Postoperative parenteral nutrition in adults

INTRODUCTION — Malnutrition is associated with postoperative complications and an increased risk for death after surgery [1]. Many surgical diseases result in malnutrition, particularly those that are associated with a hypermetabolic state. Advanced age is also associated with malnutrition in hospitalized patients, many of whom require emergency operations [2]. Although optimization of nutrition prior to surgery is an option for a select group of patients, this is not possible in most cases [3]. (See "Overview of perioperative nutrition support", section on 'Preoperative nutrition support'.)

Enteral nutrition support is preferred for all patients who do not have a contraindication. The insertion and maintenance of the enteral feeding tube is safer for the patient and less costly compared with central venous catheter insertion and maintenance for the administration of parenteral nutrition. Enteral nutrition also contributes to the health of the intestinal border over time, preventing secondary complications, while long-term parenteral nutrition causes steatohepatitis. By comparison, even though premanufactured parenteral nutrition is available, individualized parenteral nutrition must be prepared in a sterile environment using a recipe customized for the patient, refrigerated due to limited stability, and, when used in the home, must be transported directly to the patient and used within 24 to 48 hours after removal from refrigeration [4]. After the initiation of therapy, enteral nutrition recipients require less monitoring, whereas parenteral nutrition recipients require regular monitoring (eg, laboratory tests on a weekly and later monthly basis) [5]. In spite of these drawbacks, parenteral nutrition support is necessary for postoperative patients with contraindications to enteral nutrition and/or who are unable to meet their nutritional goals using the enteral route.

The indications and implementation of parenteral nutrition in the stable (ie, not critically ill) postoperative patient are reviewed. The consequences of malnutrition, nutritional assessment, indications for preoperative nutrition support, and implementation of enteral nutrition in surgical patients are reviewed elsewhere. (See "Overview of perioperative nutrition support" and "Clinical assessment and monitoring of nutrition support in adult surgical patients".)

Nutrition support in specific patient populations is also discussed in more detail in separate topic reviews:

●(See "Nutrition support in intubated critically ill adult patients: Initial evaluation and prescription".)

●(See "Overview of nutrition support in burn patients".)

●(See "The role of parenteral and enteral/oral nutritional support in patients with cancer".)

INDICATIONS — Parenteral nutrition provides support for patients in whom enteral feeding is insufficient in meeting nutritional goals or is contraindicated. In postoperative general surgery patients, enteral nutrition cannot be used due to the following conditions:

●Gastrointestinal anastomotic failure may cause peritoneal soiling and peritonitis. Uncontrolled anastomotic leak leads to contamination of the peritoneal cavity and is a contraindication to enteral feeding.

●Gastrointestinal fistulas can arise for a variety of reasons (eg, anastomotic leak, pancreatitis, inflammatory bowel disease) [6]. Enteral feeding can increase drainage through a fistula, which can delay fistula closure. Management of gastrointestinal fistula may include parenteral nutrition for those with a high-output fistula (eg >500 mL/day), especially for those who become volume depleted, have electrolyte abnormalities, or develop malnutrition or lose weight in spite of appropriate enteral supplements and medical therapy to control the fistula [7]. (See "Enterocutaneous and enteroatmospheric fistulas".)

●Postoperative mechanical bowel obstruction is associated with stasis of intestinal contents proximal to the obstruction, which results in bowel distension, nausea, vomiting, and pain. Postoperative ileus can similarly cause nausea, vomiting, and pain. While enteral nutrition is preferred during the interim period of conservative management of postoperative ileus or bowel obstruction, parenteral nutrition support may become necessary. (See "Management of small bowel obstruction in adults" and "Postoperative ileus".)

●Postoperative delayed gastric emptying is commonly observed after gastric and pancreatic surgery. In some cases, when intubation of the small intestine is not possible, delayed gastric emptying precludes safe enteral nutrition in the early postoperative period [8].

●Chyle leak is an uncommon complication of surgery for which supportive care that includes oral or enteral dietary modification is often effective treatment. For patients who do not respond, withholding an enteral diet and administering parenteral nutrition is an option [9]. (See "Chylous, bloody, and pancreatic ascites" and "Management of chylothorax".)

●Supplemental parenteral nutrition should also be considered in surgical patients whose nutritional needs are not being met using an enteral route, such as in those with inadequate absorptive capacity (short bowel syndrome), malabsorption (eg, inflammatory bowel disease, radiation enteritis), or in those with underlying conditions associated with chronic gastrointestinal dysmotility (eg, chronic pseudo-obstruction) [10-12].

●Feeding tube complications can arise either early or late with feeding tube use (eg, gastrostomy, jejunostomy placement). When complications arise, enteral nutrition is often suspended to minimize further complications (eg, leaking into the abdomen) [13].

●Pre-existing malnutrition contributes to postoperative complications, including prolonged hospital stay, poor wound healing at surgical sites, and postoperative infection (eg, pneumonia). In a study of underweight patients undergoing gastrointestinal surgery with short-term contraindications to early enteral nutrition, parenteral nutrition support prolonged hospital stay by less than a day without a significant impact on rates of hospital-acquired pneumonia, urinary tract infection, or 28-day in-hospital mortality [14]. Outcome benefits in other subset populations, including the trauma patient population, may be less impressive and are being studied.

●For patients who began parenteral nutrition support preoperatively due to chronic malnutrition, parenteral nutrition support is continued until sufficient enteral nutrition can be given to meet the nutritional demands of the patient [15,16]. (See "Overview of perioperative nutrition support", section on 'Preoperative nutrition support'.)

CONTRAINDICATIONS

Absolute — Absolute contraindications to parenteral nutrition in the postoperative surgical patient include the following:

●Ability to receive enteral feeding – Patients who can eat or receive enteral nutritional therapy should not receive parenteral nutrition. For patients who can eat but are unable to meet their nutritional goals by eating, enteral nutrition can usually be administered through a temporary feeding tube, which can be inserted and maintained with less risk of complications compared with central venous access, which is necessary to administer parenteral nutrition. Enteral formulas are also less expensive and do not require refrigeration.

●Anticipated duration of fasting is less than five days – During a period of fasting, glycogen serves as fuel for the body for up to 24 hours [17]. Insulin secretion falls with fasting as well, which can lead to muscle breakdown for fuel [18]. Basal insulin secretion can be maintained with the administration of dextrose-containing fluid (eg, dextrose 5% or dextrose 10%), thereby initially preventing muscle breakdown, but eventually this effect is lost, and nutrition support must be implemented to avoid muscle breakdown. Healthy patients generally tolerate periods of fasting up to 14 days without a significant impact on their recovery. However, malnourished patients who are starved after surgery have higher morbidity and mortality. For these reasons, nutrition guidelines recommend delaying the initiation of parenteral nutrition for up to seven days in healthy patients but starting parenteral nutrition as early as possible in malnourished patients [19]. (See 'Timing' below.)

●Terminal illness with short life expectancy – Parenteral nutrition support is unlikely to prolong life or improve quality-of-life in terminally ill patients with a short life expectancy (variably defined, but certainly less than one month) [20]. Whether patients with intermediate-length life expectancy benefit from provision from parenteral nutrition is controversial [21-23].

On the other hand, for patients with a longer life expectancy (>3 months), it is reasonable to conclude that parenteral nutrition will prolong their lives, but with added risk. As an example, in postsurgical patients with recurrence of gynecologic malignancy, malignant bowel obstruction is common. In most cases, curative surgery is not an option, but symptoms can be palliated with gastrointestinal decompression. In the absence of hydration, survival is measured in days. With parenteral nutrition support, longer survival is possible. In a historical cohort study of 55 patients with terminal intestinal obstruction due to ovarian cancer, parenteral nutrition was associated with a significantly longer median survival [24]. The median survival from time of diagnosis of intestinal obstruction for patients receiving parenteral nutrition (TPN) was 72 days (range of 16 to 485 days) compared with 41 days (range of 4 to 133 days) for those not receiving parenteral nutrition; however, the difference in survival was not significant when adjusted for chemotherapy. These benefits are not seen in other inflammatory cancers, and it is theorized that patients with noninflammatory cancers benefit from nutrition support, whereas those with cancers that cause systemic inflammation and catabolism may not.

●Refusal of or advanced directive against artificial feeding – Parenteral nutrition is a medical therapy and should not be used for patients who have declared their wishes against artificial feeding in general, or parenteral nutrition specifically, in person, in an advanced directive, or via a surrogate.

Relative — Patients with appropriate nutritional indications for parenteral nutrition support may have a clinical course (eg, shock, sepsis) that mandates the delay of initiation, or discontinuation of established parenteral nutrition support. Other complicating conditions (eg, volume overload, severe electrolyte disturbance) may make the parenteral nutrition prescription difficult to formulate. The most important of these are discussed briefly below. When the complicating condition has been adequately controlled, parenteral nutrition support can be resumed.

●Shock – Administration of parenteral nutrition in the setting of hemodynamic instability remains controversial. Parenteral nutrition can worsen the condition of a patient in shock. The optimal timing of when parenteral nutrition can be safely instituted as shock resolves remains unclear. (See "Nutrition support in intubated critically ill adult patients: Initial evaluation and prescription".)

●Catheter-related infection – For patients with bacteremia/fungemia, withholding nutrition support may be necessary if an intervening "line-free" period is needed to clear the infection. It is common for providers to withhold parenteral nutrition until there is evidence of resolution of bacteremia/fungemia (eg, negative subsequent blood cultures). In a meta-analysis, parenteral nutrition was a major risk factor for catheter-associated invasive candida infection in critically ill patients [25]. (See "Intravascular non-hemodialysis catheter-related infection: Treatment".)

●Acid-base disturbance – Acid-base disturbances are common in the postoperative patient, resulting from large-volume gastrointestinal fluid loss, bleeding, under-resuscitation, and resuscitation with certain intravenous fluids. These include lactic acidosis and contraction alkalosis. Caution is advised in administering parenteral nutrition prior to completion of resuscitation, as parenteral nutrition can contribute to lactic acidosis in a stressed surgical patient [26]. Similarly, parenteral nutrition can complicate the interpretation of contraction alkalosis through the delivery of acetate, which is converted to bicarbonate and contributes to alkalosis. (See "Overview of postoperative electrolyte abnormalities", section on 'Acid-base imbalance'.)

●Severe electrolyte disturbance – In the setting of abnormalities in potassium, phosphate, and magnesium levels, aggressive electrolyte monitoring and correction is warranted to avoid complications. Caution is advised in prescribing parenteral nutrition electrolytes in cases of severe electrolyte disturbance and/or renal dysfunction. As an example, in cases of hyperkalemia associated with acute renal failure, the administration of potassium-containing parenteral nutrition can lead to arrhythmia. (See "Overview of postoperative electrolyte abnormalities", section on 'Etiologies of postoperative electrolyte abnormalities'.)

●Refeeding syndrome – In the setting of chronic malnutrition, especially with chronic electrolyte losses, the administration of parenteral nutrition can result in refeeding syndrome. With the initiation of parenteral nutrition, brisk decrements in serum electrolyte levels are often observed, resulting in hypokalemia, hypophosphatemia, and/or hypomagnesemia. To avoid complications of refeeding syndrome, electrolytes should be monitored closely for the first few days of parenteral nutrition administration, even as often as several times daily when there are severely low levels or sudden drops in levels. Electrolyte deficits should be aggressively corrected by intravenous replacement to prevent complications. Similarly, it may be appropriate to reduce the caloric content for the first few days of parenteral nutrition administration, or to stop the parenteral nutrition until severe deficiencies are corrected. Any patient with chronic starvation should be screened for thiamine deficiency (eg, thiamine) or receive a prophylactic dose of thiamine [27]. (See "Overview of postoperative electrolyte abnormalities", section on 'Refeeding syndrome'.)

●Volume overload – Parenteral nutrition is typically prepared in a predesignated volume of fluid required by the patient. However, the concentration of parenteral nutrition is limited by the concentrations of its component base macronutrients (eg, dextrose, amino acids, and intravenous lipid emulsion). Minerals (eg, calcium, phosphorus), which will precipitate if overconcentrated, may also determine minimum parenteral nutrition volumes. In patients with clinical features of volume overload (eg, respiratory insufficiency secondary to pulmonary edema), the risk of further overload must be considered prior to administering parenteral nutrition. (See 'Complications' below.)

●Lack of central venous access – Since parenteral nutrition support requires venous access for delivery, lack of appropriate access prevents administration. While parenteral nutrition can be given into a peripheral vein, large volumes are required, and higher concentrations tend to cause phlebitis, often necessitating a central access route. Parenteral nutrition that is not delivered into a central vein should only be used for brief periods of time, to prevent possible complications. (See "Central venous access: Device and site selection in adults", section on 'Nature of infusate' and "Extravasation injury from cytotoxic and other noncytotoxic vesicants in adults".)

TIMING

Initiation — For those with appropriate indications, the timing of initiation of parenteral nutrition is largely based on the patient's clinical condition and preoperative nutritional status. Randomized trials comparing early initiation of parenteral nutrition to later initiation in critically ill patients suggest overall harm from early initiation [28]. (See "Nutrition support in intubated critically ill adult patients: Parenteral nutrition".)

For most non-critically ill patients, the risks associated with postoperative parenteral nutrition usually outweigh the benefit during the first five days of fasting. Recommendations for non-critically ill patients are based on direct evidence from small randomized trials, observational studies, and indirect evidence from larger trials in critically ill patients. From retrospective reviews, authors have concluded that patients who do not suffer from chronic malnutrition tolerate brief periods of fasting [29]. Patients who were well nourished prior to surgery often tolerate a period of starvation up to 14 days [30]. In the absence of a severe stress response (eg, shock, systemic inflammatory response syndrome [SIRS]), glycogen stores provide calories during the first 24 hours of fasting. Thereafter, providing dextrose-containing intravenous fluids postoperatively stimulates basal insulin secretion and minimizes proteolysis during fasting [18,31]. Approximately 400 to 500 kcal/day are needed for a 70 kg patient. Randomized trials comparing parenteral nutrition to prolonged starvation (>14 days) among non-critically ill surgical patients have not been done, but in observational studies, prolonged starvation has been associated with higher complication rates compared with administration of parenteral nutrition. Thus, among patients who cannot receive enteral nutrition, the general consensus as expressed in published guidelines is as follows [19,31]:

●The institution of parenteral nutrition in non-critically ill patients should be delayed until at least five days after surgery in the absence of preoperative malnutrition.

●After 14 days of postoperative starvation in otherwise healthy patients, parenteral nutrition should be initiated.

●For patients with chronic malnutrition and anticipated prolonged bowel rest, it may be appropriate to institute parenteral nutrition earlier than five to seven days. In this subset of surgical patients, parenteral nutrition has been associated with lower complication rates and lower mortality compared with fasting.

Early studies expected parenteral nutrition support to improve outcomes in surgical patients but were disappointing [32-35]. Lactic acidosis was observed, and no mortality benefit was demonstrated, likely related to the stress response of surgery [26,36]. When enteral feeding is not feasible, early initiation of parenteral nutrition has not been associated with benefit and may increase complications. Most studies addressing the timing of parenteral nutrition have been performed in critically ill patients. In particular, a large randomized trial supported later, rather than earlier, initiation of parenteral nutrition in these patients [28]. Furthermore, in reviews of parenteral nutrition for critically ill patients without malnutrition, a short course of parenteral nutrition in the early postoperative period (<5 days) compared with no nutrition failed to show an outcome difference and suggested increased morbidity [28,37]. A systematic review identified 27 trials evaluating nutrition support in patients undergoing elective general surgery procedures and found no significant differences in mortality or major complications for parenteral nutrition compared with standard care (usual oral diet plus intravenous dextrose) [38]. Major complications included pneumonia, intra-abdominal abscess, sepsis, catheter-related infection, myocardial infarction, pulmonary embolism, heart failure, stroke, renal failure, liver failure, and anastomotic leak. There were also no differences for parenteral nutrition initiated preoperatively compared with postoperatively. However, for malnourished patients, parenteral nutrition (overall) was associated with a significant reduction in complications (relative risk 0.52, 95% CI 0.30-0.91) but no mortality differences. Among trials focused on the postoperative period, parenteral nutrition was instituted between 4 and 18 days postoperatively [32,33,35,39-51].

Cessation — The duration of parenteral nutrition is typically dictated by the patient's ability to receive oral or enteral nutrition. For some patients, it may be obvious that prolonged parenteral nutrition support will be necessary or, rarely, permanent (eg, short gut syndrome) [52].

For patients in transition from parenteral to enteral therapy, the timing of discontinuation of parenteral therapy is variable. Once a patient's oral/enteral intake meets a significant portion of their daily fluid and caloric needs, parenteral therapy may be decreased or stopped. Many recovering patients require a transition period, during which parenteral therapy is weaned down as oral/enteral intake increases. In such cases, it is common to discontinue parenteral therapy when fluid goals are met and caloric goals exceed 60 percent of the patient's daily requirement. For patients with intestinal failure, bowel rehabilitation and protracted weaning may be needed. (See "Overview of postoperative fluid therapy in adults", section on 'Monitoring and adjustments' and 'Transition to enteral and oral feeding' below.)

The practice of weaning parenteral nutrition over hours to prevent acute hypoglycemia is rarely necessary and stems from experiences during the early period of parenteral nutrition when patients were overfed. In addition, it depends on the formula the patient receives; some contain insulin, which may affect serum glucose levels around the time of discontinuation. If there is concern for hypoglycemia, reducing the rate over a few hours is a common practice. Alternatively, point-of-care glucose assessments might be a means of ensuring safety and alleviating the need for tapering rates.

ESTIMATING NUTRITIONAL REQUIREMENTS — The goal of parenteral nutrition support is to meet the nutritional requirements of the postoperative surgical patient. Underfeeding and overfeeding should both be avoided, as each is associated with complications (eg, hyperglycemia, hepatic steatosis, poor wound healing, infection, inability to wean from ventilator) [53].

The caloric needs of the postoperative patient have largely been determined by estimating the degree of the patient's surgical stress. Major surgical interventions (entry into a major body cavity [cranium, chest, abdomen], joint surgery, spine surgery), trauma (eg, long-bone fracture, traumatic brain injury), or burn injury induce a metabolic response and catabolic state [18,54-56]. For patients with sepsis, pancreatitis, systemic inflammatory response syndrome [SIRS], acute respiratory distress syndrome [ARDS], massive transfusion requirement, or large surface area burns, the catabolic response may be more profound. The more profound the catabolism, the more severe the muscle wasting that can occur [57,58]. Patients with large surface area burn wounds are the most highly stressed, catabolic patients [59]. An extensive, separate literature guides therapy for this patient population, whose needs exceed those of most other surgery patients. (See "Overview of postoperative fluid therapy in adults", section on 'Physiologic stress response to surgery' and "Hypermetabolic response to moderate-to-severe burn injury and management" and "Overview of nutrition support in burn patients".)

There is no universally accepted way to determine the difference between a mildly and more heavily stressed patient. Interventions, such as better control of pain, temperature, anxiety, and optimizing ventilation, can lower measured calorie consumption in severely stressed surgical, trauma, and burn patients. Indirect calorimetry remains the most accurate way to estimate the patient's calorie consumption, but it is cumbersome and not widely available. Matching calorie prescription to calorie delivery has not proven to be ideal for postoperative patients. Proxy markers of stress, including markers of a systemic inflammatory response (tachycardia, tachypnea, fever, leukocytosis, or leukopenia), are often used together with clinical judgment to determine the severity of stress. Mathematical formulas have been proposed to estimate the nutritional requirements of the postoperative patient, but there is no single universally accepted formula to estimate the patient's requirements. The American Society for Parenteral and Enteral Nutrition (ASPEN) has recommended the use of a simple weight-based (ideal body weight) estimate of 25 kcal/kg/day when indirect calorimetry is not available [19]. (See "Nutritional demands and enteral formulas for adult surgical patients", section on 'Calculating energy requirements'.)

For the average surgery patient without large wounds or other catabolic states, 25 kcal/kg/day is a good starting point. For patients with catabolic states (eg, sepsis, head injury) or large wounds, we increase the estimate to 30 to 35 kcal/kg/day.

The initial caloric estimate is just a starting point for parenteral nutrition prescription. Adjustment to the daily calorie goal is based on many factors, including wound healing, weight gain, and ventilator weaning. Historically, serum protein production (eg, albumin, prealbumin, transferring, and C-reactive protein) was considered to be a marker of adequate nutrition, but this practice has been abandoned as our understanding that there is a lack of relationship between intake and the levels of these proteins has increased [60,61]. A more global assessment, including examination (eg, wound healing), functional status, and nitrogen balance, is used to judge the adequacy of nutrition in the postoperative patient. (See "Clinical assessment and monitoring of nutrition support in adult surgical patients", section on 'Monitoring nutrition support'.)

Collection of urine for 24-hour measurement of nitrogen excretion has long been used to drive protein prescriptions. However, the collection depends on good renal function, which may be compromised in the postoperative patient. It remains an imperfect estimate [62].

PRESCRIPTION — The process of ordering, reviewing, compounding, labeling, and administration of parenteral nutrition is complex and prone to error. Institutional practices and policy should be based on published safe practices guidelines [63-67].

Parenteral nutrition is ordered for 24-hour periods, often with a 12-hour lead time based on laboratory studies from hours prior; it should be viewed as a two-day process. Fluids and electrolytes are best managed with conservative prescriptions that anticipate what the patient will need once stabilized, with external supplementation (eg, intravenous doses) of electrolytes or fluid, as needed. In this way, overdosing is avoided, and supplementation occurs in a timely manner.

Initial prescription — The initial prescription of parenteral nutrition requires the following basic elements:

●Fluid volume – The initial volume of fluid should meet the anticipated fluid goals of the patient for 24 hours. For the average postoperative patient who is not expected to have significant ongoing fluid losses, 30 to 40 mL/kg/day is adequate. The exact minimum volume needed to prevent ingredient precipitation depends upon the content of the recipe. For the typical prescription in the average 70 kg adult, the minimum volume is typically approximately 1.5 to 1.75 liters. For patients in whom volume restriction is important, the minimum volume can be confirmed with the pharmacy. Similarly, for patients with high fluid losses (eg, enterocutaneous fistula, large open wound), a fluid prescription of up to 5 L/day is not unusual. (See "Overview of postoperative fluid therapy in adults", section on 'Maintenance fluid therapy'.)

For patients with unpredictable fluid losses, supplemental intravenous fluid can be administered to meet daily fluid goals and titrated for ongoing fluid losses. Typically, the fluid chosen to supplement parenteral nutrition is based on the bodily fluid that is lost. As an example, large-volume gastric fluid loss (eg, bowel obstruction) is high in chloride; as such, 0.9% saline may be chosen as a replacement fluid. (See "Overview of postoperative fluid therapy in adults", section on 'Fluid resuscitation' and "Overview of postoperative electrolyte abnormalities".)

●Protein – Approximately 20 to 25 percent of the daily caloric needs of the average surgical patient is provided as protein. Based on consensus opinion, the average surgical patient should receive 1.2 to 2 g/kg/day of protein. Each gram of protein contains 4 kcal of energy and is delivered in parenteral nutrition by incorporating an amino acid solution. It is suggested that surgical patients with large wounds receive an additional 0.5 to 1 g/kg/day of protein.

●Lipids – Typically, 10 to 30 percent of the daily caloric requirement is met with lipid emulsion. Each gram of lipid contains 9 kcal of energy. For patients receiving lipid-based medications, such as propofol sedation, the lipid content contributed by the medication should be considered. In some institutions, lipids are administered as a separate infusion from the daily parenteral nutrition prescription. Consensus guidelines favor admixture of intravenous lipid emulsion (ie, soybean oil-containing intralipid) with the parenteral nutrition, the so-called total nutrient admixture (TNA) or 3-in-1 parenteral nutrition [67]. However, randomized trials are lacking. In our practice, we do not administer lipids in the setting of hypertriglyceridemia exceeding 350 mg/dL, since lipoprotein lipase is saturated [68]. However, if the triglyceride level is drawn during infusion of lipid (eg, parenteral nutrition administration, propofol administration), levels up to 400 mg/dL are widely accepted. If the triglyceride level is drawn off lipid infusion, a lower threshold of 250 mg/dL is often used in holding lipid treatment.

For patients requiring parenteral nutrition for two weeks or less, a plant oil-based lipid emulsion (eg, Intralipid, Clinolipid, Nutrilipid) typically suffices. The authors use Intralipid 20% emulsion, which contains 20% soybean oil and 1.2% egg yolk phospholipid. (See 'Short-term parenteral nutrition administration' below.)

Recipients of long-term parenteral nutrition support are at risk for cholestatic liver injury and parenteral nutrition-associated liver disease, which is commonly attributed to derangements in serum levels of alpha-tocopherol, phytosterol, and omega-6 polyunsaturated fatty acid [69]. To limit adverse effects in patients who require parenteral nutrition for more than two weeks, the authors substitute the plant oil-based lipid emulsion with an alternative (ie, SMOFlipid), which is felt to confer immunological, anti-inflammatory, and pro-regenerative benefits to the recipient at risk for parenteral nutrition-associated liver disease [70]. SMOFlipid contains soybean oil (30%), medium chain triglycerides (30%), olive oil (25%), and fish oil (15%) [71]. (See 'Long-term parenteral nutrition administration' below.)

A subset of postoperative patients (critically ill with multiorgan dysfunction) may benefit from the immunomodulatory properties of fish oil-enriched parenteral nutrition (ie, SMOFlipid). However, few high-quality data exist to support this. In a noninferiority study, investigators demonstrated noninferiority of SMOFlipid compared with traditional olive oil-based formula (ie, Clinoleic) [72]. In a small, randomized study of fish oil-containing lipid emulsion (ie, Lipidem), nonsurgical patients with severe pancreatitis experienced improved outcomes and a decreased inflammatory response [73]. Many patients with necrotizing pancreatitis will go on to require surgical interventions; this remains an active focus of study. (See 'Dyslipidemia' below.)

●Carbohydrates – Carbohydrate content (in the form of dextrose) typically makes up 50 to 60 percent of the patient's daily calories. Each gram of dextrose contains 3.4 kcal of energy. Carbohydrate content is adjusted based on the protein and lipid content to maintain the calorie goal with adjustments in protein or lipid additives:

●Electrolytes and acid-base balance – Avoiding hypokalemia/hypophosphatemia/hypomagnesemia is important in optimizing return of bowel function and avoiding complications of electrolyte deficiency (eg, arrhythmia, seizure, weakness). Daily electrolyte monitoring is essential in tailoring a prescription to meet the postoperative patient's needs. Each day, electrolyte content can be added, subtracted, or adjusted to meet the patient's ongoing needs. (See 'Prescription' above and "Overview of postoperative electrolyte abnormalities", section on 'Laboratory monitoring'.)

Daily electrolyte requirements for parenteral nutrition are as follows:

•Sodium 1 to 2 mEq/kg/day (in the form of sodium chloride or sodium acetate).

•Chloride and acetate are adjusted for acid-base balance.

•Potassium 1 to 2 mEq/kg/day (in the form of potassium chloride, potassium acetate, or potassium phosphate). It is highly recommended that only one form of potassium be used on any given day, when possible, to avoid mathematical errors in potassium dosage.

•Magnesium 8 to 20 mEq/day (in the form of magnesium sulfate).

•Calcium 10 to 15 mEq/day (in the form of calcium gluconate).

•Phosphorus 20 to 40 mmol/day (in form of sodium phosphate or potassium phosphate).

When choosing a potassium or sodium additive, the acid-base balance and chloride level should be considered. Acetate is added to parenteral nutrition as a buffer and to achieve acid-base balance as an alternative to chloride in patients with hyperchloremia. Acetate is converted to bicarbonate, which is helpful for patients with ongoing bicarbonate loss. For patients without this issue, acetate content is minimized. As an example, a patient with a pancreatic fistula (rich in bicarbonate) would benefit from potassium acetate rather than potassium chloride to replete bicarbonate. Conversely, a patient with hyperkalemia and renal failure will achieve acid-base balance with dialysis; thus, potassium and acetate content should be minimized in such a patient.

●Vitamins and trace elements – For most patients, vitamins and trace elements are provided as the intravenous equivalent of a multivitamin. Supplementary amounts can be added to treat specific deficiencies that may be present.

Ongoing monitoring and adjustments — Parenteral nutrition can be prescribed on a daily or weekly basis in hospitalized patients, or on a monthly, quarterly, or even annual basis for those requiring chronic nutrition support.

Most postoperative patients require laboratory monitoring in the perioperative period [67]. Early in the course of parenteral nutrition support, it is common to check a daily chemistry panel. This is useful for identifying and treating electrolyte abnormalities, monitoring for refeeding syndrome, hyperglycemia, and hydration status. In the early postoperative period, it is common to add potassium, magnesium, and acetate until a stable recipe is achieved, as evidenced by normal blood chemistry.

As the fluid and electrolyte needs of the patient stabilize over days, parenteral nutrition recipes can be prescribed by the week. There is no standard time course for daily, weekly, and monthly ordering of parenteral nutrition, as it typically requires individual customization. Early on, weekly assessments will include triglyceride levels and liver function studies to monitor further for complications such as hyperlipidemia and steatohepatitis. Chronic parenteral nutrition recipients may have their prescriptions extended beyond one week and may undergo less frequent laboratory monitoring.

Parenteral nutrition can also be administered in a cycled fashion. Typically, stable recipes can be converted from 24-hour infusions to 12-, 16-, or 20-hour infusions, allowing the patient some time to be disconnected from the infusion. However, in some cases, ongoing fluid losses do not permit extended periods of time off parenteral nutrition. In these cases, a continuous infusion of a large volume of parenteral nutrition may be used. Alternatively, a standard volume of parenteral nutrition can be supplemented with separate administration of intravenous fluid.

Lastly, for patients in the hospital, parenteral nutrition is often ordered daily, regardless of the chronicity of administration. Similarly, the preparation is compounded in a specialized pharmacy, which may be offsite. It is common to obtain parenteral nutrition from a private, outside pharmacy. In such cases, the deliveries are often batched daily for cost reasons. Parenteral nutrition has a short shelf life; on average, it should be administered within 30 hours at room temperature or nine days if refrigerated [65].

Other prescription adjustments — The nutritional needs of a postoperative patient are largely defined by the underlying disease process, the procedure performed, and any complications that result from the disease or procedure. In creating a customized parenteral nutrition prescription for the postoperative patient, one should consider the individual needs of each patient, including the following:

Renal disease — Protein adjustment for renal disease is controversial. For patients with chronic kidney disease, dietary protein restriction may slow the progression of their renal disease. Conversely, patients on peritoneal dialysis regularly lose protein from peritoneal fluid, and aggressive supplementation has been proposed. Consultation with a renal nutritionist and/or nephrologist is advised to determine protein prescription requirements for patients with significant renal dysfunction. (See "Dietary recommendations for patients with nondialysis chronic kidney disease".)

Liver disease — Patients with liver disease may metabolize amino acids unpredictably, depending on their degree of liver function. With progressive liver failure, the patient's ability to metabolize amino acids will be increasingly compromised, resulting in hepatic encephalopathy [74]. The American Society for Parenteral and Enteral Nutrition (ASPEN) 2016 guidelines recommend using the patient's dry weight or usual weight, rather than an actual weight, to estimate protein requirements in patients with liver failure and cirrhosis. This recommendation is based on expert consensus and recognizes that complications of liver failure (ascites, edema) often make actual weight an unreliable measurement for estimating a patient's protein needs. ASPEN recommends against restricting protein in liver failure [19]. The optimal dosing of amino acids for patients with liver failure is unknown [75], and many institutions do not alter the protein additive for liver disease [74]. Protein restriction may rarely be used as a last resort for brief periods in patients with hepatic encephalopathy unresponsive to medical interventions, such as a patient with worsening acute liver failure awaiting liver transplantation who suffers from hepatic encephalopathy refractory to medical treatment. Similarly, ASPEN advises against using branched-chair amino acid formulas in patients with hepatic encephalopathy due to a lack of evidence of benefit.

Hyperglycemia — Hyperglycemia is common in hospital and ambulatory patients receiving parenteral nutrition [76]. No specific alteration in carbohydrate content is recommended for patients with diabetes. However, blood glucose monitoring should be performed in a scheduled fashion to ensure normoglycemia, especially early in the administration of parenteral nutrition [77]. Insulin can be added to the parenteral nutrition once a stable recipe and a stable insulin regimen are established.

High-volume fluid loss — Voluminous and difficult to predict fluid shifts can occur during and after major cavity surgery, large surface area debridement, small bowel obstruction, ascites production in liver failure, open abdomen with a closed system suction dressing, and enterocutaneous fistula. The volume of parenteral nutrition can be adjusted to accommodate large fluid losses. Typically, the starting parenteral nutrition volume is weight based (30 to 40 mL/kg/day). Additional fluid requirements are supplemented with intravenous fluids to meet the goal volume for 24 hours. The parenteral nutrition volume can also be adjusted, often in increments of 500 to 1000 mL on a daily basis, for convenience. Uncommonly, chronic daily fluid needs exceeding 4 to 5 L/day can be met with the administration of supplemental intravenous fluids. (See "Overview of postoperative fluid therapy in adults", section on 'Fluid resuscitation'.)

Chyle leak — The loss of lymphatic fluid, such as in traumatic or surgical disruption of lymphatic channels of the neck, chest, abdomen, or lymph node basins, can lead to the loss of fluid, protein, and lymphocytes. Nonoperative management includes a variety of enteral feeding modification approaches, from a zero-fat diet to nothing by mouth. Parenteral nutrition is prescribed to meet deficits of nutrition during this therapy [9]. (See "Management of chylothorax" and "Chylous, bloody, and pancreatic ascites".)

Large wounds — Consensus recommendations suggest additional protein and other supplements (eg, vitamin C, zinc) to promote collagen production when healing large wounds, particularly those healing by secondary intention. While these are integral nutrients in healing, data supporting their supplementation in the absence of deficiency are lacking in the general surgery patient. Some studies of burn patients have demonstrated decrements in serum vitamin C early in the treatment course [78]. Whether to provide supplement to burn patients is reviewed separately. (See "Overview of nutrition support in burn patients".)

Obesity — Enteral, but not parenteral, nutritional goals for patients who are obese and critically ill (body mass index [BMI] ≥30 kg/m²) are detailed in the American Society for Parenteral and Enteral Nutrition (ASPEN) and Society of Critical Care Medicine (SCCM) 2016 nutrition guidelines [19]. This consensus statement recommends underfeeding patients who are obese (ie, providing 65 to 70 percent of target energy requirements) based on caloric estimates obtained by indirect calorimetry. When indirect calorimetry is not available, weight-based caloric estimates and protein requirements can be calculated as follows:

•BMI 30 to 40 kg/m² – 11 to 14 kcal/kg actual body weight/day; 2 g/kg/day of protein

•BMI 40 to 50 kg/m² – 11 to 14 kcal/kg actual body weight/day; 2.5 g/kg/day of protein

•BMI >50 kg/m² – 22 to 25 kcal/kg ideal body weight/day; 2.5 g/kg/day protein

Hypermetabolic states — Postoperative patients with hypermetabolic states (eg, burns, head injury, polytrauma, sepsis, malignancy, inflammation, gastrointestinal dysfunction, and burns have nutritional requirements that exceed the average postoperative goal of 25 kcal/kg/day [79-82]. As an example, critically ill burn patients receiving parenteral nutrition may be provided as much as 7 g/kg/day of carbohydrate alone, which can result in hyperglycemia. In some extreme cases, more-than-usual lipid content (eg, 30 percent lipid in burn patient) may be used. Similarly, burn patients may be provided a minimum of 1.5 to 2 g/kg/day of protein. It is important to recognize patients that fall into the extreme hypermetabolic state and to support them appropriately. For these patients, obtaining an accurate resting energy expenditure (eg, by metabolic cart) is essential in avoiding underfeeding and overfeeding. (See "Hypermetabolic response to moderate-to-severe burn injury and management".)

Similarly, patients with traumatic brain injury require more aggressive nutrition support. The Mifflin St. Jeor, Ireton-Jones, and Penn State equations are often used to estimate the caloric needs of traumatic brain injury patients, although the gold standard for measuring energy expenditure is indirect calorimetry. These patients also require higher protein content than the average postoperative patient (1.5 to 2 g/kg/day) [54]. (See "Overview of dietary trace elements" and "Overview of perioperative nutrition support" and "Nutrition support in intubated critically ill adult patients: Initial evaluation and prescription".)

Respiratory failure — Caloric adjustments are often made in the support provided to patients with persistent respiratory failure. When indirect calorimetry is not available to guide the estimate of a calorie goal, it is easy to underfeed or overfeed a patient. In the case of overfeeding, excess calorie administration results in prolonged time to weaning from ventilator. For this reason, empiric adjustments may be made when a patient's respiratory failure persists without explanation.

Dyslipidemia

Short-term parenteral nutrition administration — Derangements in total cholesterol, lipoprotein, and triglyceride levels are commonly observed in the adult surgical population. The relationship between pre-existing dyslipidemia and parenteral nutrition-associated dyslipidemia remains poorly understood; it is complex, depending upon lipid dosing, lipid metabolism prior to parenteral nutrition administration, and clinical factors. Hypertriglyceridemia (triglyceride >200 mg/dL) appears to be the most widely studied derangement and is attributed to overfeeding. Hypertriglyceridemia results when the parenteral nutrition infusion rate exceeds the capacity of lipoprotein lipase to clear triglyceride from the blood. Risk factors for hypertriglyceridemia due to parenteral nutrition identified in a small, retrospective study of non-critically ill patients included: pre-existing hypertriglyceridemia, increased BMI, and glucose dosing >3.1 g/kg/day. In this study, omega-3 fatty acid supplementation resulted in a nonsignificant reduction of hypertriglyceridemia. In a multicenter, prospective study, 26.2 percent of patients receiving parenteral nutrition developed hypertriglyceridemia; the majority of these patients (91.2 percent) presented one or more risk factors, including renal failure, hyperglycemia, corticoid administration, pancreatitis, and sepsis [83]. The patients in this study received a mean lipid dose that was low (0.83 g/kg/day). These risk factors are commonly observed in critically ill surgical patients. Thus, we advise close lipid monitoring in surgical patients, especially those who are critically ill and/or who receive lipid-based medication (eg, propofol) [84].

Scant data are available to provide standardized recommendations regarding the incorporation of specific lipid products when starting parenteral nutrition [63]. Historically, soybean oil was the first commercially available lipid emulsion. In spite of its inflammatory properties, its persistent use may be due to its lower cost and higher amount of fatty acids (eg, linoleic acid, alpha-linolenic acid [85,86]). Most "starter" parenteral nutritional formulas contain soybean oil, which may drive dyslipidemia in the first few weeks of parenteral nutrition administration. (See 'Initial prescription' above.)

●In a study of eight patients receiving Intralipid 10% emulsion, all demonstrated elevated low-density lipoprotein, phospholipid, and cholesterol [87].

●In a comparative study, metabolic complications were more likely to develop in older patients receiving parenteral nutrition, but hypertriglyceridemia was infrequently observed (1.2 percent) in patients age ≥65 [88].

●In a retrospective review of 172 patients undergoing parenteral nutrition for medical and surgical reasons, hypertriglyceridemia was observed far less frequently in older patients age >80 years compared with patients age <80 years (7 versus 36 percent) [2].

For patients who develop dyslipidemia within the first two weeks of parenteral nutrition, the authors use the management strategies listed below:

●Exclude lipid from parenteral nutrition in the first one to two weeks of therapy. In one retrospective review, the withdrawal of lipid emulsion in hospitalized patients with hypertriglyceridemia was associated with a significant reduction in triglyceride level, although later reintroduction correlated to an increase in triglyceride level [89]. Caution is advised in withholding lipid for greater than two weeks, which can contribute to essential fatty acid deficiency.

●For a triglyceride level >250 mg/dL measured from a blood sample obtained while the patient is off lipid infusion, we advise discontinuation of the lipid component of parenteral nutrition. We monitor triglyceride levels weekly until they return to normal and reimplement lipid therapy at a lower dose thereafter.

●We tolerate a triglyceride level up to 400 mg/dL when measured from a blood sample that is drawn while lipid is infusing. If the level exceeds 400 mg/dL, we advise discontinuation of the lipid component of parenteral nutrition. We monitor triglyceride levels weekly until they return to normal, and reimplement lipid therapy at a lower dose thereafter.

●Administer the lipid component of parenteral nutrition separately. Space doses as far apart as possible to optimize recovery time in between.

●Minimize the lipid dose as much as possible.

●Fish oil and omega-3 fatty acid administration may reduce the triglyceride level through different mechanisms.

●Switch to an alternative lipid formulation (eg, SMOFlipid) as with dyslipidemia associated with long-term parenteral nutrition. (See 'Long-term parenteral nutrition administration' below.)

Liver enzymes (aspartate transaminase, alanine transaminase, alkaline phosphatase, gamma-glutamyl transpeptidase) are obtained daily, and a lipid panel is obtained intermittently to detect clinically impactful hepatic inflammation after starting parenteral nutrition. Caution is advised in interpreting typical inflammatory markers (eg, C-reactive protein, erythrocyte sedimentation rate) in the first six to eight weeks after surgery, when they are commonly deranged due to the inflammatory response associated with surgery.

Long-term parenteral nutrition administration — The impact of dyslipidemia associated with long-term parenteral nutrition administration is better established. Longstanding liver inflammation may result in dyslipidemia. This relationship is well established and is tied to the composition of the lipid formulation; omega-6 fatty acid, phytosterol, and omega-6 polyunsaturated fatty acid have all been implicated [85,86,90]. Long-term parenteral nutrition recipients (ie, >2 weeks) benefit from further evaluation of their systemic inflammatory status; 2016 ASPEN guidelines recommend monitoring C-reactive protein, erythrocyte sedimentation rate, lipid panel, and liver enzymes.

Alternative fat emulsions (eg, SMOFlipid), including those containing medium-chain triglycerides, olive oil, and fish oil, have proven to be beneficial in minimizing or preventing chronic inflammation. When compared with soybean oil, these alternative formulations contain less omega-6 fatty acid, less or no phytosterol, and may confer beneficial immunomodulatory effects due to omega-3 fatty acid, increased antioxidants, and a reduction in inflammatory mediators [91].

●In a randomized trial of SMOFlipid 20% versus medium-chain/long-chain triglycerides 20%, recipients of SMOFlipid demonstrated lower triglyceride levels, which was attributed to a better balanced fatty acid complement [92]. Case reports exist of reversal of liver disease with fish oil-based lipid formulations.

●In a meta-analysis of six trials, an association was found between SMOFlipid and lower levels of hepatic enzymes, low-density lipoprotein, triglyceride, and C-reactive protein levels, suggesting decreased liver toxicity with SMOFlipid administration compared with traditional soybean emulsion [93].

For patients who develop dyslipidemia or signs of worsening systemic inflammation (ie, elevated liver enzymes) while on long-term parenteral nutrition, it is important to consider the patient's clinical context. Inflammatory markers and liver enzymes may worsen with conditions commonly observed in the surgical patients including sepsis, medication-induced hepatotoxicity, and the sequelae of traumatic liver injury.

The following management strategies are used to adjust the lipid emulsion in response to dyslipidemia:

●If traditional soybean-based emulsion is employed, switch to SMOFlipid with ongoing monitoring.

●If dyslipidemia or worsening systemic inflammation occurs on SMOFlipid, consider decreasing the amount of lipid administered, administration of supplements that lower triglyceride levels (eg, fish oil), cycling lipid infusion (allowing 6 to 12 hours of lipid-free time for recovery), or providing lipid separately administered once or twice weekly.

TRANSITION TO ENTERAL AND ORAL FEEDING — There is no universally accepted method of weaning from parenteral nutrition. Weaning the volume or time spent on parenteral nutrition is usually done in a methodical fashion [94,95]. To be fully liberated from nutrition support, the patient should be able to maintain their weight, nutrition, and hydration on an oral diet. (See "Clinical assessment and monitoring of nutrition support in adult surgical patients", section on 'Monitoring nutrition support'.)

For patients with a reversible illness that is resolving with treatment (eg, bowel obstruction, sepsis, pancreatitis), the patient will reach a point where they will tolerate enteral feeding.

●After a short course of parenteral nutrition (ie, two to four weeks), parenteral nutrition can be discontinued when >60 percent of the caloric needs are met enterally, although this number varies widely in practice. It is generally accepted that a recovering patient will continue to progressively improve and increase their caloric intake.

●For patients receiving parenteral nutrition for a prolonged period of time (ie, >1 month), a slower course of weaning from parenteral nutrition is advised. Typically, the patient should meet at least 80 percent of daily nutrition goals through enteral intake to maintain body weight and hydration.

In some cases, enteroclysis can be used to help liberate the patient from parenteral nutrition. The principle behind this process is the collection and refeeding of enteral drainage from one part of the gastrointestinal track to another. As an example, a patient with both a high-output duodenal fistula and jejunal fistula can undergo collection of duodenal effluent, which can be fed into the jejunal fistula with a soft feeding tube. This process allows for physiologic enteral feeding, which is superior to parenteral nutrition. However, the technique is challenging for providers and patients alike. Despite this, some patients are successfully liberated from parenteral nutrition in this manner [7].

In some cases, patients who undergo parenteral therapy in hopes that a fistula will close fail to achieve fistula closure. As an example, a patient with an enterocutaneous fistula may have persistent drainage after months of enteral rest and parenteral nutrition. Such a patient will require surgery to be liberated from parenteral nutrition.

Intestinal rehabilitation — In cases of extensive bowel resection (eg, short gut syndrome) or fistulization, some patients may not be able to survive without parenteral nutrition due to intestinal dumping and malabsorption, which leads to dehydration, malnutrition, and a host of vitamin deficiencies. Many borderline patients require formal gut rehabilitation to be safely liberated from parenteral nutrition. Patients who lose >75 percent of their small bowel are expected to require dietary modification and/or parenteral nutrition. In particular, ileal resections result in bile salt and B12 malabsorption, which will worsen steatorrhea and vitamin absorption. Additionally, loss of the colon and/or ileocecal valve may contribute to a patient's parenteral nutrition dependence. (See "Management of short bowel syndrome in children", section on 'Protocol for feeding advancement' and "Management of short bowel syndrome in adults", section on 'Weaning parenteral nutrition'.)

Intestinal rehabilitation programs have yielded some useful insights into the transition from parenteral to enteral feeding. As patients recover from their illness, enteral nutrition can be modified as follows:

●Continuous or frequent small feeds saturate intestinal absorptive channels; thus, gradual introduction is recommended.

●Diets should favor complex carbohydrates over simple sugars to avoid the osmotic load that generates diarrhea.

●Protein-rich diets (ie, 20 percent protein) are more physiologic and are better tolerated.

●High fat (ie, 30 to 40 percent fat) is recommended; fat is tolerated better than a highly osmotic carbohydrate load.

●Fiber supplementation enhances adaptation and water absorption and can be constipating, which is good for patients with malabsorption.

●Oxalate restriction and avoidance of foods high in vitamin C may be needed for patients with ileal resection >100 cm; oxalate is absorbed more readily in states of bile acid deficit, leading to kidney stones.

●Vitamin/mineral deficiencies may persist, depending on fat and vitamin absorption.

●Antimotility medication therapy is often needed to manage symptoms of malabsorption.

COMPLICATIONS — Complications of parenteral nutrition are reviewed separately. (See "Nutrition support in intubated critically ill adult patients: Parenteral nutrition", section on 'Complications'.)

SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Nutrition support (parenteral and enteral nutrition) in adults".)

SUMMARY AND RECOMMENDATIONS

●An enteral route for nutrition is generally recommended, but parenteral nutrition may be necessary for postoperative patients who have a contraindication to enteral nutrition (eg, postoperative anastomotic failure, mechanical bowel obstruction). Supplemental parenteral nutrition should also be considered in surgical patients whose nutritional needs are chronically not being met using an enteral route. Patients with appropriate nutritional indications for parenteral nutrition support may have a clinical course that complicates initiation or mandates discontinuing established parenteral support (eg, shock, refeeding syndrome). (See 'Indications' above and 'Contraindications' above.)

●The optimal timing for initiation of parenteral nutrition in the postoperative period remains controversial. For those with appropriate indications, the timing of initiation is largely based on the patient's clinical condition, expected time to recovery, and preoperative nutritional status. (See 'Timing' above.)

•For non-critically ill postoperative patients who are adequately nourished, we suggest not initiating parenteral nutrition prior to five days postoperatively (Grade 2B). Most surgical patients will tolerate a short period of supported fasting, during which intravenous fluids are often sufficient. Prolonged starvation (>14 days) has been associated with higher complication rates compared with administration of parenteral nutrition. Thus, there is a general consensus that after 14 days, parenteral nutrition should be started.

•For non-critically ill postoperative patients who are malnourished at the time of surgery and who are not expected to meet enteral nutrition goals within five days, we suggest early initiation of parenteral nutrition (Grade 2B). In this subset of surgical patients, parenteral nutrition has been associated with lower complication rates when compared with fasting. (See "Overview of postoperative fluid therapy in adults", section on 'Maintenance fluid therapy'.)

•Recommendations for critically ill patients are similar and are discussed separately. (See "Nutrition support in intubated critically ill adult patients: Initial evaluation and prescription".)

●The parenteral nutrition prescription is determined by estimating the caloric, protein, lipid, and fluid needs of the postoperative patient. For the average surgical patient, a good initial estimate of required calories is 25 kcal/kg/day. A higher number of calories may be necessary for those with large wounds or other hypermetabolic states. Initial protein, lipid, and carbohydrate needs are reviewed above. Adjustments are made to customize the formula to account for coexisting conditions and ongoing nutritional and fluid and electrolyte needs. (See 'Initial prescription' above and 'Other prescription adjustments' above.)

●The transition from parenteral to enteral nutrition is commonly achieved when the patient recovers from surgery and/or complications of surgery. Some patients require prolonged parenteral nutrition support and bowel rehabilitation. Rarely, chronic parenteral nutrition is necessary for those with intestinal failure. (See 'Cessation' above and 'Transition to enteral and oral feeding' above.)