or nature of aversions that form in cancer patients administered radiation therapy. Forty-nine newly diagnosed patients with a variety of cancer types were monitored for the duration of their treatment or up to six months. Aversions were identified by questionnaires and responses to an experimental food administered at stipulated time points. Fifty-nine percent of patients formed new aversions after initiating treatment and for 53% of patients, this involved items in their customary diet. A wide array of items were targeted, but there was direct relationship between exposure frequency and aversion formation. The aversions were highly specific (median=2.5 items/ afflicted patient). The onset time was approximately 1-2 weeks, and they were transient, generally lasting 1-3 weeks. There was no evidence of a discrepant incidence among patients treated in different sites. Aversion incidence was not significantly associated with age, gender or other selected patient characteristics. Food aversions

Cancer

Radiotherapy

Cancer treatment

item during their course of therapy. Again, treatment in the abdominal region was especially effective. Finally, Carrell et al. (6) noted that the 5 patients in their sample of 34 who became nauseated during treatment ingested significantly less of a fruit juice provided in association with treatments. Given the clear evidence that ionizing radiation is a potent conditioning stimulus in rats, an animal model from which generalizations to humans are often made, the less consistent findings with humans was problematic. The present study was designed to better quantify the incidence of food aversions in radiotherapy patients and to characterize the aversions with respect to the types of foods targeted, aversion duration and the temporal association between food ingestion and aversion formation.

IONIZING radiation and emitogenic pharmacologic agents have been used to condition aversions to foods in various species (8,16). Studies with rats indicate that exposure to radiation, especially in the abdominal region (20), is a particularly effective conditioning paradigm. Several human studies have documented and characterized food aversions in patients with cancer receiving chemotherapy (1, 2, 4, 12, 13), but there is limited information available on the incidence and nature of aversions in patients with cancer undergoing radiotherapy (6, 19, 20). In a pioneering study, Smith et al. (19) reported that patients treated in the pelvic region complained of loss of appetite and decreased preference for selected items, but no specific aversions. One of two patients treated in the upper abdomen did exhibit a clear food aversion. In a follow-up study, this group obtained equivocal findings (20). While all patients indicated an unwillingness to consume as much of a juice previously paired with treatment, only 4 individuals reported a decreased preference for the beverage. Questionnaire data from a separate group of 56 radiotherapy patients revealed that only 4-14 individuals (depending on the definition of aversion used) formed an aversion to a specific

METHOD

Protocol Patients were interviewed during the hour preceding their first course of radiotherapy and during the same period 3, 5, 7, 14,

1Requests for reprints should be addressed to Richard Mattes, Ph.D., Monell Chemical Senses Center, 3500 Market St., Philadelphia, PA 191043308.

1103

11(!4

MAI'FES t:l Ai_

TABLE I

Type

N

Sex (m/f)

Breast Lung Lymphoma Testicle Hodgkin's Prostate Colon Bladder Endometrium Head and neck Gastric Esophagus Primary? Total

8 11 2 1 6 9 1 1 1 6 1 1 I 49

0/8 9/2 1/1 1/0 2/4 9/0 1/0 1/0 0/t 3/3 1/0 1/0 1/0 30/19

Age (years)

Pretrt Weight (kg)

Current Weight (kg)

Height Im~

J)~s~: I(iY

52 ± 12 61 ± 8 40 _+ 30 24 30 ± 13 70 ± 8 62 72 70 54 ± 20 57 77 37 55 _+ 18

66.4 = 10.4 84.5 ± 13.5 78.4 +_ 20.9 90.4 62.6 -+ 11.1 78.4 _+ 8.2 88.2 75.9 84.5 72.0 ± 11.6 92.7 63.6 70.0 75.7 ± 13.3

66.4 ± 10.0 81.4 ~- 13.2 83.6 ± 19.9 9 I. 8 59.4 -+ 12.4 77.9 ± 8.7 84.5 75.4 82.7 71.7 ± 12.4 89.5 60.0 65.9 74.4 _+ 13.3

1.63 +- ~10 1.72 ± .10 1.71 z..05 I~72 1.60 ± .06 176 ~_ .03 1.80 1.75 I. 52 1.69 1.75 1.68 1.79 1.70 :~ .10

6~.8 z: 5 4 60.9 -~ 25.9 4I).0 ± 0.0 25.2 46.3 -~ 15.5 66.4 -+ 27.0 50.4 45.0

68.4 ± 28.3 39.2 -. -

Health, demographic and treatment data on patients with cancers at specific sites. Values represent the mean + S.D.

21 and 28 days later as well as 2, 4 and 6 months following the initiation of therapy. Information was collected on health, demographic and dietary variables via questionnaires. In addition, at the initial meeting, patients were sequentially assigned to one of three study groups. The first received a sample of vanillaflavored halvah [a sesame confection (Joyva Corp., Brooklyn, NY)] to taste and evaluate with respect to its sensory properties. The second group was provided a sample of chocolate-flavored halvah, and group three received no food sample. This was repeated at each visit until the fourth week, when samples of both flavors were presented to all subjects for subsequent meetings. The amount of halvah consumed was recorded on all occasions. A 24-hour diet record was taken at the first meeting, and patients rated the acceptability of all ingested items on a 9-point scale with descriptors ranging from " I would eat this every opportunity I h a d " to " I would eat this only if I were forced t o . " The patients kept a record of all items ingested for the following 24 hours, again indicating the acceptability of all items, and returned the completed form at the next meeting (two days later). All items recorded on the two diet records were rerated at each interview with the same scale. Blood samples were drawn during the first and final sessions to allow assessment of selected indices of nutritional status. Information about treatment regimen and outcome was obtained from medical records, Subjects

Patients were recruited from consecutively presenting outpatients of the Departments of Radiation Oncology at the Fox Chase Cancer Center and Hospital of The University of Pennsylvania over a three-year period. The sample characteristics are presented in Table 1. All patients received photon radiation on a medical linear accelerator with energies ranging from 6 to 18 MeV. Treatments were generally delivered 5 days per week with daily doses of 1.8 to 2.0 Gy per treatment. Dosimetry was calculated using a modern computer-based treatment planning system, and radiation exposure was prescribed as a minimum target dose. The mean (SD) doses delivered to patients with cancers at different sites are listed in Table 1. A definitive primary site was not determined on 1 patient, and information about the cumulative dose administered was not available for 3 patients. All par-

ticipants were scheduled to receive their first course of radiotherapy. Patients were excluded if they had diabetes, impending bowel obstruction, physical impairments of normal ingestion, liver or brain metastases, a Karnofsky score less than 60 or were to receive any chemotherapy during the next six months. An attempt was made to follow all patients for 6 months, but due to deaths, treatment refusal and change of treatment site, only 82%, 24.5%. 10.2% and 2.0% of participants were followed for 4. 8, 16 and 24 weeks, respectively. Learned Food Aversion Assessment Instruments

Food aversions were documented by three measures. The first was by self-report on open-ended questionnaires. To reduce inclusion of trivial changes in ratings, the patient had to indicate that the thought of ingesting the item may have been sufficient to make them ill before the item was rated as aversive. Items reported to be aversive prior to the start of therapy were not included in the analysis. The second measure required a drop of at least 5 units on the 9-point category scale used to rate the acceptability of items ingested during the 48 hours surrounding the first course of treatment. Items that dropped by at least 3 units but received the lowest possible rating were also considered aversive. The final measure involved changes in responses to the halvah presented to about two-thirds of patients prior to their first treatment and all participants at later sessions. If the flavor of halvah presented prior to the first treatment constituted less than 30% of the total amount of halvah ingested when the two flavors were presented together, it was viewed as being aversive. A 50% or greater reduction in the total amount of sample ingested was also interpreted as an aversion. Pilot studies with healthy subjects indicated that the two flavors were equally palatable and that repeated ingestion for 6 months did not result in a significant decrement in intake. Indeed, the pilot tests indicated the halvah was initially viewed as novel and became more acceptable over time. Statistical Analyses

Because responses on certain factors were only on the ordinal level and were not normally distributed, nonparametric sta-

RADIOTHERAPY AND FOOD AVERSION

tistical tests were used. Cramer's V statistic was computed for tests of the association between the incidence of aversions and other patient characteristics. Where significant relationships were noted, lambda was calculated to assess the proportion of the variance in the dependent variable explained by the independent variable. Differences between patients were evaluated by the McNemer test. A probability of p

A descriptive study of learned food aversions in radiotherapy patients.

While animal studies have documented the efficacy of ionizing radiation as a conditioning stimulus for food aversion learning, little is known about t...
675KB Sizes 0 Downloads 0 Views