Preoperative Fasting: An Outdated Concept?
From the * Department of Surgery, VU Medisch
Centrum (Free University Medical Centre), Amsterdam, The Netherlands; and Correspondence: Dr. K. van Norren, Numico Research B.V., PO Box 7005, 6704 PH Wageningen, The Netherlands. Electronic mail may be sent to klaske.vannorren{at}numico-research.nl. Recent studies have shown that fasting during the preoperative period for elective surgery induces a metabolic state that seems unfavorable for patients. Results from animal studies indicate that rapid depletion of liver glycogen before surgery leads to mobilization of muscle glycogen after surgery, in turn leading to reduced muscle strength. Depletion of liver glycogen also influences the function of the mononuclear phagocytic system (MPS), which is located predominantly in the liver. The MPS is essential in restricting endotoxin, which may translocate from the gut. In addition, surgery per se puts a substantial physical strain on the patient, and fasting may adversely affect the metabolic response to surgery. This paper presents experimental and clinical data that, when combined together, prove that fasting before surgery has adverse consequences for the patient. Despite the fact that preoperative fasting is being debated—or has even been abandoned—in a number of countries, it is still standard practice in many Western countries for patients before undergoing surgery.1–3 The motivation for preoperative fasting lies in the concern of an increased risk of aspiration of acidic gastric contents during anesthesia. In 1988 and 1994, Canada and Norway were the first countries to adopt new national guidelines on preoperative fasting.4,5 This means that patients are allowed to drink clear fluids such as coffee or tea until 2 hours before induction of anesthesia. Studies have shown that this does not lead to an increased incidence in aspiration.6 Subsequently, a number of countries, including the UK, the US, Sweden, and Denmark, have followed the Norwegian initiative. The intake of clear fluids such as water, tea, coffee, and fruit juices without solid pulp has proven to be safe up to 2 hours before surgery. These fluids are, however, unlikely to have a major effect on metabolism, and therefore patients are still in a fasted state at operation. Furthermore, because of operating schedule changes, patients are often fasting for much longer than planned, even up to 1–2 days. This paper focuses on experimental and clinical data demonstrating that fasting before surgery has a number of adverse consequences.
Already in 1945, it was clear that rats subjected to burns lose more nitrogen in a fasting state than when fed before the stress.7 This indicates an adverse change in the metabolic response to stress as a result of a short period of fasting. Later, Quiros and Ware evaluated cardiac function in fasted and fed animals, respectively.8,9 Fed rats subjected to hemorrhagic shock had significantly better cardiac output and, correspondingly, a better blood circulation in the skin and muscles than rats that had been fasted before exposure to the stress. In the 1980s, Ljungqvist et al10 studied the effect of feeding and fasting on the mortality rate in rats after a hemorrhagic shock, creating a mean arterial blood pressure of 55 mmHg for 60 minutes. In this study, all fed rats survived, whereas all fasted rats died. A similar study was conducted by Alibegovic et al,11 showing that glucose infusion results in higher liver glycogen, plasma refill, and a 100% recovery and survival rate 7 days postoperatively. All fasted animals died within 3 hours postoperatively. These findings were confirmed in studies by Nettelbladt et al.12 Rats were fasted for a 24-hour period, after which 3 groups were formed: 1 group was given only IV glucose, 1 group received oral carbohydrate, and a third group was given an oral liquid containing a sweetener before a 75-minute hemorrhagic period. The latter "fasted" group had the highest mortality rate. The mechanisms behind these observations need to be further elucidated. Another factor influenced by fasting is muscle strength. Friberg et al13 demonstrated that after 24-hour fasting, there appeared to be a loss of muscle strength, even before the animals had been exposed to hemorrhagic shock. After induction of hemorrhagic shock, muscle strength in the fasted animals decreased compared with fed animals, and this effect lasted for several days. This suggests that functional parameters after acute stress were already affected by the metabolic state of the animal at the beginning of the stress, and that these effects may last for some time. The effects of fasting may also have an impact on events at the intestinal level. Esahili et al14 evaluated the effects of an LD-50 dose of endotoxin on the survival of 24-hour fasted and fed rats. There was a significant difference in the survival rate in favor of the fed rats, with fasted rats showing a 200% greater mortality rate. The fasted rats showed symptoms of liver damage, and more rats died prematurely in this group. The influence of the circadian cycle on the experimental outcome was minimized in these experiments by maintaining lipopolysaccharide infusions for 6 hours and similar experimental starting times of infusions in all groups. Furthermore, Nettelbladt et al15 demonstrated that bacterial homeostasis in the cecum was already affected at 24-hour fasting. After at least 24-hour fasting, an increase of the number of bacteria in the cecum of the rats was found, and after 48 hours of fasting bacterial adherence to the cecal epithelium had also increased. Bacterial overgrowth and adherence can be an important initial step in the process of bacterial translocation, which becomes increasingly important in conditions of a hampered mononuclear phagocytic system (MPS) function and associated decreased endotoxin and bacterial clearance function. Some studies have shown that fasted rats subjected to blood loss are more susceptible to bacterial translocation to the intestinal lymph nodes than rats fed before the stress.16 Tanigawa et al17 have shown that lipid oxidation is significantly increased in fasted rats when compared with fed animals. The increase of lipid oxidation products has been suggested to be involved in the induction of organ failure.18–20 Van Hoorn et al21,22 showed in an intestinal ischemia/reperfusion model that preoperative supplementation with a carbohydrate (CHO) mixture leads to lower plasma asymmetrical dimethylarginine (ADMA) concentrations than in fasted rats. Nijveldt et al23,24 recently reported that ADMA concentrations are concentration dependent, and associated with the incidence and severity of liver failure and intensive care unit (ICU) death. In another type of ischemia/reperfusion (IR)-induced injury model, namely liver transplantation, Arnault et al25 have shown that alanine supplementation as a direct energy substrate improves liver function parameters when compared with livers reperfused with physiologic saline. Recently, van Hoorn et al21,22 have shown that IR intestinal injury after occlusion of the superior mesenteric artery and reperfusion also was associated with significantly lower cardiac output in IR-fasted animals, which was completely avoided by preoperative nutrition to the level of sham-operated rats. Another parameter in postoperative organ dysfunction, oxidative stress (malondialdehyde assessment [MDA]), was significantly lowered in the lung and intestine of preoperatively fed animals when compared with the elevated MDA concentration seen in IR-fasted animals.21,22 Although the animal studies discussed above and listed in Table I present a good indication of some major organ systems influenced by presurgical fasting, the current results obtained from animal models are mostly descriptive and observational. The mechanisms underlying the effects of fasting vs feeding have not yet been elucidated. It can be speculated that preoperative feeding better preserves the energy stores of both liver and intestine, resulting in better preserved integrity of both liver (MPS) and intestine, in turn resulting in a better defense against endotoxins and translocation of bacteria.21,22
There are no human studies describing the effect of preoperative fasting upon morbidity and mortality. However, it is generally accepted that glycogen reserves are depleted after an overnight fast.26 As a result, fasted patients will be forced to utilize muscle proteins for gluconeogenesis upon severe metabolic stress (surgery).27 In cardiovascular patients, it has been shown that insulin treatment decreases protein wasting indicated by decreased levels of creatinin and urea. Supplementation of insulin during operation further decreases mortality rates in patients requiring >5 days of intensive care (p = .005).28 Furthermore, overnight fasting induces postoperative insulin resistance. Insulin resistance is characterized by high glucose concentrations despite the presence of an adequate insulin concentration. This symptom is due to decreased cellular insulin sensitivity, resulting in decreased cellular glucose uptake. The condition resulting from increased cellular insulin resistance is quite similar to type II diabetes mellitus. Moreover, this high glucose concentration has been shown to be a risk factor for the induction of inflammatory complications.29,30 The development of insulin resistance has been shown to be an important factor in postoperative outcome. Thorell et al31 have shown that insulin resistance is an independent predictor of the length of hospital stays for patients undergoing elective surgery. Further research indicated that peripheral tissues are important for the induction of this postsurgery complication. In particular, the glucose transporter protein GLUT-4 seems to play a major role in preventing insulin resistance.28 In a subsequent study, Nygren et al32 have demonstrated that intake of a carbohydrate drink in hip replacement surgery (800 mL the evening before surgery and 400 mL 2 hours before anesthesia) decreased postoperative insulin resistance compared with a physiologic saline drink. Retaining higher glycogen reserves by preoperative glucose infusion rather than an overnight fast has been claimed to be an important factor in reducing postoperative insulin resistance.26 A postoperative phenomenon often observed after, for example, elective surgery or cardiopulmonary bypass surgery is increased nitrogen loss, which may signal loss of muscle function. A study published in 1984 showed that preoperative administration of a high dose of glucose resulted in decreased protein turnover.33 Preoperative IV carbohydrate supplementation was first investigated in heart surgery in the 1980s. These studies demonstrated a lowering of postoperative heart injury and heart complications.34 Although IV glucose infusion has been shown to be effective, high dosages are needed to release sufficient insulin to counteract major metabolic changes.26,35 This may cause irritation to the vein, which limits the clinical use. A good alternative is an oral supplement. Use of oral carbohydrate must meet 2 criteria. Firstly, the drink should be safe and empty rapidly from the stomach to decrease the risk of aspiration. Secondly, the drink should induce a sufficiently high endogenous insulin response to change metabolism in the desired way. To meet these conditions, a special drink was developed containing 12.5% carbohydrate. This induces an insulin response quite similar to that seen after a standard meal.36 Due to the relatively low osmolarity (265 mOsm/L), gastric emptying time of carbohydrate is similar to that of water.37 Further safety tests showed that this iso-osmolar drink could safely be administrated until 2 hours before anesthesia. Clinical studies have demonstrated improved patient well-being, both pre- and postoperatively, and reduced postoperative thirst. Studies on the metabolic effect of oral carbohydrate by intake of 800 mL the evening before the operation and 400 mL 2–3 hours before surgery decreased postoperative insulin resistance as effectively as IV glucose infusion.32,38 Intake of carbohydrate reduces preoperative thirst, hunger, and anxiety,39 preoperative40 and postoperative41 nausea, postoperative vomiting,41 and postoperative loss of lean body mass.42 It may also speed recovery43 after surgery, as there are trends toward a reduction in length of hospital stay.43–45 Preoperative carbohydrate feeding has been taken up in the evidence-based Fast Track Surgery protocols or the Enhanced Recovery After Surgery protocols; an initiative of the European Society of Parenteral and Enteral Nutrition.46 The data from all human studies are summarized in Table II. There are major similarities between the animal models and clinical trials, pointing to the working mechanism behind the preoperative feeding at the physiologic level. First, feeding results in a lower insulin resistance; second, preoperative feeding appears to decrease nitrogen loss. Although the animal and human studies listed in Tables I and II give a good indication of major physiologic systems influenced by presurgical fasting, current knowledge on presurgical feeding is insufficient to explain the protective effect of feeding at the biochemical level. Data from a recent study by van Hoorn et al21,22 indicate that fasting compromises the recovery in the first period after IR injury. The cardiac output 3 hours after intestinal ischemia was lower in overnight fasted animals than in fed animals. This is ascribed to the depletion of the energy stores of the body, especially the liver, in a time of maximum energy need and utilization. These data are in accordance with the data of Alibegovic et al11 Ljungqvist et al,10 and suggesting a role for liver glycogen in posthemorrhagic survival. Furthermore, fasting has been reported to cause an imbalance in the antioxidant system after ischemia, resulting in increased oxidative stress in the lung and intestine.21,22 Therefore, the contribution of presurgical energy status and oxidative capacity to postoperative organ function requires further exploration. Moreover, the contribution of the immune system to postsurgical organ dysfunction warrants further study. The same holds true for the role of the intestine in postoperative organ function.
To answer these questions, in vivo experiments are to be performed to explore whether and to what extent the suggested factors contribute to postoperative organ dysfunction. An intestinal IR model could be used.21,22 By inducing intestinal ischemia, intestinal ATP levels have been reported to drop, which might result in increased intestinal permeability. Elucidation of the postsurgical biochemical mechanism of feeding or fasting in animals might lead to initiatives for human clinical trials. Such trials should focus on the analysis of risk factors induced by fasting, extrapolated from the underlying biochemical mechanisms. These studies, together with the already reported possible beneficial effects of presurgical nutrition in humans, may predict a change in our present view that presurgical fasting is optimal.
Animal experimental studies mimicking surgical trauma demonstrate that a short period of fasting affects the response to stress. Mortality in fasted animals is higher than that in fed animals. Fasting before operation has serious consequences for organs such as the gut, liver, kidney, heart, and lungs, probably because it exhausts the energy reserve capacity of the body. The liver is one of the main organs important in reducing postoperative stress. This organ is also needed to maintain a normal metabolic response, which is a prerequisite for regeneration or for healing of the patient. Compromising liver energy status may harm the patient, resulting in higher postoperative morbidity. Clinical studies have shown that fasting before surgery decreases muscle strength (by nitrogen loss) and increases insulin resistance when compared with feeding the patient. Preoperative oral carbohydrate loading reduces the development of insulin resistance by approximately 50% on the day after surgery.32 Moreover, intake of carbohydrate reduces preoperative thirst, hunger, and anxiety,39 preoperative40 and postoperative41 nausea, postoperative vomiting,41 and postoperative loss of lean body mass.42 We conclude that fasting before operation is an outdated clinical approach. Preoperative carbohydrate supplementation has proven to be safe and to improve well-being pre- and postoperatively. Although feeding compared with fasting in animal models improves organ function, the precise mechanism needs to be further unraveled. Future strategies to improve patient outcome may include preoperative enhancement of physiologic energy reserves such as glycogen stores, by ensuring that patients receive appropriate feeding before surgery, thus enabling a better postoperative metabolic stress response. Received for publication February 18, 2004. Accepted for publication April 4, 2005.
Journal of Parenteral and Enteral Nutrition, Vol. 29, No. 4,
298-304 (2005) This article has been cited by other articles:
|
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
|
|
 |
|
Top
ANIMAL EXPERIMENTAL STUDIES
HUMAN STUDIES
