- Email: [email protected]
Changes in taste preference after colorectal surgery: A longitudinal study
Clinical Nutrition xxx (2014) 1e4
Contents lists available at ScienceDirect
Clinical Nutrition journal homepage: http://www.elsevier.com/locate/clnu
Original article
Changes in taste preference after colorectal surgery: A longitudinal study Sophie Welchman a, Perryhan Hiotis a, Steven Pengelly a, Georgina Hughes d, Jason Halford d, Paul Christiansen d, Stephen Lewis b, c, * a
Dept of Surgery, Derriford Hospital, Plymouth, Devon, UK Dept Gastroenterology, Derriford Hospital, Plymouth, Devon, UK National Institute of Health Research Bristol Biomedical Research Centre, UK d Dept of Psychological Sciences, University of Liverpool, Merseyside, UK b c
a r t i c l e i n f o
s u m m a r y
Article history: Received 4 June 2014 Accepted 15 September 2014
Background & aims: Nutrition is a key component of surgical enhanced recovery programmes. However, alterations in food preferences are often reported as reasons for patients not eating in the early postoperative period. We hypothesised that taste preferences are altered in the early postoperative period and this dysgeusia affects patients' food choices during this critical time. Methods: This is a longitudinal study looking at taste preferences of patients recovering from surgery. Patients undergoing colonic resections were recruited. Using visual analogue scales participants completed a questionnaire, taste tests and preference scoring of food images for the 6 groups of taste (bitter, salty, savoury, sour, spicy and sweet) preoperatively and on postoperative days 1e3. Patients were also offered snacks postoperatively, which represented foods from the six groups and consumption was measured. Differences from baseline were assessed using the Friedman's and Wilcoxon tests. Results: 31 patients were studied. In the immediate postoperative period participants reported deterioration in their sense of taste (p 0.001), increased nausea (p < 0.001) and hunger (p ¼ 0.03). Sweet, savoury and spicy tastes were the most popular during the perioperative period. However, only palatability for salty taste increased (p ¼ 0.001) following surgery. The highest rated images were for savoury food with only the ratings for salty food increasing after surgery (p < 0.05). These findings concurred with the sweet, savoury and salty snacks being the most consumed foods in the postoperative period. Bitter, sour and spicy foods were the least frequently consumed. Discussion: This is the first study to investigate postsurgical patients' food preferences. A consistent change in all the individual tastes with the exception of salty in the postoperative period was observed. The most desirable tastes were for savoury and sweet, reflecting patients' preoperative preferences. An improved understanding of taste may improve the resumption of eating after colonic surgery. © 2014 Elsevier Ltd and European Society for Clinical Nutrition and Metabolism.
Keywords: Post-operative Nutrition Enhanced recovery Dysgeusia
1. Introduction Current surgical practice encourages patients to resume eating rapidly after their operations. Nutrition within twenty-four hours of colorectal surgery is associated with improved outcome and shorter postoperative hospital stay [5,1]. Early postoperative nutritional intake is a key component of enhanced recovery programmes (ERP), however, for a variety of reasons post-operative
* Corresponding author. Dept of Gastroenterology, Derriford Hospital, Plymouth, Devon PL6 8DH, UK. E-mail address: [email protected] (S. Lewis).
nutritional intake is often poor [6]. Dysguesia, loss of hunger and alterations in food palatability are often reported. Surgery induces a catabolic response with the release of stress hormones (glucagon, cortisol, catecholamines) and inflammatory mediators [7] many of which are known to suppress appetite. These responses, with the loss of insulin sensitivity (insulin resistance) [7,16], are claimed by some to be dysfunctional. In addition intestinal dysfunction (ileus) is common after certain types of surgery. Pain, drugs and environment may also influence a postoperative patient's appreciation of food. The palatability of a food by an individual has both chemical and psychological components [3], the perceived palatability being a combination of the food's taste, smell, texture, temperature, visual
http://dx.doi.org/10.1016/j.clnu.2014.09.017 0261-5614/© 2014 Elsevier Ltd and European Society for Clinical Nutrition and Metabolism.
Please cite this article in press as: Welchman S, et al., Changes in taste preference after colorectal surgery: A longitudinal study, Clinical Nutrition (2014), http://dx.doi.org/10.1016/j.clnu.2014.09.017
2
S. Welchman et al. / Clinical Nutrition xxx (2014) 1e4
appearance, sound and irritative sensations [19,15]. Generally, increased palatability results in increased oral intake [12]. This may work both through enhanced hedonic experience and a delay in within-meal satiation (the natural inhibitory feedback generated through meal consumption). Loss of taste and smell as a result of direct damage to the taste and olfactory system (chemotherapy, radiotherapy or maxillofacial surgery) are well described and associated with reduced energy intake [17,8,9]. Non-specific changes in taste appreciation are also described in many other medical conditions [11,10,13,2]. Alteration of food palatability as a result of coexistent stresses such as abdominal surgery, are not well characterised and the mechanisms unclear. We hypothesised that taste preferences and desires are altered in the early post-operative period and this affects patients' food choices during this critical time. We set out to describe changes in perioperative taste and food preferences with a view to being able to optimise food choices and thus improve intake. To do this we recorded patients' experience of and preference for individual tastes (bitter, salty, savoury, sour, spicy and sweet) in three separate experiments designed to assess distinct components of food palatability and preference. 2. Materials and methods 2.1. Subjects Patients undergoing a colorectal resection at Derriford Hospital, Plymouth, were eligible for inclusion in this study. Exclusion criteria were: age below 18 years, inability to give informed consent, any changes in taste over preceding six months, participation in another trial, pregnancy, diabetes, steroids or immunosuppressant use, abnormal gastric emptying, intestinal obstruction, concurrent nutritional support, taking medication known to alter taste or being a smoker. Written informed consent was obtained. Postoperatively participants who developed clinical complications requiring cessation of oral intake were withdrawn from the study. The study was approved by Plymouth Local Research Ethics Committee. 2.2. Study design All study participants were managed under the care of their attending surgeon within an ‘Enhanced Recovery’ protocol (ERP). As such patients were encouraged to eat immediately after surgery. Baseline demographic and operative data were recorded. Postoperatively clinical complications were noted. Participants deemed unable to eat in the opinion of their attending surgeon immediately after surgery were excluded from the study. Participants were asked to complete a questionnaire and three experimental protocols to assess changes in patients' general appetite, any changes in taste palatability and desire as well as actual taste preference. Changes were assessed using visual analogue scales (VAS), which have been widely used in evaluation of appetite [18]. These protocols were done at their preoperative assessment appointment and then on postoperative days (POD) 1, 2 and 3 at 18:00 before the evening meal was served when patients would normally be hungry. 2.3. Questionnaire Clinical data were recorded. Current medications including antiemetics were noted. To assess overall perception of their hunger, nausea, thirst and sense of taste, visual analogue scales (VAS, 0e100%) were completed.
2.4. Protocol 1 To assess any change in the palatability over the perioperative period dilutions of the six different tastes were scored by participants. These were added into a neutral flavoured ‘balanced’ polymeric energy drink (Fresubin® 2 Kcal/ml, Fresenius-Kabi, Runcorn, Cheshire, UK) to associate specific taste modalities with caloric intake. Patients were asked to rate the taste of six flavoured energy dense liquids pre-operatively and on POD 1, 2 and 3. The liquids were flavoured with concentrations of the six core tastes (adapted from Steingruber et al. [14] and adjusted by the research team such that the final strength of taste was similar (bitter (quinine sulphate 0.01 g ml1), salty (sodium chloride 0.05 g ml1), savoury (unami 0.005 g ml1), sour (citric acid 0.01 g ml1), spicy (chilli powder 0.005 g ml1) and sweet (glucose 0.05 g ml1)). The solutions were prepared daily in plastic beakers and labelled 1e6. The labelling allocation of taste to code (1e6) was random and held on a preprepared spreadsheet. Both participants and researchers were blinded to which taste was being served. A ‘mouth care’ sponge was dipped in each flavoured liquid and patients were asked to taste and score each taste using a VAS (Horrible-Nice: 0e100%). Participants rinsed their mouths with water after each taste and allowed at least 5 min between each tasting. 2.5. Protocol 2 To examine participants desire to eat food items representative of the core tastes: bitter (grapefruit), salty (popcorn), savoury (cheese on toast), sour (gherkins), spicy (curry) and sweet (sugar mice: mice shaped from icing sugar, egg whites and golden syrup), patients were shown photographs of food items (presented on an identical plain white plate and neutral background). They were asked to rate each of them in terms of desire to consume at the time using a VAS scale (Not appetizing-very appetizing: 0e100%). 2.6. Protocol 3 To examine what foods patients would actually eat we gave patients a box (at 18:00) containing individually packaged (80 cm3 clear plastic container) snacks incorporating a small amount of foods representative of the six core tastes (bitter (dark chocolate), salty (pretzels), savoury (cheese), sour (gherkins), spicy (Bombay mix: a commonly available dried mix of fried pulses, peanuts, corn, flaked rice and curry leaves) and sweet (fudge)). Patients were asked to eat any of the snacks they wished over the next 24 h, leaving the empty container of eaten or partially consumed snacks within the box. The boxes were collected 24 h later and note was made of the items fully consumed. 2.7. Sample size and statistical analysis The study was descriptive and as no previous data were available to inform sample size we decided that thirty patients completing the study would provide a clinically meaningful number. Statistical analysis was done by a researcher that was blinded to allocation (SL, PC) using SPSS 21 (IBM, Chicago, Illinois, USA). Data were non-parametric thus presented as median values with interquartile ranges. Differences in taste perception across the four assessment points (baseline and POD 1e3) were examined using Friedman tests, with significant changes explored further with Wilcoxon rank sum tests to examine differences between scores at baseline and POD 1-3 (Wilcoxon tests are presented as p values with 95% confidence intervals).
Please cite this article in press as: Welchman S, et al., Changes in taste preference after colorectal surgery: A longitudinal study, Clinical Nutrition (2014), http://dx.doi.org/10.1016/j.clnu.2014.09.017
S. Welchman et al. / Clinical Nutrition xxx (2014) 1e4
3
3. Results A total of 36 patients were recruited, of which five participants were excluded owing to postoperative complications making them unable to eat (four participants) or withdrawal of consent (one participant). Baseline demographics of the 31 participants who completed the study are presented in Table 1. There were no significant clinical complications or the use of intravenous fluids after the operative day in the remaining participants. 3.1. Questionnaire Visual analogue scores (VAS, 0e100%) for the questionnaires asked about the participants' sense of taste, whether food tasted normal, hunger, thirst and nausea are presented in Fig. 1. Friedman tests showed changes for taste (p < 0.001), hunger (p ¼ 0.003) and nausea (p < 0.001). No changes in VAS scores for thirst were seen. At baseline all participants reported a VAS score of 0% for nausea and 97% scores of 100% for their sense of taste. Planned comparisons showed scores for taste decreased from baseline compared to postoperative days (POD) 1 (p < 0.001, 45, 20), 2 (p < 0.001, 35, 15) and 3 (p < 0.001, 15, 5), with scores being lowest on POD 1 and slowly returning towards preoperative levels. Reported hunger scores were greater at POD 3 (p ¼ 0.033, 5, 30) compared to baseline. Nausea scores were greater than baseline on POD 1 (p < 0.001, 10, 20), POD 2 (p ¼ 0.004, 0, 5), POD 3 (p ¼ 0.022, 0, 5).
Fig. 1. Questionnaire, Change in sense of taste; does food taste normally, hunger, thirst and nausea across the perioperative period, median (IQ range).
3.2. Protocol 1 Changes in the VAS scores (Horrible-Nice: 0e100%) for the palatability of the six tastes are presented in order of baseline VAS score (lowest score on left to highest on right hand side) in Fig. 2. Scores for bitter were extremely low throughout, whilst savoury, spicy and sweet were the highest. Only scores for salty increased across time points (p < 0.001), with scores being raised 50% from baseline on POD 1 (p < 0.001, 5, 20), POD 2 (p < 0.001, 10, 20) and POD 3 (p < 0.001, 5, 15). 3.3. Protocol 2 The VAS scores (Not appetizing-very appetizing: 0e100%) of how appetizing the foods in the photographs looked are given in Fig. 3 (the least appetizing foods on the left and most on the right hand side). Friedman tests revealed changes in scores for sour (p ¼ 0.009), salty (p ¼ 0.025), bitter (p ¼ 0.001) and savoury (p ¼ 0.047). The savoury food was the most appetizing. Only salty VAS scores increased postoperatively (POD 1, p ¼ 0.03, 0, 20), normalising by POD 2. Scores for sour (POD 1, p ¼ 0.013, 20, 0, POD 2, p ¼ 0.009, 15, 0) and spicy (POD 1, p ¼ 0.025, 25, 0 POD 2, p ¼ 0.05, 25, 0) initially fell but were no different from baseline by POD 3. Similarly scores for bitter (POD 1, p ¼ 0.002, 30, 10) and savoury (POD 1, p ¼ 0.048, 15, 0) all initially fell before recovering by POD 2.
Fig. 2. Protocol 1, Change in taste preference across the perioperative period, median (IQ range).
3.4. Protocol 3 The snack item that was most frequently eaten (fully consumed) postoperatively by patients was the salty-pretzels (POD1-2, 32%, POD2-3 32%) and sweet-fudge (POD1-2 32%, POD2-3 39%) followed by savoury-cheese (POD1-2, 26%, POD2-3 29%). Few people consumed the other tastes (gherkins, dark chocolate or Bombay mix). Due to the small sample size and few patients consuming food from the snack box no statistical description of these results is offered. 4. Discussion We have demonstrated subjective and objective change in participants' description of taste and food preference, or dysgeusia, across the perioperative period. To the authors' knowledge this is
Table 1 Baseline Demographics of 31 patients completing the study. Demographics Median age, years (IQ range) Median BMI Kg m1 (IQ range) Indication for resection Inflammatory bowel disease Benign polyp Operation Right hemicolectomy Left hemicolectomy
76 (68, 81) 28 (22, 33)
Male:Female ratio
14:17
2 4
Malignancy Diverticular disease
24 1
12 6
Anterior Resection Total colectomy
12 1
Fig. 3. Protocol 2, Preoperative food image ‘appetizing’ score, median (IQ range).
Please cite this article in press as: Welchman S, et al., Changes in taste preference after colorectal surgery: A longitudinal study, Clinical Nutrition (2014), http://dx.doi.org/10.1016/j.clnu.2014.09.017
4
S. Welchman et al. / Clinical Nutrition xxx (2014) 1e4
the first report of alterations in patient taste preferences and food choices in the perioperative period excluding surgery to the oropharyngeal region. Postoperatively participants' reported a reduced sense of taste, increased nausea and hunger. Participants preferred savoury, spicy and sweet flavoured supplement drink. Although participants' satisfaction from a salty taste was not as high, this increased in the postoperative period. Participants' theoretical desire to eat different tasting foods (photographs) after surgery was generally low, with savoury tasting food being the most desirable. Again the only food type to increase in desirability was salty. In practice participants predominantly ate salty, sweet and savoury snacks. There was internal consistency of trends across the three strands of the study with sweet and savoury tastes being the most popular and only salty taste increasing significantly. The first experiment suggested that participants' appreciation of the different tastes apart from salty did not change from baseline in the postoperative period. However, the desirability to eat foods of different taste often reduced (apart from salty). The absolute ranking of the desirability of the different tastes at any given postoperative day (POD) seem to reflect their ranking at baseline. As expected the measured values had started to return towards baseline by POD 3. Following surgery the body's stress response results in salt retention in order to retain fluid through activation of the reninangiotensin system. However, humans do not have an easily demonstrable sodium or salt appetite in response to sodium loss or cortisol administration [4]. This may well explain the increase in desire for salty taste. Salt is also a well-recognised taste enhancer. All studies of alterations in taste perception or preference are limited by the complexity of this sense and the variety of factors that can impact on it. In reality all foods are a combination of multiple tastes and other factors such as presentation, texture, temperature, smell and food groups which all contribute to the enjoyment of eating. Thus focussing on the six core tastes is simplistic. Similarly a patient's experience of recovery from operative stress will be governed by a multitude of factors both physical and psychological for which control was not possible. It would be difficult to define a true control group and so taste changes are compared against baseline instead. This may be confounded by the ‘baseline’ measurements not reflecting the patients ‘normal’, given they were done in a hospital environment and the associated anxieties the preoperative patient would be experiencing. The bitter, sour and spicy tastes depending on their strength can be very unpalatable. Their presence in people's diets will be highly variable. Likewise specific food items may or may not be part of patients' diets. For example some people may have little experience of eating pickled gherkins. We found that photographs for bitter tasting grapefruit were popular, whilst the taste and snacks were not. Foods like grapefruit often taste tart or bitter but some varieties can taste sweet; participants' previous experiences or lack of them may have been reflected in their scoring of that taste. At a time when patients are unwell there may be a desire to stick to less adventurous tastes. Early post-operative nutrition improves surgical outcome and is an essential element of modern surgical practice [5]. Previous work has shown postoperative patients prefer simple solid foods to clear fluids [20]. We know that the early postoperative period nutritional intake is often poor [6], thus understanding taste preferences and changes from usual food choices are vital in guiding hospitals to design menus that encourage patients to eat during this period. Our study suggests that sweet and savoury foods may be the most palatable to the postoperative patient. As desire to eat salty food
increases, there may be a role for modest amounts of salt or ‘low sodium’ salt as a taste enhancer. It is clear from this study that patients taste preferences change following surgery. In the context of a clinical drive to restore oral intake in the immediate postoperative period this is an important observation. More work in this field is required to fully define these changes and to investigate the impact of meal options on the time to established diet and calorie intake. Likewise whilst popular liquid nutritional supplements come in many flavours, there is less variety in terms of tastes (bitter to sweet). It is possible that a better understanding of this subject could result in a tailored postoperative menu that offers patients the tastes and foods that they are most likely to eat. This may help to provide patients with the energy and nutrition that they need to support healing, restore bowel function and global recovery. On the other hand, it may be that posttraumatic anorexia, as observed in the animal world, is a healthy physiological response and the way ahead to introduce early enteral feeding is to reduce the inflammatory insult rather than change the taste of the food. Conflict of interest None declared. References [1] Andersen HK, Lewis SJ, Thomas S. Early enteral nutrition within 24 h of colorectal surgery versus later commencement of feeding for postoperative complications. Cochrane Database 2011:CD004080. [2] Depoortere I. Taste receptors of the gut: emerging roles in health and disease. Gut 2014;63(1):179e90. [3] Farmer MN, Raddin RS, Roberts JD. The relationship between taste, olfaction, and nutrition in the cancer population. J Support Oncol 2009;7(2):70e2. [4] Leshem M. The excess salt appetite of humans is not due to sodium loss in adulthood. Physiol Behav 2009;98(3):331e7. [5] Lewis SJ, Egger M, Sylvester PA, Thomas S. Early enteral feeding versus “nil by mouth” after gastrointestinal surgery: systematic review and meta-analysis of controlled trials. Br Med J 2001;323:773e6. [6] Lidder P, Thomas S, Fleming S, Hosie K, Shaw S, Lewis S. A randomised placebo controlled trial of preoperative carbohydrate drinks and early postoperative nutritional supplement drinks in colorectal surgery. Colorectal Dis Off J Assoc Coloproctol G B Irel 2013. [7] Ljungqvist O, Soreide E. Preoperative fasting. Br J Surg 2003;90:400e6. [8] Nguyen HM, Reyland ME, Barlow LA. Mechanisms of taste bud cell loss after head and neck irradiation. J Neurosci Off J Soc Neurosci 2012;32(10):3474e84. [9] Porter SR, Fedele S, Habbab KM. Taste dysfunction in head and neck malignancy. Oral Oncol 2010;46(6):457e9. [10] Schiffman SS. Taste and smell in disease (first of two parts). N Engl J Med 1983;308(21):1275e9. [11] Schiffman SS. Taste and smell in disease (second of two parts). N Engl J Med 1983;308(22):1337e43. [12] Sorensen LB, Moller P, Flint A, Martens M, Raben A. Effect of sensory perception of foods on appetite and food intake: a review of studies on humans. Int J Obes Relat Metab Disord J Int Assoc Study Obes 2003;27(10): 1152e66. [13] Spielman AI. Chemosensory function and dysfunction. Crit Rev Oral Biol Med An Off Publ Am Assoc Oral Biol 1998;9(3):267e91. € hme C. Development of an assessment instru[14] Steingruber R, Gillessen S, Bo ment for chemotherapy associated dysgeusia and its implications for patients. Eur J Cancer 2005;3:446. [15] Stroebele N, De Castro JM. Effect of ambience on food intake and food choice. Nutrition 2004;20(9):821e38. [16] Svanfeldt M, Thorell A, Hausel J, Soop M, Rooyackers O, Nygren J, et al. Randomized clinical trial of the effect of preoperative oral carbohydrate treatment on postoperative whole-body protein and glucose kinetics. Br J Surg 2007;94(11):1342e50. [17] van Dam FS, Hilgers FJ, Emsbroek G, Touw FI, van As CJ, de Jong N. Deterioration of olfaction and gustation as a consequence of total laryngectomy. Laryngoscope 1999;109(7 Pt 1):1150e5. [18] Whybrow S, Stephen JR, Stubbs RJ. The evaluation of an electronic visual analogue scale system for appetite and mood. Eur J Clin Nutr 2006;60(4): 558e60. [19] Yeomans MR. Taste, palatability and the control of appetite. Proc Nutr Soc 1998;57(4):609e15. [20] Yeung SE, Fenton TR. Colorectal surgery patients prefer simple solid foods to clear fluids as the first postoperative meal. Dis Colon Rectum 2009;52(9): 1616e23.
Please cite this article in press as: Welchman S, et al., Changes in taste preference after colorectal surgery: A longitudinal study, Clinical Nutrition (2014), http://dx.doi.org/10.1016/j.clnu.2014.09.017
Contents lists available at ScienceDirect
Clinical Nutrition journal homepage: http://www.elsevier.com/locate/clnu
Original article
Changes in taste preference after colorectal surgery: A longitudinal study Sophie Welchman a, Perryhan Hiotis a, Steven Pengelly a, Georgina Hughes d, Jason Halford d, Paul Christiansen d, Stephen Lewis b, c, * a
Dept of Surgery, Derriford Hospital, Plymouth, Devon, UK Dept Gastroenterology, Derriford Hospital, Plymouth, Devon, UK National Institute of Health Research Bristol Biomedical Research Centre, UK d Dept of Psychological Sciences, University of Liverpool, Merseyside, UK b c
a r t i c l e i n f o
s u m m a r y
Article history: Received 4 June 2014 Accepted 15 September 2014
Background & aims: Nutrition is a key component of surgical enhanced recovery programmes. However, alterations in food preferences are often reported as reasons for patients not eating in the early postoperative period. We hypothesised that taste preferences are altered in the early postoperative period and this dysgeusia affects patients' food choices during this critical time. Methods: This is a longitudinal study looking at taste preferences of patients recovering from surgery. Patients undergoing colonic resections were recruited. Using visual analogue scales participants completed a questionnaire, taste tests and preference scoring of food images for the 6 groups of taste (bitter, salty, savoury, sour, spicy and sweet) preoperatively and on postoperative days 1e3. Patients were also offered snacks postoperatively, which represented foods from the six groups and consumption was measured. Differences from baseline were assessed using the Friedman's and Wilcoxon tests. Results: 31 patients were studied. In the immediate postoperative period participants reported deterioration in their sense of taste (p 0.001), increased nausea (p < 0.001) and hunger (p ¼ 0.03). Sweet, savoury and spicy tastes were the most popular during the perioperative period. However, only palatability for salty taste increased (p ¼ 0.001) following surgery. The highest rated images were for savoury food with only the ratings for salty food increasing after surgery (p < 0.05). These findings concurred with the sweet, savoury and salty snacks being the most consumed foods in the postoperative period. Bitter, sour and spicy foods were the least frequently consumed. Discussion: This is the first study to investigate postsurgical patients' food preferences. A consistent change in all the individual tastes with the exception of salty in the postoperative period was observed. The most desirable tastes were for savoury and sweet, reflecting patients' preoperative preferences. An improved understanding of taste may improve the resumption of eating after colonic surgery. © 2014 Elsevier Ltd and European Society for Clinical Nutrition and Metabolism.
Keywords: Post-operative Nutrition Enhanced recovery Dysgeusia
1. Introduction Current surgical practice encourages patients to resume eating rapidly after their operations. Nutrition within twenty-four hours of colorectal surgery is associated with improved outcome and shorter postoperative hospital stay [5,1]. Early postoperative nutritional intake is a key component of enhanced recovery programmes (ERP), however, for a variety of reasons post-operative
* Corresponding author. Dept of Gastroenterology, Derriford Hospital, Plymouth, Devon PL6 8DH, UK. E-mail address: [email protected] (S. Lewis).
nutritional intake is often poor [6]. Dysguesia, loss of hunger and alterations in food palatability are often reported. Surgery induces a catabolic response with the release of stress hormones (glucagon, cortisol, catecholamines) and inflammatory mediators [7] many of which are known to suppress appetite. These responses, with the loss of insulin sensitivity (insulin resistance) [7,16], are claimed by some to be dysfunctional. In addition intestinal dysfunction (ileus) is common after certain types of surgery. Pain, drugs and environment may also influence a postoperative patient's appreciation of food. The palatability of a food by an individual has both chemical and psychological components [3], the perceived palatability being a combination of the food's taste, smell, texture, temperature, visual
http://dx.doi.org/10.1016/j.clnu.2014.09.017 0261-5614/© 2014 Elsevier Ltd and European Society for Clinical Nutrition and Metabolism.
Please cite this article in press as: Welchman S, et al., Changes in taste preference after colorectal surgery: A longitudinal study, Clinical Nutrition (2014), http://dx.doi.org/10.1016/j.clnu.2014.09.017
2
S. Welchman et al. / Clinical Nutrition xxx (2014) 1e4
appearance, sound and irritative sensations [19,15]. Generally, increased palatability results in increased oral intake [12]. This may work both through enhanced hedonic experience and a delay in within-meal satiation (the natural inhibitory feedback generated through meal consumption). Loss of taste and smell as a result of direct damage to the taste and olfactory system (chemotherapy, radiotherapy or maxillofacial surgery) are well described and associated with reduced energy intake [17,8,9]. Non-specific changes in taste appreciation are also described in many other medical conditions [11,10,13,2]. Alteration of food palatability as a result of coexistent stresses such as abdominal surgery, are not well characterised and the mechanisms unclear. We hypothesised that taste preferences and desires are altered in the early post-operative period and this affects patients' food choices during this critical time. We set out to describe changes in perioperative taste and food preferences with a view to being able to optimise food choices and thus improve intake. To do this we recorded patients' experience of and preference for individual tastes (bitter, salty, savoury, sour, spicy and sweet) in three separate experiments designed to assess distinct components of food palatability and preference. 2. Materials and methods 2.1. Subjects Patients undergoing a colorectal resection at Derriford Hospital, Plymouth, were eligible for inclusion in this study. Exclusion criteria were: age below 18 years, inability to give informed consent, any changes in taste over preceding six months, participation in another trial, pregnancy, diabetes, steroids or immunosuppressant use, abnormal gastric emptying, intestinal obstruction, concurrent nutritional support, taking medication known to alter taste or being a smoker. Written informed consent was obtained. Postoperatively participants who developed clinical complications requiring cessation of oral intake were withdrawn from the study. The study was approved by Plymouth Local Research Ethics Committee. 2.2. Study design All study participants were managed under the care of their attending surgeon within an ‘Enhanced Recovery’ protocol (ERP). As such patients were encouraged to eat immediately after surgery. Baseline demographic and operative data were recorded. Postoperatively clinical complications were noted. Participants deemed unable to eat in the opinion of their attending surgeon immediately after surgery were excluded from the study. Participants were asked to complete a questionnaire and three experimental protocols to assess changes in patients' general appetite, any changes in taste palatability and desire as well as actual taste preference. Changes were assessed using visual analogue scales (VAS), which have been widely used in evaluation of appetite [18]. These protocols were done at their preoperative assessment appointment and then on postoperative days (POD) 1, 2 and 3 at 18:00 before the evening meal was served when patients would normally be hungry. 2.3. Questionnaire Clinical data were recorded. Current medications including antiemetics were noted. To assess overall perception of their hunger, nausea, thirst and sense of taste, visual analogue scales (VAS, 0e100%) were completed.
2.4. Protocol 1 To assess any change in the palatability over the perioperative period dilutions of the six different tastes were scored by participants. These were added into a neutral flavoured ‘balanced’ polymeric energy drink (Fresubin® 2 Kcal/ml, Fresenius-Kabi, Runcorn, Cheshire, UK) to associate specific taste modalities with caloric intake. Patients were asked to rate the taste of six flavoured energy dense liquids pre-operatively and on POD 1, 2 and 3. The liquids were flavoured with concentrations of the six core tastes (adapted from Steingruber et al. [14] and adjusted by the research team such that the final strength of taste was similar (bitter (quinine sulphate 0.01 g ml1), salty (sodium chloride 0.05 g ml1), savoury (unami 0.005 g ml1), sour (citric acid 0.01 g ml1), spicy (chilli powder 0.005 g ml1) and sweet (glucose 0.05 g ml1)). The solutions were prepared daily in plastic beakers and labelled 1e6. The labelling allocation of taste to code (1e6) was random and held on a preprepared spreadsheet. Both participants and researchers were blinded to which taste was being served. A ‘mouth care’ sponge was dipped in each flavoured liquid and patients were asked to taste and score each taste using a VAS (Horrible-Nice: 0e100%). Participants rinsed their mouths with water after each taste and allowed at least 5 min between each tasting. 2.5. Protocol 2 To examine participants desire to eat food items representative of the core tastes: bitter (grapefruit), salty (popcorn), savoury (cheese on toast), sour (gherkins), spicy (curry) and sweet (sugar mice: mice shaped from icing sugar, egg whites and golden syrup), patients were shown photographs of food items (presented on an identical plain white plate and neutral background). They were asked to rate each of them in terms of desire to consume at the time using a VAS scale (Not appetizing-very appetizing: 0e100%). 2.6. Protocol 3 To examine what foods patients would actually eat we gave patients a box (at 18:00) containing individually packaged (80 cm3 clear plastic container) snacks incorporating a small amount of foods representative of the six core tastes (bitter (dark chocolate), salty (pretzels), savoury (cheese), sour (gherkins), spicy (Bombay mix: a commonly available dried mix of fried pulses, peanuts, corn, flaked rice and curry leaves) and sweet (fudge)). Patients were asked to eat any of the snacks they wished over the next 24 h, leaving the empty container of eaten or partially consumed snacks within the box. The boxes were collected 24 h later and note was made of the items fully consumed. 2.7. Sample size and statistical analysis The study was descriptive and as no previous data were available to inform sample size we decided that thirty patients completing the study would provide a clinically meaningful number. Statistical analysis was done by a researcher that was blinded to allocation (SL, PC) using SPSS 21 (IBM, Chicago, Illinois, USA). Data were non-parametric thus presented as median values with interquartile ranges. Differences in taste perception across the four assessment points (baseline and POD 1e3) were examined using Friedman tests, with significant changes explored further with Wilcoxon rank sum tests to examine differences between scores at baseline and POD 1-3 (Wilcoxon tests are presented as p values with 95% confidence intervals).
Please cite this article in press as: Welchman S, et al., Changes in taste preference after colorectal surgery: A longitudinal study, Clinical Nutrition (2014), http://dx.doi.org/10.1016/j.clnu.2014.09.017
S. Welchman et al. / Clinical Nutrition xxx (2014) 1e4
3
3. Results A total of 36 patients were recruited, of which five participants were excluded owing to postoperative complications making them unable to eat (four participants) or withdrawal of consent (one participant). Baseline demographics of the 31 participants who completed the study are presented in Table 1. There were no significant clinical complications or the use of intravenous fluids after the operative day in the remaining participants. 3.1. Questionnaire Visual analogue scores (VAS, 0e100%) for the questionnaires asked about the participants' sense of taste, whether food tasted normal, hunger, thirst and nausea are presented in Fig. 1. Friedman tests showed changes for taste (p < 0.001), hunger (p ¼ 0.003) and nausea (p < 0.001). No changes in VAS scores for thirst were seen. At baseline all participants reported a VAS score of 0% for nausea and 97% scores of 100% for their sense of taste. Planned comparisons showed scores for taste decreased from baseline compared to postoperative days (POD) 1 (p < 0.001, 45, 20), 2 (p < 0.001, 35, 15) and 3 (p < 0.001, 15, 5), with scores being lowest on POD 1 and slowly returning towards preoperative levels. Reported hunger scores were greater at POD 3 (p ¼ 0.033, 5, 30) compared to baseline. Nausea scores were greater than baseline on POD 1 (p < 0.001, 10, 20), POD 2 (p ¼ 0.004, 0, 5), POD 3 (p ¼ 0.022, 0, 5).
Fig. 1. Questionnaire, Change in sense of taste; does food taste normally, hunger, thirst and nausea across the perioperative period, median (IQ range).
3.2. Protocol 1 Changes in the VAS scores (Horrible-Nice: 0e100%) for the palatability of the six tastes are presented in order of baseline VAS score (lowest score on left to highest on right hand side) in Fig. 2. Scores for bitter were extremely low throughout, whilst savoury, spicy and sweet were the highest. Only scores for salty increased across time points (p < 0.001), with scores being raised 50% from baseline on POD 1 (p < 0.001, 5, 20), POD 2 (p < 0.001, 10, 20) and POD 3 (p < 0.001, 5, 15). 3.3. Protocol 2 The VAS scores (Not appetizing-very appetizing: 0e100%) of how appetizing the foods in the photographs looked are given in Fig. 3 (the least appetizing foods on the left and most on the right hand side). Friedman tests revealed changes in scores for sour (p ¼ 0.009), salty (p ¼ 0.025), bitter (p ¼ 0.001) and savoury (p ¼ 0.047). The savoury food was the most appetizing. Only salty VAS scores increased postoperatively (POD 1, p ¼ 0.03, 0, 20), normalising by POD 2. Scores for sour (POD 1, p ¼ 0.013, 20, 0, POD 2, p ¼ 0.009, 15, 0) and spicy (POD 1, p ¼ 0.025, 25, 0 POD 2, p ¼ 0.05, 25, 0) initially fell but were no different from baseline by POD 3. Similarly scores for bitter (POD 1, p ¼ 0.002, 30, 10) and savoury (POD 1, p ¼ 0.048, 15, 0) all initially fell before recovering by POD 2.
Fig. 2. Protocol 1, Change in taste preference across the perioperative period, median (IQ range).
3.4. Protocol 3 The snack item that was most frequently eaten (fully consumed) postoperatively by patients was the salty-pretzels (POD1-2, 32%, POD2-3 32%) and sweet-fudge (POD1-2 32%, POD2-3 39%) followed by savoury-cheese (POD1-2, 26%, POD2-3 29%). Few people consumed the other tastes (gherkins, dark chocolate or Bombay mix). Due to the small sample size and few patients consuming food from the snack box no statistical description of these results is offered. 4. Discussion We have demonstrated subjective and objective change in participants' description of taste and food preference, or dysgeusia, across the perioperative period. To the authors' knowledge this is
Table 1 Baseline Demographics of 31 patients completing the study. Demographics Median age, years (IQ range) Median BMI Kg m1 (IQ range) Indication for resection Inflammatory bowel disease Benign polyp Operation Right hemicolectomy Left hemicolectomy
76 (68, 81) 28 (22, 33)
Male:Female ratio
14:17
2 4
Malignancy Diverticular disease
24 1
12 6
Anterior Resection Total colectomy
12 1
Fig. 3. Protocol 2, Preoperative food image ‘appetizing’ score, median (IQ range).
Please cite this article in press as: Welchman S, et al., Changes in taste preference after colorectal surgery: A longitudinal study, Clinical Nutrition (2014), http://dx.doi.org/10.1016/j.clnu.2014.09.017
4
S. Welchman et al. / Clinical Nutrition xxx (2014) 1e4
the first report of alterations in patient taste preferences and food choices in the perioperative period excluding surgery to the oropharyngeal region. Postoperatively participants' reported a reduced sense of taste, increased nausea and hunger. Participants preferred savoury, spicy and sweet flavoured supplement drink. Although participants' satisfaction from a salty taste was not as high, this increased in the postoperative period. Participants' theoretical desire to eat different tasting foods (photographs) after surgery was generally low, with savoury tasting food being the most desirable. Again the only food type to increase in desirability was salty. In practice participants predominantly ate salty, sweet and savoury snacks. There was internal consistency of trends across the three strands of the study with sweet and savoury tastes being the most popular and only salty taste increasing significantly. The first experiment suggested that participants' appreciation of the different tastes apart from salty did not change from baseline in the postoperative period. However, the desirability to eat foods of different taste often reduced (apart from salty). The absolute ranking of the desirability of the different tastes at any given postoperative day (POD) seem to reflect their ranking at baseline. As expected the measured values had started to return towards baseline by POD 3. Following surgery the body's stress response results in salt retention in order to retain fluid through activation of the reninangiotensin system. However, humans do not have an easily demonstrable sodium or salt appetite in response to sodium loss or cortisol administration [4]. This may well explain the increase in desire for salty taste. Salt is also a well-recognised taste enhancer. All studies of alterations in taste perception or preference are limited by the complexity of this sense and the variety of factors that can impact on it. In reality all foods are a combination of multiple tastes and other factors such as presentation, texture, temperature, smell and food groups which all contribute to the enjoyment of eating. Thus focussing on the six core tastes is simplistic. Similarly a patient's experience of recovery from operative stress will be governed by a multitude of factors both physical and psychological for which control was not possible. It would be difficult to define a true control group and so taste changes are compared against baseline instead. This may be confounded by the ‘baseline’ measurements not reflecting the patients ‘normal’, given they were done in a hospital environment and the associated anxieties the preoperative patient would be experiencing. The bitter, sour and spicy tastes depending on their strength can be very unpalatable. Their presence in people's diets will be highly variable. Likewise specific food items may or may not be part of patients' diets. For example some people may have little experience of eating pickled gherkins. We found that photographs for bitter tasting grapefruit were popular, whilst the taste and snacks were not. Foods like grapefruit often taste tart or bitter but some varieties can taste sweet; participants' previous experiences or lack of them may have been reflected in their scoring of that taste. At a time when patients are unwell there may be a desire to stick to less adventurous tastes. Early post-operative nutrition improves surgical outcome and is an essential element of modern surgical practice [5]. Previous work has shown postoperative patients prefer simple solid foods to clear fluids [20]. We know that the early postoperative period nutritional intake is often poor [6], thus understanding taste preferences and changes from usual food choices are vital in guiding hospitals to design menus that encourage patients to eat during this period. Our study suggests that sweet and savoury foods may be the most palatable to the postoperative patient. As desire to eat salty food
increases, there may be a role for modest amounts of salt or ‘low sodium’ salt as a taste enhancer. It is clear from this study that patients taste preferences change following surgery. In the context of a clinical drive to restore oral intake in the immediate postoperative period this is an important observation. More work in this field is required to fully define these changes and to investigate the impact of meal options on the time to established diet and calorie intake. Likewise whilst popular liquid nutritional supplements come in many flavours, there is less variety in terms of tastes (bitter to sweet). It is possible that a better understanding of this subject could result in a tailored postoperative menu that offers patients the tastes and foods that they are most likely to eat. This may help to provide patients with the energy and nutrition that they need to support healing, restore bowel function and global recovery. On the other hand, it may be that posttraumatic anorexia, as observed in the animal world, is a healthy physiological response and the way ahead to introduce early enteral feeding is to reduce the inflammatory insult rather than change the taste of the food. Conflict of interest None declared. References [1] Andersen HK, Lewis SJ, Thomas S. Early enteral nutrition within 24 h of colorectal surgery versus later commencement of feeding for postoperative complications. Cochrane Database 2011:CD004080. [2] Depoortere I. Taste receptors of the gut: emerging roles in health and disease. Gut 2014;63(1):179e90. [3] Farmer MN, Raddin RS, Roberts JD. The relationship between taste, olfaction, and nutrition in the cancer population. J Support Oncol 2009;7(2):70e2. [4] Leshem M. The excess salt appetite of humans is not due to sodium loss in adulthood. Physiol Behav 2009;98(3):331e7. [5] Lewis SJ, Egger M, Sylvester PA, Thomas S. Early enteral feeding versus “nil by mouth” after gastrointestinal surgery: systematic review and meta-analysis of controlled trials. Br Med J 2001;323:773e6. [6] Lidder P, Thomas S, Fleming S, Hosie K, Shaw S, Lewis S. A randomised placebo controlled trial of preoperative carbohydrate drinks and early postoperative nutritional supplement drinks in colorectal surgery. Colorectal Dis Off J Assoc Coloproctol G B Irel 2013. [7] Ljungqvist O, Soreide E. Preoperative fasting. Br J Surg 2003;90:400e6. [8] Nguyen HM, Reyland ME, Barlow LA. Mechanisms of taste bud cell loss after head and neck irradiation. J Neurosci Off J Soc Neurosci 2012;32(10):3474e84. [9] Porter SR, Fedele S, Habbab KM. Taste dysfunction in head and neck malignancy. Oral Oncol 2010;46(6):457e9. [10] Schiffman SS. Taste and smell in disease (first of two parts). N Engl J Med 1983;308(21):1275e9. [11] Schiffman SS. Taste and smell in disease (second of two parts). N Engl J Med 1983;308(22):1337e43. [12] Sorensen LB, Moller P, Flint A, Martens M, Raben A. Effect of sensory perception of foods on appetite and food intake: a review of studies on humans. Int J Obes Relat Metab Disord J Int Assoc Study Obes 2003;27(10): 1152e66. [13] Spielman AI. Chemosensory function and dysfunction. Crit Rev Oral Biol Med An Off Publ Am Assoc Oral Biol 1998;9(3):267e91. € hme C. Development of an assessment instru[14] Steingruber R, Gillessen S, Bo ment for chemotherapy associated dysgeusia and its implications for patients. Eur J Cancer 2005;3:446. [15] Stroebele N, De Castro JM. Effect of ambience on food intake and food choice. Nutrition 2004;20(9):821e38. [16] Svanfeldt M, Thorell A, Hausel J, Soop M, Rooyackers O, Nygren J, et al. Randomized clinical trial of the effect of preoperative oral carbohydrate treatment on postoperative whole-body protein and glucose kinetics. Br J Surg 2007;94(11):1342e50. [17] van Dam FS, Hilgers FJ, Emsbroek G, Touw FI, van As CJ, de Jong N. Deterioration of olfaction and gustation as a consequence of total laryngectomy. Laryngoscope 1999;109(7 Pt 1):1150e5. [18] Whybrow S, Stephen JR, Stubbs RJ. The evaluation of an electronic visual analogue scale system for appetite and mood. Eur J Clin Nutr 2006;60(4): 558e60. [19] Yeomans MR. Taste, palatability and the control of appetite. Proc Nutr Soc 1998;57(4):609e15. [20] Yeung SE, Fenton TR. Colorectal surgery patients prefer simple solid foods to clear fluids as the first postoperative meal. Dis Colon Rectum 2009;52(9): 1616e23.
Please cite this article in press as: Welchman S, et al., Changes in taste preference after colorectal surgery: A longitudinal study, Clinical Nutrition (2014), http://dx.doi.org/10.1016/j.clnu.2014.09.017