- Email: [email protected]
Dynamics of oral intake resumption after general anesthesia and operation in rats
Physiology&Behavior,Vol.52, pp. 597-601, 1992
0031-9384/92 $5.00 + .00 Copyright@ 1992PergamonPressLtd.
Printedin the USA.
Dynamics of Oral Intake Resumption After General Anesthesia and Operation in Rats M I C H A E L M. M E G U I D , l Z H O N G - J I N Y A N G , C A R L O R A T T O , R O B E R T A. M E G U I D , D A V I D C. H I T C H , A K I O K U B O T A A N D A N T O N I O C A R L O S L . C A M P O S
Surgical Metabolism and Nutrition Laboratory, Department of Surgery, University Hospital, S U N Y Health Science Center, Syracuse, N Y 13210 Received 9 D e c e m b e r 1991 MEGUID, M. M., Z.-J. YANG, C. RATTO, R. A. MEGUID, D. C. HITCH, A. KUBOTA AND A. C. L. CAMPOS. Dynamics of oral intake resumption aftergeneralanesthesia and operation in rats. PHYSIOL BEHAV 52(3) 597-601, 1992.--The influence of general anesthesia and operation on dynamics of postoperative food intake resumption was investigated in eight rats. A laparotomy was performed on each rat under anesthesia induced by intraperitoneally injected chloral hydrate. Spontaneous food intake and feeding indexes were continuously measured using an Automated Computerized Rat Eater Meter (ACREM) before and after operation. Although spontaneous food intake and all feeding indexes were depressed immediately followinganesthesia and operation, each feedingindex was depressed to a greater degree during the dark vs. the light cycle. Initially, rats fully capable of eating ate fewer, smaller, and shorter meals. The return to normal of each t~eding index differed temporally. Thus, although meal number normalized by the third postoperative day, meal size by the sixth postoperative day, and meal duration by the fourth postoperative day, normalization of meal number during the light cycle occurred sooner than during the dark cycle, while the converse occurred with meal size and meal duration. Rat General anesthesia Feeding behavior
Abdominal operation
Food intake
Feedingindexes
Nocturnal
Diurnal
dietary history); the other patient had a prolonged recovery of SFI. However, in none of these studies were the food intake patterns, i.e., meal number, meal size, and meal duration, following anesthesia and operation measured. Since patients usually are not discharged from the hospital until they are eating enough to maintain themselves, describing the pattern of normal SFI after operation so as to more clearly understand nutritional factors that prolong hospital stay poses a significant practical problem. The magnitude of the problem becomes evident when one supposes that the postoperative hospital stay is prolonged for 3 days at $300/day in one patient per month in each of 500 hospitals. These conservative but entirely warranted assumptions culminate in extra health care costs totaling $5,400,000 per year ($300/day × 3 days X 12 months × 500 hospitals). We undertook our current study in rats to
THE customary eating pattern is disrupted by general anesthesia and operation. Even after nongastrointestinal operations, spontaneous food intake is initially depressed, then increases as convalescence progresses until the patient once more eats an adequate amount. The period from operation until patients eat sufficiently to maintain themselves, has been defined as the inadequate oral nutrient intake period (10). Durations of the inadequate oral nutrient intake period were measured in 101 patients with benign (14) or 365 patients with a variety of malignant diseases undergoing abdominal operations (11). The amounts eaten daily were recorded to assess that oral intake needed to assure timely discharge without subsequent readmission because of nutritional problems. The factors related to shortening or prolongation of the postoperative inadequate oral nutrient intake period included the primary diagnosis, the type of operation, and the types of postoperative complications. Attempts have been made by others (4) to quantitate the rate of postoperative food intake recovery in 12 patients following elective colectomy. Total spontaneous food intake (SFI) was measured, and in 11 patients, SFI was begun 2 to 7 days after operation, reaching their normal SFI between 7 and 14 days (based on estimated preoperative home SFI levels obtained by
1. gain insight into the dynamics of the resumption of oral feeding after general anesthesia and operation, and 2. see whether this postoperative period is associated with alterations in individual feeding indexes which define food intake patterns.
Work presented in this paper was supported in part by Grant AM36275 from the National Institute of Arthritis, Diabetes, Digestive & Kidney Diseases, NIH. Requests for reprints should be addressed to Michael M. Meguid, M.D., Ph.D., F.A.C.S., Department of Surgery, University Hospital, SUNY Health Science Center, 750 East Adams Street, Syracuse, NY 13210.
597
598
MEGUID ET AL.
While findings from rat studies may not exactly correlate with feeding behavior changes in patients, they are less likely to be affected by long-standing differences in interindividual food preferences and eating habits. Therefore, they may be useful for developing an increased understanding of what happens to food intake recovery mechanism(s) after anesthesia and operation. METHOD
Experimental Subjects Adult male Fischer 344 rats (Charles River, Inc., Wilmington, MA), weighing 280-300 g, were housed in holding cages for 710 days following purchase in order to acclimatize them to the study surroundings. Throughout the experiment the environmental conditions were kept constant (12 h light/dark cycles; room temperature 26 ° + 1°C; and 45% humidity). Rats were allowed free access to water and fresh coarsely ground Purina chow (Diet No. 5008; Ralston Purina, St. Louis, MO). After acclimatization, rats which gained weight and were healthy were studied. The study was approved by the Committee for the Humane Use of Animals.
Anesthesia and Operation Selection oJ'chloral hydrate as anesthetic agent. Intraperitoneal chloral hydrate was selected instead of inhalation anesthesia, because a more uniform level of narcosis and analgesia was obtained in rats and because it avoided the drying of the oral cavity and upper airway which might have interfered with early postoperative resumption of SFI. Furthermore, chloral hydrate has a negligible effect on blood glucose and other metabolites, historically considered to be factors influencing food intake (6,9). Three percent chloral hydrate (0.3 g/kg body weight) was given intraperitoneally via a lower abdominal midline injection site for anesthesia. Then the abdomen was shaved and the skin prepared with povidone-iodine 10% (Betadine®). A midline incision was made from the xyphoid to the symphysis pubis. The liver, the stomach, and the distal esophagus were explored by gentle mobilization. Normal saline was used to irrigate the abdominal cavity. The abdominal incision was then closed in separate layers. This procedure took about 30 rain. Anesthesia and operation were usually performed between 1400 and 1600 h; rats took up to 90 rain to wake from anesthesia and appeared fully recovered by late evening.
number (g/meal) during each light and dark cycle, or in 24 h; and meal duration - total duration of all the meals per total number of meals eaten (rain/meal) in each light and dark cycle, or during 24 h. A fresh supply of water and coarsely ground Purina chow, from the same initial stock, was provided each day at 0800 h throughout the study. Food intake and feeding indexes were continuously measured for 3 days preoperatively, and their mean served as normal preoperative feeding pattern. After the standard laparotomy, all rats were replaced in their own individual cages. Food intake and feeding indexes were then continuously measured after operation until the normal preoperative feeding pattern was resumed.
Statistical Analysis Data were analyzed using analysis of variance contrast t-test comparing post to preoperative events. Data are expressed as mean + standard error. RESULTS
Spontaneous Food Intake (SFI) As shown in Fig. 1A, mean daily preoperative SFI was 14~0 _+ 0.7 g/day (49 _+ 2.5 kcal/day). As shown in Fig. 1B, rats ate twice as much food during the dark cycle than during the light FOOD INTAKE BEFORE AND AFTER GENERAL ANESTHESIA AND OPERATION 24
20
=<
12
z
8 C~ C~ r~
4
0
1
2
3
4
5
6
7
DAY
Experimental Design Three days before surgery eight healthy rats were transferred from their holding cages to individual metabolic cages equipped with an Automated Computerized Rat Eater Meter (ACREM). The ACREM consists of commercially available metabolic cages, with the supplied feeding cup replaced by an electronic scale balance and two photoelectric cells centered above the food dish (12). Access to food is via a feeding tunnel monitored by photocells, while food consumption is measured by an electronic scale which weighs the amount eaten during each access to food. A remote computerized data-collection device integrates feeding activity, as measured by the electronic scale and the photocells, with real time. The signals, so generated, are processed and interpreted by a computer and recorded. As previously defined (12), a meal is a bite or series of bites preceded and followed by at least 5 minutes of feeding inactivity. The ACREM was used to continually measure and quantitate feeding indexes during each light and dark cycle. These include: meal number -- total number of meals during each 12-h light and dark cycle, or each 24 h; meal size = total amount of food consumed per total meal
FOOD INTAKE DURING LIGHT AND DARK CYCLES:
EFFECT OF GENERAL ANESTHESIA AND OPERATION 16" I ProOperation
Post-Operation
[]
UGHTSON
•
LIGHTSOFF
0 0
1
2
3
4
5
6
7
DAY
FIG. 1. (A-B) After anesthesia and operation, food intake was depressed, the greatest influence occurring during the dark cycle. *p < 0.05; **p < 0.01 vs. preoperation.
EFFECT OF ANESTHESIA AND OPERATION ON FOOD INTAKE
NUMBER OF MEALS BEFORE AND AFTER GENERAL ANESTHESIA AND OPERATION
599
NUMBER OF MEALS DURING LIGHT AND DARK CYCLES: EFFECT OF GENERAL ANESTHESIA AND OPERATION
24
.a. 18 ] 161pr."
A
~141~ Operatlon• E
16
< uJ
12
~ 8
~
[1 UGHrSON] • LJGHTSOFFI
Post Operation
,o
.°
4 0
0
0
1
2
3
4
5
6
~
7
0
1
2
3
4
5
6
7
DAY
DAY
MEAL SIZE DURING LIGHT AND DARK CYCLES: EFFECT OF GENERAL ANESTHESIA AND OPERATION
MEAL SIZE BEFORE AND AFTER GENERAL ANESTHESIA AND OPERATION
1.8
1.6 t ~emt~n Pre1.4 t
Po~43pemUon
o"
~ 1.2 I.(]
0.8
,..2 <
0.6 0.4 0.2 O.C 0
I
2
3
4
5
6
7
0
1
2
3
MEAL DURATION BEFORE AND AFTER GENERAL ANESTHESIA AND OPERATION
5
6
7
MEAL DURATION DURING LIGHT AND DARK CYCLES: EFFECT OF GENERAL ANESTHESIA AND OPERATION
10 Pre-
Operation
4 DAY
DAY
-~"
Post-Operation
1°1
r'l u ~ v s ~
Pre-
|
~e
z ~ 4
ev
°
,,
"
~2
0
I
2
3
4 DAY
5
6
7
0 0
1
2
3
&
5
6
7
DAY
FIG. 2. (A-B) Anesthesia and operation had marked effects on the different feeding indexes and its influence was apparent during different light/ dark cycles. Initially, rats fully capable of eating ate fewer, smaller, and shorter meals, thereby reducing food intake. The return to normal of the feeding indexes differed temporally. *p < 0.05; **p < 0.01 vs. preoperation.
600
MEGUID ET AL.
cycle; SFI being 4.4 _+ 0.3 g in the light cycle and 9.6 + 0.5 g in the dark cycle. The combined effects of anesthesia and operation were to depress SFI. On postoperative days 1 and 2, SFI was 3.0 + 0.9 g/day (p < 0.01) and 2.3 _+ 0.7 g/day (p < 0.01), respectively. Then, SFI gradually increased and returned to preoperative level by postoperative day 7 (p = NS). The decrease in daily SF! was due to a decrease in both the light and dark cycle food intake after anesthesia and operation. The ratio between dark and light cycle food intake became smaller. After the fourth postoperative day, the low daily SFI was due primarily to the significantly lower dark cycle SFI (see Fig. 1B), because the light cycle SF! had returned to preoperative level. The dark vs. light cycle food intake ration gradually increased, being 1.7 on 7th postoperative day.
DISCUSSION
Prior to proceeding to studying the combined influence of general anesthesia and operation on the dynamics of food intake in rats, a series of pilot studies were perforrned in order to distinguish the contributory effects of the anesthesia. When chloral hydrate was given intravenously (IV = 0.1 g/kg body weight), anesthesia was briefly induced but was not maintained for a sufficiently long period nor at a level of analgesia adequate to perform the exploratory laparotomy. There was no significant influence on rat's daily SFI after intravenous chloral hydrate alone. In contrast, intraperitoneal chloral hydrate depressed SFI for 6-8 h. Hence, the delay in resumption of SFI during the early recovery stage is related to the combined effects of a more prolonged anesthesia plus operation; thereafter, the dominant influence on SFI was the operative procedure. Although an abdominal incision was used, an additional viscera-altering procedure was avoided, because we wished to investigate the influence of laparotomy, unconfounded by factors consequent to gastrointestinal alterations which might mechanically influence food intake. Prior to operation, the overall food intake, feeding indexes, and feeding behavior pattern of the rats were normal, while the ratios for the light-to-dark values for each index was similar to that previously reported (12,15). The influence which anesthesia and operation had on each index was different during each cycle. Since SFI is normally greater during the dark versus the light cycle, the influence which anesthesia and operation had on SFI depression was greater and for a longer time during the dark cycle; light cycle SFI normalizing sooner. In contrast, meal size and meal duration normalized sooner during the dark than the light cycle. The rat continued to have an abnormal meal pattern during the light cycle by having fewer meals of shorter duration. Our data show that the daily pattern, as well as the diurnal pattern, of SFI and feeding indexes are profoundly disrupted by the combined effects of general anesthesia and operation, even following a benign procedure, such as a straightforward laparotomy. Disruption would be expected to be greater after more invasive procedures involving excision of a portion of the gastrointestinal tract followed by an anastomosis. Based on this model, it is apparent that after anesthesia and operation, the individual indexes making up food intake are changed, not only relative to one another, but nychthemerally as well. These factors are being taken into consideration in our current studies of patients after undergoing a variety of operations performed under general anesthesia. Although such measurements are reproducible in more stable situations, in postoperative patients they are fraught with potential methodological problems. These include: 1. patient subjectivity, 2. the hospital environment, 3. patient's self esteem, 4. depression and mood changes, 5. pain status, and 6. medication status (2,3,5). However, the insights into the dynamics of oral intake resumption after general anesthesia and operation in rats, gained from the use of the rat model, provides an objective and reproducible framework for studies in patients. They will materially assist in elucidating the complex problem of the dynamics of human food intake following operation.
The rat was chosen as our experimental model because spontaneous food intake, and the feeding behavior of rats have been well established and documented in detail (1,7,8,13).
ACKNOWLEDGEMENTS We thank Mrs. Darlene Thompson for editorial assistance and William G. Hammond, M.D., for continued counsel and advice.
Meal Number, Size and Duration As shown in Fig. 2A, the mean daily meal number, size, and duration before anesthesia and operation were 14.6 _+ 0.7 meals/day, 0.99 + 0.04 g/meal, and 4.7 _+ 0.4 rain/meal, respectively. The meal number, meal size, and meal duration during the light cycle were 4.6 _+ 0.4 meals/24 h, 1.04 + 0.07 g/meal, and 4.9 + 0.3 min/meal; and 10.0 + 0.4 meals/24 h, 0.96 _+ 0.04 g/meal, and 4:3 + 0.5 min/meal during the dark cycle (Fig. 2B). Meal number was about two times more during the dark cycle than light cycle. Meal size and duration were nearly the same in both cycles. Figure 2A shows that following anesthesia and operation, daily meal number on the first 2 days were significantly below preoperative values; being 5.6 + 2.1 meals/day (p < 0.01) and 6.1 + 1.8 meals/day (p < 0.01). The reduction in meal number was greater during the dark cycle (p < 0.01) than during the light cycle (p < 0.05) on the first 2 days after surgery (Fig. 2B). By postoperative day 3, meal number in both cycles returned to preoperative level and, consequently, resulted in the normalized daily meal number, although the ratio between both cycles was still lower than preoperative level. As shown in Fig. 2A, during the first 3 postoperative days, meal size was only about 40% of preoperative value. Thereafter, meal size rose to about 70% of preoperative value on postoperative days 4 and 5. By postoperative day 6, it had normalized relative to the preoperative value, being 0.80 + 0.04 g/meal vs. 0.99 + 0.04 g/meal (NS). As shown in Fig. 2B, both the light and dark cycle meal size was significantly depressed by anesthesia and operation, resulting in a significantly decreased mean daily meal size. The dark cycle meal size recovered to preoperative level on the fourth day after operation, 1 day earlier than during the light cycle. Figure 2A also shows that meal duration was significantly shortened by anesthesia and operation, being 2.3 + 0.8, 2.0 + 0.4, and 2.9 + 0.7 min/meal on postoperative day 1, day 2, and day 3, respectively. Meal duration normalized by postoperative day 4. As shown in Fig. 2B, meal duration during the dark cycle was affected by anesthesia and operation for only 1 day. The shortened daily meal duration was mainly due to a shortened light cycle meal duration. There was a minimal effect of anesthesia and operation on the ratio of meal size and meal duration between two cycles except on the second postoperative day.
EFFECT OF ANESTHESIA AND OPERATION ON FOOD INTAKE
601
REFERENCES 1. Balagura, S.; Coscina, D. V. Periodicity of food intake in the rat as measured by an operant response. Physiol. Behav. 3:641-643; 1968. 2. Garner, D. M.; Garfinkel, P. E. The eating attitudes test: An index of the symptoms of anorexia nervosa. Psychol. Med. 9:273-279; 1979. 3. Garner, D. M.; Olmsted, M. P.; Bohr, Y.; Garfinkel, P. E. The eating attitudes test: Psychometric features and clinical correlates. Psychol. Med. 12:871-878; 1982. 4. Hackett, A. F.; Yeung, C. K.; Hill, G. L. Eating patterns in patients recovering from major surgery--A study of voluntary food intake and energy balance. Br. J. Surg. 66:415-418; 1979. 5. Kammann, R.; Flett, R. Affectometer 2: A scale to measure current level of general happiness. Aust. J. Psychol. 35:259-265; 1983. 6. Kennedy, G. C. The role of a depot fat in the hypothalamic control of food intake in the rat. Proc. R. Soc. Lond. [Biol] 140:578-592; 1956. 7. Kissileff, H. R. Free feeding in normal and recovered lateral rats monitored by a Pellet-Detecting Eatometer. Physiol. Behav. 5:163173; 1970. 8. Levitsky, D. A. Feeding patterns of rats in response to fasts and changes in environmental conditions. Physiol. Behav. 5:291-300; 1970.
9. Mayer, J. Regulation of energy intake and the body weight: The glucostatic theory and the lipostatic theory. Ann. NY Acad. Sci. 63: 15-42; 1955. 10, Meguid, M. M.; Campos, A. C. L.; Meguid, V.; Debonis, D.; Terz, J. J. 1ONIP: A criterion of surgical outcome and patient selection for perioperative nutritional support. Br. J. Clin. Pract. 42:8-14; 1988. 11. Meguid, M. M.; Debonis, D.; Meguid, V.; Hill, L. R.; Terz, J. J. Complications of abdominal operations for malignant disease. Am. J. Surg. 156:341-345; 1988. 12. Meguid, M. M.; Kawashima, Y.; Campos, A. C. L. et al. Automated computerized rat eater meter: Brief description and application. Physiol. Behav. 48:'159-763; 1990. 13. Morrison, S. D. Amount of feeding activity and size of meals in free-feeding rats. Physiol. Behav. 25:893-399; 1980. 14. Mughal, M. D.; Meguid, M. M. The effect of nutritional status on morbidity after elective surgery for benign gastrointestinal disease. J. Parenter. Ent. Nutr. 11:140-143; 1987. 15. Yang, Z.-J.; Bodoky, G.; Meguid, M. M. Nocturnal cyclic versus continuous total parenteral nutrition: Food intake and feeding pattern in rats. Physiol. Behav. 51:431-435; 1992.
0031-9384/92 $5.00 + .00 Copyright@ 1992PergamonPressLtd.
Printedin the USA.
Dynamics of Oral Intake Resumption After General Anesthesia and Operation in Rats M I C H A E L M. M E G U I D , l Z H O N G - J I N Y A N G , C A R L O R A T T O , R O B E R T A. M E G U I D , D A V I D C. H I T C H , A K I O K U B O T A A N D A N T O N I O C A R L O S L . C A M P O S
Surgical Metabolism and Nutrition Laboratory, Department of Surgery, University Hospital, S U N Y Health Science Center, Syracuse, N Y 13210 Received 9 D e c e m b e r 1991 MEGUID, M. M., Z.-J. YANG, C. RATTO, R. A. MEGUID, D. C. HITCH, A. KUBOTA AND A. C. L. CAMPOS. Dynamics of oral intake resumption aftergeneralanesthesia and operation in rats. PHYSIOL BEHAV 52(3) 597-601, 1992.--The influence of general anesthesia and operation on dynamics of postoperative food intake resumption was investigated in eight rats. A laparotomy was performed on each rat under anesthesia induced by intraperitoneally injected chloral hydrate. Spontaneous food intake and feeding indexes were continuously measured using an Automated Computerized Rat Eater Meter (ACREM) before and after operation. Although spontaneous food intake and all feeding indexes were depressed immediately followinganesthesia and operation, each feedingindex was depressed to a greater degree during the dark vs. the light cycle. Initially, rats fully capable of eating ate fewer, smaller, and shorter meals. The return to normal of each t~eding index differed temporally. Thus, although meal number normalized by the third postoperative day, meal size by the sixth postoperative day, and meal duration by the fourth postoperative day, normalization of meal number during the light cycle occurred sooner than during the dark cycle, while the converse occurred with meal size and meal duration. Rat General anesthesia Feeding behavior
Abdominal operation
Food intake
Feedingindexes
Nocturnal
Diurnal
dietary history); the other patient had a prolonged recovery of SFI. However, in none of these studies were the food intake patterns, i.e., meal number, meal size, and meal duration, following anesthesia and operation measured. Since patients usually are not discharged from the hospital until they are eating enough to maintain themselves, describing the pattern of normal SFI after operation so as to more clearly understand nutritional factors that prolong hospital stay poses a significant practical problem. The magnitude of the problem becomes evident when one supposes that the postoperative hospital stay is prolonged for 3 days at $300/day in one patient per month in each of 500 hospitals. These conservative but entirely warranted assumptions culminate in extra health care costs totaling $5,400,000 per year ($300/day × 3 days X 12 months × 500 hospitals). We undertook our current study in rats to
THE customary eating pattern is disrupted by general anesthesia and operation. Even after nongastrointestinal operations, spontaneous food intake is initially depressed, then increases as convalescence progresses until the patient once more eats an adequate amount. The period from operation until patients eat sufficiently to maintain themselves, has been defined as the inadequate oral nutrient intake period (10). Durations of the inadequate oral nutrient intake period were measured in 101 patients with benign (14) or 365 patients with a variety of malignant diseases undergoing abdominal operations (11). The amounts eaten daily were recorded to assess that oral intake needed to assure timely discharge without subsequent readmission because of nutritional problems. The factors related to shortening or prolongation of the postoperative inadequate oral nutrient intake period included the primary diagnosis, the type of operation, and the types of postoperative complications. Attempts have been made by others (4) to quantitate the rate of postoperative food intake recovery in 12 patients following elective colectomy. Total spontaneous food intake (SFI) was measured, and in 11 patients, SFI was begun 2 to 7 days after operation, reaching their normal SFI between 7 and 14 days (based on estimated preoperative home SFI levels obtained by
1. gain insight into the dynamics of the resumption of oral feeding after general anesthesia and operation, and 2. see whether this postoperative period is associated with alterations in individual feeding indexes which define food intake patterns.
Work presented in this paper was supported in part by Grant AM36275 from the National Institute of Arthritis, Diabetes, Digestive & Kidney Diseases, NIH. Requests for reprints should be addressed to Michael M. Meguid, M.D., Ph.D., F.A.C.S., Department of Surgery, University Hospital, SUNY Health Science Center, 750 East Adams Street, Syracuse, NY 13210.
597
598
MEGUID ET AL.
While findings from rat studies may not exactly correlate with feeding behavior changes in patients, they are less likely to be affected by long-standing differences in interindividual food preferences and eating habits. Therefore, they may be useful for developing an increased understanding of what happens to food intake recovery mechanism(s) after anesthesia and operation. METHOD
Experimental Subjects Adult male Fischer 344 rats (Charles River, Inc., Wilmington, MA), weighing 280-300 g, were housed in holding cages for 710 days following purchase in order to acclimatize them to the study surroundings. Throughout the experiment the environmental conditions were kept constant (12 h light/dark cycles; room temperature 26 ° + 1°C; and 45% humidity). Rats were allowed free access to water and fresh coarsely ground Purina chow (Diet No. 5008; Ralston Purina, St. Louis, MO). After acclimatization, rats which gained weight and were healthy were studied. The study was approved by the Committee for the Humane Use of Animals.
Anesthesia and Operation Selection oJ'chloral hydrate as anesthetic agent. Intraperitoneal chloral hydrate was selected instead of inhalation anesthesia, because a more uniform level of narcosis and analgesia was obtained in rats and because it avoided the drying of the oral cavity and upper airway which might have interfered with early postoperative resumption of SFI. Furthermore, chloral hydrate has a negligible effect on blood glucose and other metabolites, historically considered to be factors influencing food intake (6,9). Three percent chloral hydrate (0.3 g/kg body weight) was given intraperitoneally via a lower abdominal midline injection site for anesthesia. Then the abdomen was shaved and the skin prepared with povidone-iodine 10% (Betadine®). A midline incision was made from the xyphoid to the symphysis pubis. The liver, the stomach, and the distal esophagus were explored by gentle mobilization. Normal saline was used to irrigate the abdominal cavity. The abdominal incision was then closed in separate layers. This procedure took about 30 rain. Anesthesia and operation were usually performed between 1400 and 1600 h; rats took up to 90 rain to wake from anesthesia and appeared fully recovered by late evening.
number (g/meal) during each light and dark cycle, or in 24 h; and meal duration - total duration of all the meals per total number of meals eaten (rain/meal) in each light and dark cycle, or during 24 h. A fresh supply of water and coarsely ground Purina chow, from the same initial stock, was provided each day at 0800 h throughout the study. Food intake and feeding indexes were continuously measured for 3 days preoperatively, and their mean served as normal preoperative feeding pattern. After the standard laparotomy, all rats were replaced in their own individual cages. Food intake and feeding indexes were then continuously measured after operation until the normal preoperative feeding pattern was resumed.
Statistical Analysis Data were analyzed using analysis of variance contrast t-test comparing post to preoperative events. Data are expressed as mean + standard error. RESULTS
Spontaneous Food Intake (SFI) As shown in Fig. 1A, mean daily preoperative SFI was 14~0 _+ 0.7 g/day (49 _+ 2.5 kcal/day). As shown in Fig. 1B, rats ate twice as much food during the dark cycle than during the light FOOD INTAKE BEFORE AND AFTER GENERAL ANESTHESIA AND OPERATION 24
20
=<
12
z
8 C~ C~ r~
4
0
1
2
3
4
5
6
7
DAY
Experimental Design Three days before surgery eight healthy rats were transferred from their holding cages to individual metabolic cages equipped with an Automated Computerized Rat Eater Meter (ACREM). The ACREM consists of commercially available metabolic cages, with the supplied feeding cup replaced by an electronic scale balance and two photoelectric cells centered above the food dish (12). Access to food is via a feeding tunnel monitored by photocells, while food consumption is measured by an electronic scale which weighs the amount eaten during each access to food. A remote computerized data-collection device integrates feeding activity, as measured by the electronic scale and the photocells, with real time. The signals, so generated, are processed and interpreted by a computer and recorded. As previously defined (12), a meal is a bite or series of bites preceded and followed by at least 5 minutes of feeding inactivity. The ACREM was used to continually measure and quantitate feeding indexes during each light and dark cycle. These include: meal number -- total number of meals during each 12-h light and dark cycle, or each 24 h; meal size = total amount of food consumed per total meal
FOOD INTAKE DURING LIGHT AND DARK CYCLES:
EFFECT OF GENERAL ANESTHESIA AND OPERATION 16" I ProOperation
Post-Operation
[]
UGHTSON
•
LIGHTSOFF
0 0
1
2
3
4
5
6
7
DAY
FIG. 1. (A-B) After anesthesia and operation, food intake was depressed, the greatest influence occurring during the dark cycle. *p < 0.05; **p < 0.01 vs. preoperation.
EFFECT OF ANESTHESIA AND OPERATION ON FOOD INTAKE
NUMBER OF MEALS BEFORE AND AFTER GENERAL ANESTHESIA AND OPERATION
599
NUMBER OF MEALS DURING LIGHT AND DARK CYCLES: EFFECT OF GENERAL ANESTHESIA AND OPERATION
24
.a. 18 ] 161pr."
A
~141~ Operatlon• E
16
< uJ
12
~ 8
~
[1 UGHrSON] • LJGHTSOFFI
Post Operation
,o
.°
4 0
0
0
1
2
3
4
5
6
~
7
0
1
2
3
4
5
6
7
DAY
DAY
MEAL SIZE DURING LIGHT AND DARK CYCLES: EFFECT OF GENERAL ANESTHESIA AND OPERATION
MEAL SIZE BEFORE AND AFTER GENERAL ANESTHESIA AND OPERATION
1.8
1.6 t ~emt~n Pre1.4 t
Po~43pemUon
o"
~ 1.2 I.(]
0.8
,..2 <
0.6 0.4 0.2 O.C 0
I
2
3
4
5
6
7
0
1
2
3
MEAL DURATION BEFORE AND AFTER GENERAL ANESTHESIA AND OPERATION
5
6
7
MEAL DURATION DURING LIGHT AND DARK CYCLES: EFFECT OF GENERAL ANESTHESIA AND OPERATION
10 Pre-
Operation
4 DAY
DAY
-~"
Post-Operation
1°1
r'l u ~ v s ~
Pre-
|
~e
z ~ 4
ev
°
,,
"
~2
0
I
2
3
4 DAY
5
6
7
0 0
1
2
3
&
5
6
7
DAY
FIG. 2. (A-B) Anesthesia and operation had marked effects on the different feeding indexes and its influence was apparent during different light/ dark cycles. Initially, rats fully capable of eating ate fewer, smaller, and shorter meals, thereby reducing food intake. The return to normal of the feeding indexes differed temporally. *p < 0.05; **p < 0.01 vs. preoperation.
600
MEGUID ET AL.
cycle; SFI being 4.4 _+ 0.3 g in the light cycle and 9.6 + 0.5 g in the dark cycle. The combined effects of anesthesia and operation were to depress SFI. On postoperative days 1 and 2, SFI was 3.0 + 0.9 g/day (p < 0.01) and 2.3 _+ 0.7 g/day (p < 0.01), respectively. Then, SFI gradually increased and returned to preoperative level by postoperative day 7 (p = NS). The decrease in daily SF! was due to a decrease in both the light and dark cycle food intake after anesthesia and operation. The ratio between dark and light cycle food intake became smaller. After the fourth postoperative day, the low daily SFI was due primarily to the significantly lower dark cycle SFI (see Fig. 1B), because the light cycle SF! had returned to preoperative level. The dark vs. light cycle food intake ration gradually increased, being 1.7 on 7th postoperative day.
DISCUSSION
Prior to proceeding to studying the combined influence of general anesthesia and operation on the dynamics of food intake in rats, a series of pilot studies were perforrned in order to distinguish the contributory effects of the anesthesia. When chloral hydrate was given intravenously (IV = 0.1 g/kg body weight), anesthesia was briefly induced but was not maintained for a sufficiently long period nor at a level of analgesia adequate to perform the exploratory laparotomy. There was no significant influence on rat's daily SFI after intravenous chloral hydrate alone. In contrast, intraperitoneal chloral hydrate depressed SFI for 6-8 h. Hence, the delay in resumption of SFI during the early recovery stage is related to the combined effects of a more prolonged anesthesia plus operation; thereafter, the dominant influence on SFI was the operative procedure. Although an abdominal incision was used, an additional viscera-altering procedure was avoided, because we wished to investigate the influence of laparotomy, unconfounded by factors consequent to gastrointestinal alterations which might mechanically influence food intake. Prior to operation, the overall food intake, feeding indexes, and feeding behavior pattern of the rats were normal, while the ratios for the light-to-dark values for each index was similar to that previously reported (12,15). The influence which anesthesia and operation had on each index was different during each cycle. Since SFI is normally greater during the dark versus the light cycle, the influence which anesthesia and operation had on SFI depression was greater and for a longer time during the dark cycle; light cycle SFI normalizing sooner. In contrast, meal size and meal duration normalized sooner during the dark than the light cycle. The rat continued to have an abnormal meal pattern during the light cycle by having fewer meals of shorter duration. Our data show that the daily pattern, as well as the diurnal pattern, of SFI and feeding indexes are profoundly disrupted by the combined effects of general anesthesia and operation, even following a benign procedure, such as a straightforward laparotomy. Disruption would be expected to be greater after more invasive procedures involving excision of a portion of the gastrointestinal tract followed by an anastomosis. Based on this model, it is apparent that after anesthesia and operation, the individual indexes making up food intake are changed, not only relative to one another, but nychthemerally as well. These factors are being taken into consideration in our current studies of patients after undergoing a variety of operations performed under general anesthesia. Although such measurements are reproducible in more stable situations, in postoperative patients they are fraught with potential methodological problems. These include: 1. patient subjectivity, 2. the hospital environment, 3. patient's self esteem, 4. depression and mood changes, 5. pain status, and 6. medication status (2,3,5). However, the insights into the dynamics of oral intake resumption after general anesthesia and operation in rats, gained from the use of the rat model, provides an objective and reproducible framework for studies in patients. They will materially assist in elucidating the complex problem of the dynamics of human food intake following operation.
The rat was chosen as our experimental model because spontaneous food intake, and the feeding behavior of rats have been well established and documented in detail (1,7,8,13).
ACKNOWLEDGEMENTS We thank Mrs. Darlene Thompson for editorial assistance and William G. Hammond, M.D., for continued counsel and advice.
Meal Number, Size and Duration As shown in Fig. 2A, the mean daily meal number, size, and duration before anesthesia and operation were 14.6 _+ 0.7 meals/day, 0.99 + 0.04 g/meal, and 4.7 _+ 0.4 rain/meal, respectively. The meal number, meal size, and meal duration during the light cycle were 4.6 _+ 0.4 meals/24 h, 1.04 + 0.07 g/meal, and 4.9 + 0.3 min/meal; and 10.0 + 0.4 meals/24 h, 0.96 _+ 0.04 g/meal, and 4:3 + 0.5 min/meal during the dark cycle (Fig. 2B). Meal number was about two times more during the dark cycle than light cycle. Meal size and duration were nearly the same in both cycles. Figure 2A shows that following anesthesia and operation, daily meal number on the first 2 days were significantly below preoperative values; being 5.6 + 2.1 meals/day (p < 0.01) and 6.1 + 1.8 meals/day (p < 0.01). The reduction in meal number was greater during the dark cycle (p < 0.01) than during the light cycle (p < 0.05) on the first 2 days after surgery (Fig. 2B). By postoperative day 3, meal number in both cycles returned to preoperative level and, consequently, resulted in the normalized daily meal number, although the ratio between both cycles was still lower than preoperative level. As shown in Fig. 2A, during the first 3 postoperative days, meal size was only about 40% of preoperative value. Thereafter, meal size rose to about 70% of preoperative value on postoperative days 4 and 5. By postoperative day 6, it had normalized relative to the preoperative value, being 0.80 + 0.04 g/meal vs. 0.99 + 0.04 g/meal (NS). As shown in Fig. 2B, both the light and dark cycle meal size was significantly depressed by anesthesia and operation, resulting in a significantly decreased mean daily meal size. The dark cycle meal size recovered to preoperative level on the fourth day after operation, 1 day earlier than during the light cycle. Figure 2A also shows that meal duration was significantly shortened by anesthesia and operation, being 2.3 + 0.8, 2.0 + 0.4, and 2.9 + 0.7 min/meal on postoperative day 1, day 2, and day 3, respectively. Meal duration normalized by postoperative day 4. As shown in Fig. 2B, meal duration during the dark cycle was affected by anesthesia and operation for only 1 day. The shortened daily meal duration was mainly due to a shortened light cycle meal duration. There was a minimal effect of anesthesia and operation on the ratio of meal size and meal duration between two cycles except on the second postoperative day.
EFFECT OF ANESTHESIA AND OPERATION ON FOOD INTAKE
601
REFERENCES 1. Balagura, S.; Coscina, D. V. Periodicity of food intake in the rat as measured by an operant response. Physiol. Behav. 3:641-643; 1968. 2. Garner, D. M.; Garfinkel, P. E. The eating attitudes test: An index of the symptoms of anorexia nervosa. Psychol. Med. 9:273-279; 1979. 3. Garner, D. M.; Olmsted, M. P.; Bohr, Y.; Garfinkel, P. E. The eating attitudes test: Psychometric features and clinical correlates. Psychol. Med. 12:871-878; 1982. 4. Hackett, A. F.; Yeung, C. K.; Hill, G. L. Eating patterns in patients recovering from major surgery--A study of voluntary food intake and energy balance. Br. J. Surg. 66:415-418; 1979. 5. Kammann, R.; Flett, R. Affectometer 2: A scale to measure current level of general happiness. Aust. J. Psychol. 35:259-265; 1983. 6. Kennedy, G. C. The role of a depot fat in the hypothalamic control of food intake in the rat. Proc. R. Soc. Lond. [Biol] 140:578-592; 1956. 7. Kissileff, H. R. Free feeding in normal and recovered lateral rats monitored by a Pellet-Detecting Eatometer. Physiol. Behav. 5:163173; 1970. 8. Levitsky, D. A. Feeding patterns of rats in response to fasts and changes in environmental conditions. Physiol. Behav. 5:291-300; 1970.
9. Mayer, J. Regulation of energy intake and the body weight: The glucostatic theory and the lipostatic theory. Ann. NY Acad. Sci. 63: 15-42; 1955. 10, Meguid, M. M.; Campos, A. C. L.; Meguid, V.; Debonis, D.; Terz, J. J. 1ONIP: A criterion of surgical outcome and patient selection for perioperative nutritional support. Br. J. Clin. Pract. 42:8-14; 1988. 11. Meguid, M. M.; Debonis, D.; Meguid, V.; Hill, L. R.; Terz, J. J. Complications of abdominal operations for malignant disease. Am. J. Surg. 156:341-345; 1988. 12. Meguid, M. M.; Kawashima, Y.; Campos, A. C. L. et al. Automated computerized rat eater meter: Brief description and application. Physiol. Behav. 48:'159-763; 1990. 13. Morrison, S. D. Amount of feeding activity and size of meals in free-feeding rats. Physiol. Behav. 25:893-399; 1980. 14. Mughal, M. D.; Meguid, M. M. The effect of nutritional status on morbidity after elective surgery for benign gastrointestinal disease. J. Parenter. Ent. Nutr. 11:140-143; 1987. 15. Yang, Z.-J.; Bodoky, G.; Meguid, M. M. Nocturnal cyclic versus continuous total parenteral nutrition: Food intake and feeding pattern in rats. Physiol. Behav. 51:431-435; 1992.