- Email: [email protected]
Obesity and post-operative pain
Journalof PsychosomolicReseorch, Vol. 29, No. 1, pp. 43-48. 1985. Printed in Great Britain.
COZZ-3999/85 $3.00+ .OO 0 1985 Pergamon Press Ltd.
OBESITY AND POST-OPERATIVE PAIN COLLEENS. W. RAND, JOHN M. KULDAU and RICHARDL. YOST (Received 2 May 1984; accepted in revisedform 20 July 1984)
Abstract-Patients underwent elective abdominal surgery for morbid obesity (z= 154 kgs, R= 55) or cholecystectomy (x = 71 kgs, n = 54). Post-operative narcotics were transformed into morphine equivalent units (ME). Morbidly obese patients received significantly fewer total doses than cholecystectomy patients and less total mg ME/kg over a five-day period. Sedative use for both groups was comparable. In the morbidly obese patients, preoperative psychiatric and drug usage data predicted 67% of the variance in number of doses and 69% of the variance in total mg ME/kg. INTRODUCTION PAIN usually follows major surgery. Amounts of post-operative analgesia requested by patients, however, can vary enormously. This study examined pre-existing characteristics of morbidly obese patients that may influence post-operative narcotic use: (1) their obesity, and (2) their history of psychiatric distress and previous drug exposure. Endogenous opiates are believed to have a role in regulating pain and controlling eating behaviors [l]. Higher levels of endogenous opiates appear to decrease pain sensitivity. For example, rats with increased release and utilization of endogenous opiates reacted more slowly to pain stimuli than controls [2]. Artificial elevation, through intravenous administration of synthetic beta-endorphins, provided pain relief to cancer patients [3]. Eating and weight regulation are also associated with endogenous opiates. Administration of naltrexone, an opiate antagonist, has led to decreased eating and weight loss among patient groups [4]. Elevated concentrations of beta-endorphins have been found in the pituitaries of genetically obese mice and rats [5]. Morbidly obese adults (i.e. 100 pounds or 100% over ideal weight) may, like the genetically obese rodents, owe at least part of their condition to excessively high amounts of endogenous opiates. We hypothesized that if both eating and pain sensitivity are at least in part under the control of endogenous opiates, then morbidly obese adults should experience less post-surgical pain than normal weight adults. In the course of the investigation testing this hypothesis, we found tremendous variability in analgesia requested by both morbidly obese patients following surgery for obesity and normal weight patients following cholecystectomy. We were able to address additional possibilities to explain this variability within the morbidly obese group using data obtained from another ongoing study of these patients [6]. Unfortunately, similar data were not available for the normal weight patients. Some of the variability in use of post-operative analgesia may be explained by psychic distress [7] and willingness to use drugs [8]. Beecher [9] noted that morphine provides greater pain relief when psychic distress is present. Stembach [lo], discussing
Universityof Florida, Department of Psychiatry and Department of Pharmacy, U.S.A. Correspondence and reprint requests should be addressed to: Colleen S. W. Rand, Ph.D., University Gainesville, Florida, 32610, U.S.A.
ofFlorida, Box J-256, JHMHC,
43
44
C. S. W. RAND, J. M. KULDAU and R. L. YOST
his own work and that of Beecher, proposed that analgesics often have a greater impact on the psychogenic fear-anxiety ‘reaction component’ of pain than the original sensations. Most surgical patients have high levels of anxiety both pre- and postoperatively [l 11. Post-operative analgesics may be even more effective for those patients who have elevated psychiatric distress in addition to ‘normal’ pre-operative anxiety. There are large cultural variations in willingness to use drugs, regardless of pain experienced [8]. One index of willingness is past drug use of any kind (e.g. diet pills, tranquilizers). In addition, past exposure to situations where physicians prescribed drugs may increase the probability of drug use in similar situations (e.g. prior surgeries). Our second hypothesis was that pre-operative measures of psychiatric distress and past use of drugs in morbidly obese patients would be positively related to their post-operative narcotic use. METHOD A random selection was made of patients who underwent elective abdominal surgery at the University of Florida between 1974 and 1980: jejunoileal bypass for morbid obseity (n = 55) or cholecystectomy (n=54). Each patient’s chart was reviewed to determine analgesic and sedative use for the first five post-operative days. To standardize analgesic use, all total daily doses were converted to a reference standard dose (ME) of morphine sulfate administered subcutaneously [12].* To control for the marked group differences in body weight, two measures of post-operative narcotic use were selected: total number of doses, which ignores differences in body weight, and ME/kg body weight, where body weight is taken into account. Sedative agents were simply noted as present or absent for the study days. Sedative agents included agents listed in categories 4:00, 28:16:08, and 28:04 in the American Hospital Formulary Service [13]. The pre-operative medications used for both groups were similar. The regimen generally consisted of a narcotic analgesic (morphine or meperidine), atropine, and/or a sedative (hydroxyzine or diazepam). Maintenance anesthesia usually was comprised of nitrous oxide, thiopental sodium, a skeletal muscle relaxant, and a narcotic (morphine or fentanyl). Volatile anesthetic agents such as halothane, enfluane, or methoxyflurane were given to three morbidly obese and 17 cholecystectomy patients. The surgical service at Shands Teaching Hospital in Gainesville, Florida, requests a psychiatric consultation for patients hospitalized for surgery for obesity. Systematic data collection was initiated in 1976 as part of a longitudinal prospective study of these patients [6]. The longitudinal project provided pre-operative psychiatric and psychosocial data on 32 of the morbidly obese patients in our study. These 32 patients were interviewed with a semistructured format. Medical history and several areas of social functioning were reviewed and a scoreable standardized mental status schedule, the Present State Examination (PSE), was included [14]. Patients also completed the Minnesota Multiphasic Personality Inventory (MMPI) [15]. Variables selected for this study were limited to a few demographic statistics and items indicating anxiety, distress, and exposure to medications: prior counseling, feelings of loneliness, PSE item No. 10 nervous tension, PSE levels of definition [6], all scales in the MMPI, and MMPI scales with standardized scores of at least 80; binge eating, night eating syndrome, and smoking; prior use of sedatives, sleeping pills, tranquilizers, amphetamines, antidepressants, diet pills, and pills to ‘feel better’; general health, history of hospitalizations, and prior surgeries; age, sex, and social class. Patient groups were compared using multivariate analysis of variance, r-test, and chi-square analyses. Morbidly obese patients’ pre-operative data were analyzed using stepwise multiple regression and partial correlations. RESULTS
Sample characteristics The patient groups were compared on the following characteristics: age, sex, weight, previous surgeries and physicians’ prescribing characteristics (see Table I). *Amount of analgesia equivalent to 10 ME, administered orally (PO) and intramuscularly 75 mg Meperidine IM; 45 mg Pentazocine IM; 120 mg Codeine IM; 240 mg Codeine PO; Hydromorphone IM.
(IM): 1.5 mg
45
Obesity and pain TABLE I.-PATIENT
Cholecystectomy Patients (n = 54)
Variable
Age Sex Weight at admission Total previous surgeries
44.402 15.80yr 12 men 42 women 70.92 + 17.95 Kgs 1.85 + 1.71
CHARACTERISTICS
Morbidly obese Patients (n = 55)
Probability P<
33.20 + 9.03 yr 24 men 31 women 154.46? 36.84 kgs 1.29+ 1.30
0.001* 0.05** 0.001* 0.10*
*t-test, 2-tail. **chi-square.
Morbidly obese patients were significantly younger than cholecystectomy patients @ I O.OOl), significantly heavier (p I O.OOl), and more often women (p < 0.05). The number of past surgeries was the same. Although there was variability over the six years in physicians’ prescribing practices regarding type of analgesia, quantity, and dosing rate, there were no systematic differences in physicians’ orders for the two patient groups. Morbidly obese and cholecystectomy patient comparisons Morbidly obese patients received fewer total doses than cholecystectomy patients (F=9.18 compared to 13.13;p(O.O1), and lower total ME/kg (x=0.58 compared to 1.57; p < 0.001); see Table II. These results could not be accounted for by age or sex differences between the groups. The same levels of significance were maintained for both total dose and ME/kg when an age matched subsample of 35 patients from each group was selected; pairs within each age were randomly matched. Correlations TABLE HA.-POST-SURGICAL NARCOTICUSE: NUMBEROF DOSESAND MORPHINEEQUIVALENT UNITS PER KILOGRAM(ME/KG)
Narcotic
use
Total doses Day 1 Day 2 Day 3 Day 4 Day 5
Cholecystectomy Patients (n = 54) 13.13 + 7.95 3.78? 1.96 3.15k2.10 2.57 f 2.02 1.67& 1.89 1.01 f 1.60
Morbidly obese Patients (n = 55) 9.18 2 2.56 * 2.35 + 1.84? 1.15 + 0.55 +
7.43 1.72 2.09 1.99 1.67 1.03
Probability PS 0.05 0.001 0.05 0.06 NS 0.10
TABLE IIB.-POST-SURGICAL NARCOTICUSE: NUMBER OF DOSESAND MORPHINEEQUIVALENT UNITS PER KILOGRAM(ME/KG)
Narcotic
use
Total ME/Kg Day 1 Day 2 Day 3 Day 4 Day 5
Cholecystectomy Patients (n = 54)
Morbidly obese Patients (n = 55)
1.57 + 1.21 0.53 * 0.35 0.40 * 0.35 0.27 ? 0.27 0.17kO.26 0.08 +0.15
0.58 f 0.48 0.18 +0.13 0.17 +0.16 0.11+0.14 0.05 & 0.08 0.02 + 0.04
Total doses and total ME/Kg were analyzed using multivariate analyses Individual comparisons by day were then analyzed using r-tests, 2-tailed.
Probability PI 0.001 0.001 0.001 0.001 0.003 0.005 of variance.
46
C. S.
W.
RAND, J. M. KULDAUand R. L. YOST
between age and narcotic use within each group were not significant. Narcotic use for men and women within each patient group was also statistically comparable. Morbidly obese and cholecystectomy patients did not differ significantly in postoperative use of sedative and hypnotic medications. Other potential analgesic synergists (e.g. tricyclic antidepressant agents) were not used by any patient during the study period. Pre-operative factors and post-operative analgesic use by morbidly obese patients Pre-operative factors accounted for a surprisingly large amount in the variance of post-operative analgesics: 67% of the variance in number of doses Gt,< 0.001) and 69% of the variance in total ME/kg 0, < 0.001); see Table III. Variables indicating psychiatric anxiety and distress, and prior use of medications were entered into the regressions. TAFILEIII.-PARTIAL
CORRELATIONS OF VARIABLESAPPEARINGIN MULTIPLE REGRESSIONS AND ADJUSTED R2 Partial correlations Total doses ME/kg
Variable
MMPI-D (Depression) Scale MMPI-PD (Anti-social) Scale Tranquilizers MMPI-any scale above 80th percentile Number of past surgeries Use of pills to ‘feel better’ PSE item No. lo-nervous tension Prior counseling Adjusted R2
0.54 0.63 0.70 0.74 -0.36 0.38 0.36 0.67
0.33 0.46 0.75 0.68 -0.34
0.50 0.69
DISCUSSION
Morbidly obese patients used significantly less post-operative analgesics than cholecystectomy patients. The possibility that this is due to differences in the directionality of the surgical incision is unlikely: there is no consistent opinion among surgeons as to whether a midline or diagonal (i.e. kocher, subcostal) incision is more painful. Cholecystectomy patients received either a midline or diagonal incision whereas morbidly obese patients had only midline incisions. It is also unlikely that the finding that the morbidly obese patients used less analgesics could be explained either by delayed release of lipophilic agents used during surgery or by less injury to ribs and soft tissues. Halothane, a lipophilic agent, was used during surgery for only one morbidly obese patient, as a supplement to nitrous oxide. Even in high doses, halothane lacks potent analgesic properties. Morbidly obese rather than cholecystectomy patients would be expected to incur greater rib and soft tissue damage because their bulk necessitates increased traction pressure. Several potentially important factors were not studied: We have no evaluations of psychiatric distress or prior drug use among the cholecystectomy patients, thus cannot rule out the possibility that results may be explained by differences along these dimensions. We also did not assess extent or quality of pre-operative counseling by nursing staff about post-operative pain. Finally, although we did evaluate
Obesity and pain
47
comparability of physicians’ prescribing practices, we could not determine if nurses responded differentially to patients’ post-operative pain or if patients were satisfied with medications administered. One of the factors originally prompting us to undertake this study was the repeated comments on the part of the surgeons that obese patients seemed up and about unusually quickly. Future research is needed to address these issues. There was enormous variability within each patient group in amount of postoperative analgesia used. Pre-operative data were only available on the morbidly obese patients, but within this group, a surprisingly large percentage of variability could be accounted for by prior drug use and psychiatric indexes. The positive association between post-operative analgesia and the MMPI-PD score, counseling, and the PSE item nervous tension may reflect an underlying increased anxiety. The MMPI-PD score indicates a low tolerance for frustration, impulsivity, and acting-out, qualities often associated with a readiness to use chemicals. The previous use of tranquilizers and ‘pills to feel better’ probably indicates both a culturally based reliance on the use of medications and actual distress. The inverse relationship between number of prior surgeries and amount of analgesia used was unexpected. A post facto explanation is that prior surgical experiences made the current surgical process familiar and thus less stressful. Our results are consistent with McKendall and Haier’s [l] hypothesis that obese adults have lower pain sensitivities than non-obese adults. McKendall and Haier, however, contrary to their expectations, demonstrated that obese subjects were more sensitive than non-obese subjects to pain produced by placing the index finger in a pressure device bearing a three-pound weight. Pradalier et al. [ 161 also observed hypersensitivity among obese compared to normal weight subjects in response to electrical stimulation of the sural nerve. Although we do not have a satisfactory explanation of why obese patients in our study had lower pain sensitivity while obese subjects in the other two studies had greater pain sensitivity, there are several major differences: (1) Clinical v.s experimental pain. Our study used a clinical paradigm while the other studies were experimental. There are major differences in sources, intensity and time courses of pain. It would be reasonable to expect a more profound response from major surgery than brief finger pressure or sural nerve stimulation. Beecher [17] obtained extensive data demonstrating that clinical pain and experimental pain are not comparable. He noted [9] that both placebos and morphine are more effective in clinical than experimental situations. The anxiety-stress ‘reactive component’ of pain presumably accounts for at least part of this difference. (2) Degree of obesity. Patients in our study were substantially more overweight than subjects in the other two studies. Morbidly obese patients may be biologically ‘different’ from moderately overweight subjects. (3) Food deprivation. Sanger and McCarthy [18] propose that feeding state may be related to endorphin activity. McKendall and Haier [l] suggested that if their obese subjects were operating under a state of food deprivation they might have had lower levels of endorphin activity, hence would be more sensitive to pain. This factor was better controlled in our study: all patients were food deprived during the 24 hr preceding surgery, and were on a carefully controlled (limited) diet for several days post-surgically.
C. S. W. RAND, J. M. KULDAU and R. L. YOST
48
The results of this study underscore the multi-dimensional aspects of pain. Morbidly obese adults requested less post-operative analgesia than cholecystectomy patients. Within the morbidly obese patient group, strong associations were observed between pre-operative measures of psychic distress and acculturation to medications and post-operative analgesic use. The relationship between endogenous opiates, obesity, and pain should be further studied, ideally with a direct measure of the endogenous opiates. REFERENCES 1. MCKENDALL MJ, HAIER RJ. Pain sensitivity and obesity. PsychiatRes 1983; 8: 119-125. 2. LIEBLICH I, GANCHROW JR, BLASS EM, BERGMANN F. Morphine tolerance in genetically selected rats induced by chronically elevated saccharine intake. Science 1983; 221: 871-873. 3. CATLIN DH, HUI KK, LOH HH, LIO CH. /%Endorphin: clinical studies. In Endorphins in Mental Health Research (Edited by USDIN E, BUNNEY WE JR, KLINE NS), pp. 535-544. New York: Oxford University Press, 1979. 4. STERNBACH HV, ANNITTO W, POTTASH ALC, GOLD MS. Anorexic effects of naltrexone in man. Lancet 1982; 1: 388-389. 5. MARGULES DL, LEWIS MJ, SHIBUYA H, PERT CB. /3-Endorphin is associated with overeating in genetically obese mice (ob/ob) and rats (fa/fa). Science 1978; 202: 988-991. 6. KULDAU JM, RAND CSW. Jejunoileal bypass: general and psychiatric outcome after one year.
Psychosomatics 1980; 21: 534-539. MERSKEY H. Pain and personality. In The Psychology of Pain (Edited by STERNBACH RA), pp. 111-117. New York: Raven Press, 1978. 8. CRAIG KD. Social modeling influences on pain. In The Psychology of Pain (Edited by STERNBACH RA), pp. 73-109. New York: Raven Press, 1978. 9. BEECHERHK. Increased stress and effectiveness of placebos and ‘active’ drugs. Science 1960; 132 (8):
7.
91-92. 10. STERNBACH RA.
Pain-A
PsychophysiologicalAnalysis.
New
York: Academic
Press, 1968.
11. JOHNSTONM. Anxiety in surgical patients. Psychof Med 1980; 10: 145-152. 12. JAFFE JH, MARTIN WR. Opioid analgesics and antagonists. In The Pharmacological Basis of Therapeutics (Edited by GILMAN AG, GOODMAN LS, GILMAN A), pp. 494-534. New York: MacMillan Publishing Co. Inc., 1980. 13. REILLY MJ, KEPLER JA (eds). American Hospital Formulury Service. American Society of Hospital Pharmacists, Washington, D.C., 1977. 14. WING JK, COOPER JE, SARTORIUS N. The Measurement and Classification of Psychiatric Symptoms. Cambridge, England: Cambridge University Press, 1974. 15. KORCHIN SJ. Modern ClinicalPsychology, pp. 228-232. New York: Basic Books, 1976. 16. PRADALIER A, WILLER JC, BOUREAU F, DRY J. Relationship between pain and obesity: an electrophysiological study. Physiol Behav 1981; 27: 961-964. 17. BEECHER HK. Measurement of Subjective Responses-Quantitative Effects of Drugs. New York:
Oxford University Press, 1959. 18. SANGER DJ, MCCARTHY PS. Differential deprived
and freely-feeding
rats.
effects
of morphine on food 1980; 72: 103-106.
Psychopharmacology
and water intake in food-
COZZ-3999/85 $3.00+ .OO 0 1985 Pergamon Press Ltd.
OBESITY AND POST-OPERATIVE PAIN COLLEENS. W. RAND, JOHN M. KULDAU and RICHARDL. YOST (Received 2 May 1984; accepted in revisedform 20 July 1984)
Abstract-Patients underwent elective abdominal surgery for morbid obesity (z= 154 kgs, R= 55) or cholecystectomy (x = 71 kgs, n = 54). Post-operative narcotics were transformed into morphine equivalent units (ME). Morbidly obese patients received significantly fewer total doses than cholecystectomy patients and less total mg ME/kg over a five-day period. Sedative use for both groups was comparable. In the morbidly obese patients, preoperative psychiatric and drug usage data predicted 67% of the variance in number of doses and 69% of the variance in total mg ME/kg. INTRODUCTION PAIN usually follows major surgery. Amounts of post-operative analgesia requested by patients, however, can vary enormously. This study examined pre-existing characteristics of morbidly obese patients that may influence post-operative narcotic use: (1) their obesity, and (2) their history of psychiatric distress and previous drug exposure. Endogenous opiates are believed to have a role in regulating pain and controlling eating behaviors [l]. Higher levels of endogenous opiates appear to decrease pain sensitivity. For example, rats with increased release and utilization of endogenous opiates reacted more slowly to pain stimuli than controls [2]. Artificial elevation, through intravenous administration of synthetic beta-endorphins, provided pain relief to cancer patients [3]. Eating and weight regulation are also associated with endogenous opiates. Administration of naltrexone, an opiate antagonist, has led to decreased eating and weight loss among patient groups [4]. Elevated concentrations of beta-endorphins have been found in the pituitaries of genetically obese mice and rats [5]. Morbidly obese adults (i.e. 100 pounds or 100% over ideal weight) may, like the genetically obese rodents, owe at least part of their condition to excessively high amounts of endogenous opiates. We hypothesized that if both eating and pain sensitivity are at least in part under the control of endogenous opiates, then morbidly obese adults should experience less post-surgical pain than normal weight adults. In the course of the investigation testing this hypothesis, we found tremendous variability in analgesia requested by both morbidly obese patients following surgery for obesity and normal weight patients following cholecystectomy. We were able to address additional possibilities to explain this variability within the morbidly obese group using data obtained from another ongoing study of these patients [6]. Unfortunately, similar data were not available for the normal weight patients. Some of the variability in use of post-operative analgesia may be explained by psychic distress [7] and willingness to use drugs [8]. Beecher [9] noted that morphine provides greater pain relief when psychic distress is present. Stembach [lo], discussing
Universityof Florida, Department of Psychiatry and Department of Pharmacy, U.S.A. Correspondence and reprint requests should be addressed to: Colleen S. W. Rand, Ph.D., University Gainesville, Florida, 32610, U.S.A.
ofFlorida, Box J-256, JHMHC,
43
44
C. S. W. RAND, J. M. KULDAU and R. L. YOST
his own work and that of Beecher, proposed that analgesics often have a greater impact on the psychogenic fear-anxiety ‘reaction component’ of pain than the original sensations. Most surgical patients have high levels of anxiety both pre- and postoperatively [l 11. Post-operative analgesics may be even more effective for those patients who have elevated psychiatric distress in addition to ‘normal’ pre-operative anxiety. There are large cultural variations in willingness to use drugs, regardless of pain experienced [8]. One index of willingness is past drug use of any kind (e.g. diet pills, tranquilizers). In addition, past exposure to situations where physicians prescribed drugs may increase the probability of drug use in similar situations (e.g. prior surgeries). Our second hypothesis was that pre-operative measures of psychiatric distress and past use of drugs in morbidly obese patients would be positively related to their post-operative narcotic use. METHOD A random selection was made of patients who underwent elective abdominal surgery at the University of Florida between 1974 and 1980: jejunoileal bypass for morbid obseity (n = 55) or cholecystectomy (n=54). Each patient’s chart was reviewed to determine analgesic and sedative use for the first five post-operative days. To standardize analgesic use, all total daily doses were converted to a reference standard dose (ME) of morphine sulfate administered subcutaneously [12].* To control for the marked group differences in body weight, two measures of post-operative narcotic use were selected: total number of doses, which ignores differences in body weight, and ME/kg body weight, where body weight is taken into account. Sedative agents were simply noted as present or absent for the study days. Sedative agents included agents listed in categories 4:00, 28:16:08, and 28:04 in the American Hospital Formulary Service [13]. The pre-operative medications used for both groups were similar. The regimen generally consisted of a narcotic analgesic (morphine or meperidine), atropine, and/or a sedative (hydroxyzine or diazepam). Maintenance anesthesia usually was comprised of nitrous oxide, thiopental sodium, a skeletal muscle relaxant, and a narcotic (morphine or fentanyl). Volatile anesthetic agents such as halothane, enfluane, or methoxyflurane were given to three morbidly obese and 17 cholecystectomy patients. The surgical service at Shands Teaching Hospital in Gainesville, Florida, requests a psychiatric consultation for patients hospitalized for surgery for obesity. Systematic data collection was initiated in 1976 as part of a longitudinal prospective study of these patients [6]. The longitudinal project provided pre-operative psychiatric and psychosocial data on 32 of the morbidly obese patients in our study. These 32 patients were interviewed with a semistructured format. Medical history and several areas of social functioning were reviewed and a scoreable standardized mental status schedule, the Present State Examination (PSE), was included [14]. Patients also completed the Minnesota Multiphasic Personality Inventory (MMPI) [15]. Variables selected for this study were limited to a few demographic statistics and items indicating anxiety, distress, and exposure to medications: prior counseling, feelings of loneliness, PSE item No. 10 nervous tension, PSE levels of definition [6], all scales in the MMPI, and MMPI scales with standardized scores of at least 80; binge eating, night eating syndrome, and smoking; prior use of sedatives, sleeping pills, tranquilizers, amphetamines, antidepressants, diet pills, and pills to ‘feel better’; general health, history of hospitalizations, and prior surgeries; age, sex, and social class. Patient groups were compared using multivariate analysis of variance, r-test, and chi-square analyses. Morbidly obese patients’ pre-operative data were analyzed using stepwise multiple regression and partial correlations. RESULTS
Sample characteristics The patient groups were compared on the following characteristics: age, sex, weight, previous surgeries and physicians’ prescribing characteristics (see Table I). *Amount of analgesia equivalent to 10 ME, administered orally (PO) and intramuscularly 75 mg Meperidine IM; 45 mg Pentazocine IM; 120 mg Codeine IM; 240 mg Codeine PO; Hydromorphone IM.
(IM): 1.5 mg
45
Obesity and pain TABLE I.-PATIENT
Cholecystectomy Patients (n = 54)
Variable
Age Sex Weight at admission Total previous surgeries
44.402 15.80yr 12 men 42 women 70.92 + 17.95 Kgs 1.85 + 1.71
CHARACTERISTICS
Morbidly obese Patients (n = 55)
Probability P<
33.20 + 9.03 yr 24 men 31 women 154.46? 36.84 kgs 1.29+ 1.30
0.001* 0.05** 0.001* 0.10*
*t-test, 2-tail. **chi-square.
Morbidly obese patients were significantly younger than cholecystectomy patients @ I O.OOl), significantly heavier (p I O.OOl), and more often women (p < 0.05). The number of past surgeries was the same. Although there was variability over the six years in physicians’ prescribing practices regarding type of analgesia, quantity, and dosing rate, there were no systematic differences in physicians’ orders for the two patient groups. Morbidly obese and cholecystectomy patient comparisons Morbidly obese patients received fewer total doses than cholecystectomy patients (F=9.18 compared to 13.13;p(O.O1), and lower total ME/kg (x=0.58 compared to 1.57; p < 0.001); see Table II. These results could not be accounted for by age or sex differences between the groups. The same levels of significance were maintained for both total dose and ME/kg when an age matched subsample of 35 patients from each group was selected; pairs within each age were randomly matched. Correlations TABLE HA.-POST-SURGICAL NARCOTICUSE: NUMBEROF DOSESAND MORPHINEEQUIVALENT UNITS PER KILOGRAM(ME/KG)
Narcotic
use
Total doses Day 1 Day 2 Day 3 Day 4 Day 5
Cholecystectomy Patients (n = 54) 13.13 + 7.95 3.78? 1.96 3.15k2.10 2.57 f 2.02 1.67& 1.89 1.01 f 1.60
Morbidly obese Patients (n = 55) 9.18 2 2.56 * 2.35 + 1.84? 1.15 + 0.55 +
7.43 1.72 2.09 1.99 1.67 1.03
Probability PS 0.05 0.001 0.05 0.06 NS 0.10
TABLE IIB.-POST-SURGICAL NARCOTICUSE: NUMBER OF DOSESAND MORPHINEEQUIVALENT UNITS PER KILOGRAM(ME/KG)
Narcotic
use
Total ME/Kg Day 1 Day 2 Day 3 Day 4 Day 5
Cholecystectomy Patients (n = 54)
Morbidly obese Patients (n = 55)
1.57 + 1.21 0.53 * 0.35 0.40 * 0.35 0.27 ? 0.27 0.17kO.26 0.08 +0.15
0.58 f 0.48 0.18 +0.13 0.17 +0.16 0.11+0.14 0.05 & 0.08 0.02 + 0.04
Total doses and total ME/Kg were analyzed using multivariate analyses Individual comparisons by day were then analyzed using r-tests, 2-tailed.
Probability PI 0.001 0.001 0.001 0.001 0.003 0.005 of variance.
46
C. S.
W.
RAND, J. M. KULDAUand R. L. YOST
between age and narcotic use within each group were not significant. Narcotic use for men and women within each patient group was also statistically comparable. Morbidly obese and cholecystectomy patients did not differ significantly in postoperative use of sedative and hypnotic medications. Other potential analgesic synergists (e.g. tricyclic antidepressant agents) were not used by any patient during the study period. Pre-operative factors and post-operative analgesic use by morbidly obese patients Pre-operative factors accounted for a surprisingly large amount in the variance of post-operative analgesics: 67% of the variance in number of doses Gt,< 0.001) and 69% of the variance in total ME/kg 0, < 0.001); see Table III. Variables indicating psychiatric anxiety and distress, and prior use of medications were entered into the regressions. TAFILEIII.-PARTIAL
CORRELATIONS OF VARIABLESAPPEARINGIN MULTIPLE REGRESSIONS AND ADJUSTED R2 Partial correlations Total doses ME/kg
Variable
MMPI-D (Depression) Scale MMPI-PD (Anti-social) Scale Tranquilizers MMPI-any scale above 80th percentile Number of past surgeries Use of pills to ‘feel better’ PSE item No. lo-nervous tension Prior counseling Adjusted R2
0.54 0.63 0.70 0.74 -0.36 0.38 0.36 0.67
0.33 0.46 0.75 0.68 -0.34
0.50 0.69
DISCUSSION
Morbidly obese patients used significantly less post-operative analgesics than cholecystectomy patients. The possibility that this is due to differences in the directionality of the surgical incision is unlikely: there is no consistent opinion among surgeons as to whether a midline or diagonal (i.e. kocher, subcostal) incision is more painful. Cholecystectomy patients received either a midline or diagonal incision whereas morbidly obese patients had only midline incisions. It is also unlikely that the finding that the morbidly obese patients used less analgesics could be explained either by delayed release of lipophilic agents used during surgery or by less injury to ribs and soft tissues. Halothane, a lipophilic agent, was used during surgery for only one morbidly obese patient, as a supplement to nitrous oxide. Even in high doses, halothane lacks potent analgesic properties. Morbidly obese rather than cholecystectomy patients would be expected to incur greater rib and soft tissue damage because their bulk necessitates increased traction pressure. Several potentially important factors were not studied: We have no evaluations of psychiatric distress or prior drug use among the cholecystectomy patients, thus cannot rule out the possibility that results may be explained by differences along these dimensions. We also did not assess extent or quality of pre-operative counseling by nursing staff about post-operative pain. Finally, although we did evaluate
Obesity and pain
47
comparability of physicians’ prescribing practices, we could not determine if nurses responded differentially to patients’ post-operative pain or if patients were satisfied with medications administered. One of the factors originally prompting us to undertake this study was the repeated comments on the part of the surgeons that obese patients seemed up and about unusually quickly. Future research is needed to address these issues. There was enormous variability within each patient group in amount of postoperative analgesia used. Pre-operative data were only available on the morbidly obese patients, but within this group, a surprisingly large percentage of variability could be accounted for by prior drug use and psychiatric indexes. The positive association between post-operative analgesia and the MMPI-PD score, counseling, and the PSE item nervous tension may reflect an underlying increased anxiety. The MMPI-PD score indicates a low tolerance for frustration, impulsivity, and acting-out, qualities often associated with a readiness to use chemicals. The previous use of tranquilizers and ‘pills to feel better’ probably indicates both a culturally based reliance on the use of medications and actual distress. The inverse relationship between number of prior surgeries and amount of analgesia used was unexpected. A post facto explanation is that prior surgical experiences made the current surgical process familiar and thus less stressful. Our results are consistent with McKendall and Haier’s [l] hypothesis that obese adults have lower pain sensitivities than non-obese adults. McKendall and Haier, however, contrary to their expectations, demonstrated that obese subjects were more sensitive than non-obese subjects to pain produced by placing the index finger in a pressure device bearing a three-pound weight. Pradalier et al. [ 161 also observed hypersensitivity among obese compared to normal weight subjects in response to electrical stimulation of the sural nerve. Although we do not have a satisfactory explanation of why obese patients in our study had lower pain sensitivity while obese subjects in the other two studies had greater pain sensitivity, there are several major differences: (1) Clinical v.s experimental pain. Our study used a clinical paradigm while the other studies were experimental. There are major differences in sources, intensity and time courses of pain. It would be reasonable to expect a more profound response from major surgery than brief finger pressure or sural nerve stimulation. Beecher [17] obtained extensive data demonstrating that clinical pain and experimental pain are not comparable. He noted [9] that both placebos and morphine are more effective in clinical than experimental situations. The anxiety-stress ‘reactive component’ of pain presumably accounts for at least part of this difference. (2) Degree of obesity. Patients in our study were substantially more overweight than subjects in the other two studies. Morbidly obese patients may be biologically ‘different’ from moderately overweight subjects. (3) Food deprivation. Sanger and McCarthy [18] propose that feeding state may be related to endorphin activity. McKendall and Haier [l] suggested that if their obese subjects were operating under a state of food deprivation they might have had lower levels of endorphin activity, hence would be more sensitive to pain. This factor was better controlled in our study: all patients were food deprived during the 24 hr preceding surgery, and were on a carefully controlled (limited) diet for several days post-surgically.
C. S. W. RAND, J. M. KULDAU and R. L. YOST
48
The results of this study underscore the multi-dimensional aspects of pain. Morbidly obese adults requested less post-operative analgesia than cholecystectomy patients. Within the morbidly obese patient group, strong associations were observed between pre-operative measures of psychic distress and acculturation to medications and post-operative analgesic use. The relationship between endogenous opiates, obesity, and pain should be further studied, ideally with a direct measure of the endogenous opiates. REFERENCES 1. MCKENDALL MJ, HAIER RJ. Pain sensitivity and obesity. PsychiatRes 1983; 8: 119-125. 2. LIEBLICH I, GANCHROW JR, BLASS EM, BERGMANN F. Morphine tolerance in genetically selected rats induced by chronically elevated saccharine intake. Science 1983; 221: 871-873. 3. CATLIN DH, HUI KK, LOH HH, LIO CH. /%Endorphin: clinical studies. In Endorphins in Mental Health Research (Edited by USDIN E, BUNNEY WE JR, KLINE NS), pp. 535-544. New York: Oxford University Press, 1979. 4. STERNBACH HV, ANNITTO W, POTTASH ALC, GOLD MS. Anorexic effects of naltrexone in man. Lancet 1982; 1: 388-389. 5. MARGULES DL, LEWIS MJ, SHIBUYA H, PERT CB. /3-Endorphin is associated with overeating in genetically obese mice (ob/ob) and rats (fa/fa). Science 1978; 202: 988-991. 6. KULDAU JM, RAND CSW. Jejunoileal bypass: general and psychiatric outcome after one year.
Psychosomatics 1980; 21: 534-539. MERSKEY H. Pain and personality. In The Psychology of Pain (Edited by STERNBACH RA), pp. 111-117. New York: Raven Press, 1978. 8. CRAIG KD. Social modeling influences on pain. In The Psychology of Pain (Edited by STERNBACH RA), pp. 73-109. New York: Raven Press, 1978. 9. BEECHERHK. Increased stress and effectiveness of placebos and ‘active’ drugs. Science 1960; 132 (8):
7.
91-92. 10. STERNBACH RA.
Pain-A
PsychophysiologicalAnalysis.
New
York: Academic
Press, 1968.
11. JOHNSTONM. Anxiety in surgical patients. Psychof Med 1980; 10: 145-152. 12. JAFFE JH, MARTIN WR. Opioid analgesics and antagonists. In The Pharmacological Basis of Therapeutics (Edited by GILMAN AG, GOODMAN LS, GILMAN A), pp. 494-534. New York: MacMillan Publishing Co. Inc., 1980. 13. REILLY MJ, KEPLER JA (eds). American Hospital Formulury Service. American Society of Hospital Pharmacists, Washington, D.C., 1977. 14. WING JK, COOPER JE, SARTORIUS N. The Measurement and Classification of Psychiatric Symptoms. Cambridge, England: Cambridge University Press, 1974. 15. KORCHIN SJ. Modern ClinicalPsychology, pp. 228-232. New York: Basic Books, 1976. 16. PRADALIER A, WILLER JC, BOUREAU F, DRY J. Relationship between pain and obesity: an electrophysiological study. Physiol Behav 1981; 27: 961-964. 17. BEECHER HK. Measurement of Subjective Responses-Quantitative Effects of Drugs. New York:
Oxford University Press, 1959. 18. SANGER DJ, MCCARTHY PS. Differential deprived
and freely-feeding
rats.
effects
of morphine on food 1980; 72: 103-106.
Psychopharmacology
and water intake in food-