Metabolic balance studies: A comprehensive protocol and methodological approach for assessing intestinal function Authors: Julie Hvid Møller Andersen, Astrid Verbiest, Mark Hvistendahl, Tim Vanuytsel, Palle Bekker Jeppesen Published: 2026 ABSTRACT AbstractAdequate absorption of nutrients, fluids, and electrolytes is essential for maintaining physiological function and overall health. Clinical evaluation of the intestinal absorptive capacity is therefore important for diagnosing and managing chronic gastrointestinal disorders with impaired nutrient absorption. To this end, metabolic balance studies (MBS), which assess the intestinal absorption of energy, macronutrients, fluids and electrolytes, are considered the gold standard. However, full MBS are performed in only a few specialized centers due to their complexity and high resource demands. This paper provides a detailed methodological practice guide to conducting MBS, covering the entire process from standardized sample collection to laboratory procedures such as homogenization and freeze-drying, followed by quantitative analysis of energy, fat, carbohydrates, protein (nitrogen content), wet weight, and electrolytes. By offering a clear and reproducible framework, the aim of this guide is to support the broader implementation of MBS as an objective tool for assessing digestive and absorptive capacity in patients with malabsorption.Keywords: Metabolic Balance study, Intestinal failure, Short bowel syndrome, Absorption, Intestinal function Introduction One of the functions of the gastrointestinal tract is to adequately absorb nutrients, fluids, and electrolytes. Adequate dietary intake, digestion, and absorption are essential to meet or exceed an individual’s nutritional needs, thereby preserving optimal body composition, function, and overall health [1]. Gastrointestinal disorders may present with various symptoms, including abdominal pain, vomiting, dyspepsia and changes in bowel habits and may lead to weight loss, dehydration and electrolyte disturbances resulting from an nutritional imbalance [2]. It can be challenging to determine whether the imbalance is due to inadequate nutrient intake (oral failure), an inability to compensate for malabsorption, or underlying pathophysiological conditions that elevate nutritional requirements. One such condition is intestinal failure (IF), where the ability of the intestines to absorb sufficient fluid, electrolytes and macronutrients is severely impaired. This condition can lead to malnutrition and other secondary health complications. To better manage these conditions, metabolic balance studies (MBS) [3] is the gold standard for assessing intestinal absorptive function [4], [5]. These studies, used in both clinical and research settings, provide critical insights into the functionality of the gastrointestinal tract by measuring the balance between oral intake (from meals & beverages) and output (fecal/stoma output and urine). The absorptive capacity is assessed through multiple readouts, including energy absorption, macronutrient absorption (fat, carbohydrate, protein) as well as wet weight and electrolyte absorption. While full MBS are considered the most informative, they are only conducted in a few specialized centers worldwide due to the need for dedicated resources, laboratory facilities, personnel, equipment, and expertise. This explains why many centers have attempted to simplify evaluations often relying on food diaries and/or focusing only on limited analyses of fecal or urine output. This paper provides a comprehensive guide for conducting a full metabolic balance study and performing the associated analyses. Drawing on the + 50 years expertise of a leading MBS center, this instruction manual aims to provide unexperienced centers with limited MBS knowledge the necessary information to implement and carry out these studies, ultimately aiming to improve patient care and advancing research in the area of measuring intestinal absorptive function. Fig. 1 Flow diagram illustrating the steps involved in a metabolic balance study. Metabolic balance sample collections A metabolic balance period typically lasts from two to four consecutive days (48–96 h), during which the patient is admitted to the hospital ward. During this period, identical duplicates of oral intake, as well as fecal/stoma output and urine excretions are collected. On a daily basis, the patient is provided with a standardized set of collection materials, comprising two 2.5 L containers designated for urine and beverages collection, and two 5 L buckets allocated for fecal/stoma output, and the duplicate of the meal consumed. If the volume of intake or output exceeds standard capacity, additional containers or buckets are provided accordingly. The metabolic balance period commences in the morning, after an overnight fast. Prior to the initiation of the collection period, patients are instructed to empty their bladder and, if possible, to defecate. Patients are instructed to prepare two identical portions of each meal by individually weighing all dietary components. To facilitate accurate replication of their oral intake, critical for accurate collection and reliability of subsequent analyses, patients are provided with a calibrated electronic kitchen scale (± 1 g accuracy) and measuring cups (±50 mL accuracy). One portion is consumed, while the duplicate is collected in the designated 5-liter collection bucket. In cases where a meal is not entirely consumed, the equivalent uneaten portion must be removed from the duplicate to ensure that only the ingested amount is represented in the analysis. A similar protocol is followed for beverages: two identical volumes are measured using the provided measuring jug; one is ingested, and the second is transferred to a 2.5-liter collection container using a funnel. This procedure is to be repeated for every instance of oral intake throughout the collection period. Patients are encouraged to replicate their habitual dietary patterns during the metabolic balance study to ensure representative intake data. Only edible portions should be included in the collection – inedible parts such as packaging, fruit pits or peels are excluded. Pharmaceutical preparations and medical products are likewise excluded from duplication and should not be collected. Fecal/stoma output is collected directly into the designated 5 L bucket, which can be positioned within the toilet bowl for ease of use. Toilet paper should not be collected. Urine is collected separately in a designated 2.5 L container. To facilitate accurate collection, patients are provided with a measuring jug for direct urine collection and a funnel to enable easy and hygienic transfer into the container. To reduce the risk of contamination or mixing between fecal and urinary outputs, patients are advised to urinate frequently throughout the collection period. In cases where patients are unable to reliably separate urine from fecal output—such as in individuals with incontinence or high-volume diarrhea -the temporary placement of a urinary catheter may be considered to ensure the integrity of specimen collection. Collected samples of meals beverages and urine are stored at ambient room temperature in the hospital room throughout the 24-hour collection period. In contrast, fecal or stoma output should be stored in a bedside freezer at −20°C to prevent microbial fermentation and degradation of analytes [6]. At the end of each 24-hour collection cycle, all specimens are transferred to the laboratory for further processing. Upon arrival, samples of oral intake and urine are refrigerated at 2–8°C, while fecal or stoma output is kept at −20°C. On the evening prior to homogenization, all collected fecal or stoma output is removed from the freezer and allowed to thaw at room temperature overnight in preparation for subsequent analysis. Table 1, Table 2, Table 3, Table 4, Table 5, Table 6, Table 7 Table 1Overview of steps in the homogenization process.StepMealsBeveragesFecal/stoma outputUrineCollectionDuplicate collected in 5 L bucket.Duplicatecollected in2.5 L container.Collected directly in 5 L bucket.Stored at –20 °C.Collected in2.5 L container.Pre-treatmentHeated together and gently stirred continuously until the fat-rich components are melted.Thawed overnight at room temperature prior to processing.Container manually shaken.HomogenizationThe mixture is blended for 3 min. Starting at low speed and gradually increasing to high speed.Homogenized directly in collection bucket using digital homogenizer (low speed).Not applicable.AliquotingTransferred into a 100 mL flask + a 20 mL scintillation vial.Transferred into a 100 mL flask + a 20 mL scintillation vial.Transferred into a 20 mL scintillation vial. Overview of steps in the homogenization process. Table 2Overview of steps in the freeze-drying process.StepProcedureInitial material and weighing stepsHomogenized material is transferred to round-bottomed flask:Flasks are weighed empty, with wet material and after drying.Rapid freezingFlask containing wet material is rotated in liquid nitrogen.LyophilizationThe frozen flask is placed under vacuum in freeze-dryer for ≥ 48 h.Powder preparationDried material is crushed to fine powder and stored in scintillation vials at room temperature.Result expressed asDry matter (%) and dry matter (g/day). Overview of steps in the freeze-drying process. Table 3Overview of energy analysis and key steps for determination of energy content.Method & analytical principleBomb calorimetry.Combustion of samples under high-pressure resulting in heat release and a temperature rise proportional to the energy content.What is measuredHeat released from sample combustion (J/g).Sample typeFreeze-dried powdered material of meals & beverages and fecal/stoma output.ControlsBenzoic acid (crystalline form).ReplicationSamples and control are measured as duplicates. Acceptance of 700 J/g between duplicates.Final result expressed asEnergy in kJ/day. Overview of energy analysis and key steps for determination of energy content. Table 4Overview of fat analysis and key steps for determination of fat content.Method & analytical principleModified version of Van de Kamer titration method.Alkaline saponification of triglycerides, hexane extraction of released fatty acids, followed by titration.What is measuredFree fatty acids released from triglycerides.Sample typeFreeze-dried powdered material of meals & beverages and fecal/stoma output is preferred. Wet material can be used as well.ControlsGlycerol tristate and freeze-dried fecal sample spiked withtriglycerides.CalibrationTitration of a hydrochloric standard series to establish base normality.ReplicationSamples, controls and standards are measured as duplicates.Acceptance of ≤ 20% deviation between duplicates.Final result expressed asFat content in mmol/day and g/day. Overview of fat analysis and key steps for determination of fat content. Table 5Overview of carbohydrate analysis and key steps for determination of carbohydrate content.Method & analytical principle3,5-dinitrosalicylic acid (DNS) method.What is measuredReducing sugars released from carbohydrates.Sample typeFreeze-dried powdered material of meals & beverages and fecal/stoma output.ControlsPlantago.CalibrationA five-point standard series is used to establish a calibration factor.ReplicationSamples, controls and standards are measured as duplicates.Acceptance of ≤ 20% deviation between duplicates.Final result expressed asCarbohydrate content in g/day. Overview of carbohydrate analysis and key steps for determination of carbohydrate content. Table 6Overview of protein analysis and key steps in the determination of protein content.Method & analytical principleKjeldahl method.Digestion of organic nitrogen to ammonium ions, distillation of ammonia into flask with boric acid, and back-titration with hydrochloric acid.What is measuredNitrogen content.Sample typeFreeze-dried powdered material of meals & beverages and fecal/stoma output.Controls5% Alanine solution, ammonium chloride and distilled water (blank).Method controlAlanine and blank are digested with samples to verify digestion. Ammonium chloride and blank are not digested to verify distillation.ReplicationSamples and controls are measured as duplicates.Acceptance of ≤ 20% deviation between duplicates.Final result expressed asProtein content in g/day. (conversion factor 6.25) Overview of protein analysis and key steps in the determination of protein content. Table 7Overview of electrolyte analysis and key steps in the determination of electrolyte content.Method & analytical principleFlame Atomic Absorption Spectrometry.Samples are aspirated into a flame, where atoms absorb light specific to each element. The absorbance is proportional to the concentration.Sample typeHomogenized wet material of meals & Beverages and fecal/stoma output and urine.ControlsFluid from infusion bag with known electrolyte concentrations.CalibrationStandard solutions of each element are used to create calibration curves.ReplicationSamples and controls are measured as duplicates.Final result expressed asElectrolyte content in mmol/day. Overview of electrolyte analysis and key steps in the determination of electrolyte content. Quantification of wet weight and homogenization of collections Metabolic balance collections require thorough processing to obtain a homogeneous mixture that accurately represents the entirety of the 24-hour collection period. Given the presence of diverse nutrients and electrolytes within the collected materials, thorough homogenization is critical to ensure the reliability and validity of subsequent qualitative and quantitative analyses. To quantify the daily intake and output of the patient, each 24-hour collected sample is weighed using a calibrated, leveled scale with an accuracy of ±1 g. The weight of the empty container (including the lid) is subtracted from the total weight to determine the net weight of the collection. All measurements are systematically recorded on the homogenization worksheet (see supplementary materials for worksheet template). Fecal/stoma output The collected fecal output is homogenized directly within the collections bucket using a digital homogenizer (e.g., IKA T50 Ultra Turrax, IKA-Werke, Germany) operated at the lowest speed (setting 1). To minimize the risk of splashing and ensure operator safety, a custom-designed lid with a central opening for the homogenizer shaft is utilized during the process (Suppl. Figure 2). Throughout homogenization, the bucket is manually lifted and gently rotated to promote uniform mixing of the contents. If required, distilled water may be added to facilitate dilution of the fecal/stoma output sample and achieve a uniform, smoothie-like consistency suitable for accurate handling and aliquoting. The exact volume of distilled water added must be recorded on the homogenization worksheet, as it is necessary for subsequent calculations of dry matter content. The resulting thoroughly mixed sample is referred to as the homogenate, which serves as the analytical specimen for downstream measurements. Once the desired consistency has been achieved, aliquots of the homogenate are transferred using a calibrated measuring spoon into a 100 mL flask and a 20 mL scintillation vial for storage and analysis. Meals and beverages Collected meals and beverages are combined to represent the total daily oral wet weight intake. As food components tend to adhere to the inner surface of the collection bucket, homogenization is performed in three steps. First, the entire meal collection along with half of the collected beverages is heated together in a cooking pot on an electrical stove. Heating facilitates the emulsification of fat-rich foods, such as butter and cheese. To prevent burning, the mixture must be stirred continuously with a spoon during the heating process. Once the butter and cheese are visibly melted, the stove is turned off, and the mixture is transferred to a heavy-duty stainless-steel blender (e.g., Waring ® Commercial, Pennsylvania USA). The mixture is blended for approximately three minutes – starting at low speed and gradually increasing to high speed to ensure thorough homogenization. Meanwhile, the remaining beverages are heated in the pot on the turned-off stove. The heated beverages are then poured into the now-empty collection bucket to dissolve any residual food adhering to the sides. Once all residues are dissolved, the liquid is transferred to the blender and blended with the initial mixture for an additional two minutes. Additionally, distilled water may need to be added to obtain the desired consistency if the liquid content is insufficient. If the volumes exceed the capacity of the blender, the homogenization is performed in batches. The contents are transferred back and forth between the bucket and blender to ensure a thorough homogenized and representative final mixture. Like the fecal/stoma output collection, the homogenized mixture of meals & beverages is transferred into a 100 mL flask and a 20 mL scintillation vial. As the mixture tends to produce foam, care must be taken to avoid transferring any foam into the containers. Urine Urine collection containers are shaken manually before a portion is poured into a 20 mL scintillation vial. If additional analyses are required, a sample from the urine collection can be sent to the central laboratory. All scintillation vials are stored at –20 °C, while the contents of the 100 mL flask are subjected to freeze-drying. Freeze-drying The homogenized samples undergo lyophilization, beginning with rapid freezing using liquid nitrogen, followed by vacuum drying to remove water by sublimation, resulting in a powdered material. The contents of the 100 mL flasks are transferred into a 1 L round-bottomed flask, which is then mounted on an overhead stirrer (e.g., Hei-TORQUE core, Heidolph, Germany) and positioned at a 45-degree angle (Suppl. Figure 3). The flask is rotated in a Styrofoam container filled with liquid nitrogen until the sample is completely frozen – typically indicated by audible ice cracking sounds. Once fully frozen, the flask is connected to a freeze-dryer (e.g., Heto LyoLab 3000, Thermo Scientific, USA) and placed under vacuum for a minimum of 48 h. During this period, the ice sublimates, leaving behind dry material suitable for powdering. The round-bottom flasks are weighed at three stages: when empty, when filled with the wet sample and after freeze-drying. The dry matter content is determined by comparing the weight of the dried sample to the weight of the wet sample, with the weight of the empty flask subtracted in both cases. The result is then multiplied by 100 to express the dry matter as percentage. To calculate the dry matter in grams per day (g/day), the dry matter percentage is multiplied by the total weight of the homogenate and then divided by 100. This yields an accurate estimate of the solid content remaining after water removal. Powder generation After a minimum of 48 h of freeze-drying, the dry material is removed from the flaks using a spatula and transferred into a resealable plastic bag. To facilitate transfer, the resealable plastic bag is placed securely around the neck of the inverted flask, allowing the dry material to fall into the bag with gentle shaking. Once transferred, the bag is sealed, and the contents are manually crushed into fine powder. A small corner of the bag is then cut to allow for easy transfer of the powder directly into scintillation vials. The vials are stored at room temperature for subsequent analysis of energy and macronutrient content. Determination of total energy content Total energy content of oral intake and fecal/stoma output is determined using bomb calorimetry (e.g., C 6000, IKA-Werke, Germany) on freeze-dried samples. The sample material is combusted inside a vessel, or “bomb”, under high oxygen pressure. This vessel is immersed in an insulated reservoir filled with water. The heat released during combustion is transferred to the surrounding water, resulting in a temperature increase. The change in water temperature is directly proportional to the energy content of the sample [7]. To ensure consistency, all samples from a complete metabolic balance period should ideally be measured in a single continuous run. If repeated MBS have been performed at baseline and follow-up, the corresponding samples should be analyzed together to avoid a batch effect. Each analytical run begins with the determination of benzoic acid, which serves as a quality control, with a known enthalpy of combustion of 26 460 J/g [8]. The recommended sample weight for analysis is 0.3 g, though this may vary depending on the caloric density of the sample. For low-density but high-caloric samples, a larger volume may be required to achieve the desired weight. However, this can result in a high local energy density during combustion, which in some cases may cause the quartz dish to rupture. The sample material is weighed directly into a quartz combustion dish and placed inside the calorimetric vessel. All sample weights are documented on the energy worksheet (see supplementary material). A cotton thread is fastened to the ignition wire using a cow hitch knot and placed in contact with the sample, serving as a fuse to initiate combustion (Suppl. Figure 4). The resulting combustion heat is expressed in joules per gram (J/g). Each sample, including the quality control, is measured in duplicate. A deviation of ≤ 700 J/g between duplicates is considered acceptable. If the deviation exceeds 700 J/g, a third measurement should be performed. The daily energy content is calculated by multiplying the mean energy value of the duplicates by the sample’s dry matter (g/day) and dividing it by 1000 to obtain a value in kJ/day. Determination of fat content Fat content is determined using a modified version of the Van de Kamer titration method from 1949 [9]. Although both freeze-dried and wet samples can be used, freeze-dried material is recommended for improved consistency and accuracy. Quality control and sample preparation Glycerol tristearate, a triglyceride, is used as a quality control standard due to its consistent and well-defined fat content. Additionally, in our center, a freeze-dried fecal sample spiked with triglycerides to reach a target of 30 g fat per 100 g is included as a control to account for matrix effects. All samples and controls are prepared in duplicate in screw-cap glass tubes (e.g., Pyrex). For each tube, 30–50 mg of sample or fecal control, and 10 mg of glycerol tristearate are weighed directly into the tubes. A volume of 1 mL 33% potassium hydroxide (KOH) is added to each tube, and the rack is gently shaken to ensure that all material is thoroughly drenched. The alkaline solution initiates saponification, converting triglycerides into free fatty acids [9]. To facilitate dissolution, 4 mL of an ethanol:isoamyl alcohol mixture (ratio 24:1) is added. Tubes are sealed tightly with screw caps to prevent evaporation during boiling. It is important to reuse the same cap with each tube throughout the procedure. The tubes are shaken manually and then placed in a boiling water bath (99.9 °C). After 15 min, the rack is shaken again. After a total of 30 min, the tubes are removed from the water bath and allowed to cool to room temperature. Acidification and extraction A volume of 1.7 mL 6 M hydrochloric acid is added to each tube, causing an exothermic reaction and steam release. This acidifies the mixture, stopping the saponification reaction and precipitating the free fatty acids. Then, 5 mL of hexane is added to each tube, one by one, before the cap is screwed back on. All samples are then shaken by hand for thorough mixing and centrifuged at room temperature for five minutes at 600 g, resulting in a visible phase separation. Hexane acts as a solvent, extracting the free fatty acids to the upper phase. In the final step of sample preparation, 2.5 mL of the upper (hexane) phase is transferred into soda glass test tubes. Color development and titration setup To each tube, 1 mL of 99.9% ethanol and 50 µL of thymol blue are added, producing an orange to pink coloration. Simultaneously, standards for a titration series are prepared in duplicate using three concentrations of 0.1 M hydrochloric acid (0.25 mL, 0.5 mL and 0.75 mL) in soda glass tubes. Ethanol is added to each to reach a final volume of 3.5 mL (i.e., 3.25 mL, 3.0 mL ad 2.75 mL, respectively). Each standard also receives 50 µL of thymol blue, resulting in a bright pink color. Titration procedure Samples and standards are titrated using a burette filled with a titrant composed of isobutanol and potassium hydroxide (see supplementary material for preparation). Titration continues until a distinct color change to blue indicates the endpoint. The titration begins with the first duplicate of each standard, followed by the samples, and concludes with the second duplicate of each standard. Calculations of fat content The titration series is used to calibrate the method by establishing the relationship between the volume of hydrochloric acid and the amount of titrant required for neutralization. Potassium hydroxide in the titrant reacts with the free fatty acids in the same manner as it does with the hydrochloric acid in the standards. Based on the titration of the standards, a base normality for each standard is calculated by multiplying the volume of hydrochloric acid used with its concentration, dividing the result by the volume of titrant used, and finally multiplying by 1000. The mean of these base normality values is further used to determine the concentration of free fatty acids in the samples. To calculate the theoretical value for the control, mass of the weighed material (mg) is divided by the the molecular weight (0.8915 mg/mmol) [10]. This result is then multiplied by three, as glycerol tristearate contains three fatty acid chains. The fat content of the sample is calculated based on the titration volume, where the volume of KOH is multiplied by 2 and the base normality, and this is divided by the weight of the sample material. The result is then multiplied by the sample's dry matter (g/day). A conversion factor of 0.27 g/mmol is used, corresponding to an estimated average molecular weight for the free fatty acids in the sample. A deviation of up to 20% between duplicate measurements is considered acceptable. If the deviation exceeds this threshold, the analysis should be repeated. Determination of carbohydrates content The determination of carbohydrate content is based on the 3,5-dinitrosalicylic acid (DNS) method which detects reducing sugars [11]. As the human diet contains various complex carbohydrates, these must first be hydrolyzed into reducing sugars to be measured. This is achieved by adding acid and applying heat, which breaks down polysaccharides into monosaccharides. The resulting reducing sugars, which possess free aldehyde or ketone groups, react with DNS. When heated in alkaline conditions, DNS is reduced to 3-amino-5-nitrosalicylic acid, producing a red/brown color. The intensity of this color is directly proportional to the concentration of reducing sugars and is quantified spectrophotometrically (12). Sample preparation Freeze-dried material is used for the determination of carbohydrates, with Plantago serving as a quality control. All samples and controls are prepared in duplicate in screw-cap glass tubes (e.g., Pyrex). The following amounts are weighed directly into the tubes: •Fecal/stoma sample: 50 mg Fecal/stoma sample: 50 mg •Meals & beverages sample: 25 mg Meals & beverages sample: 25 mg •Plantago control: 20 mg Plantago control: 20 mg All weights are documented on the corresponding worksheet (see supplementary material). To each tube, 1 mL of 12 M sulfuric acid is added. The caps are tightly secured, and the tubes are gently shaken to ensure complete soaking of the sample. The samples are then incubated at 35 °C for 1 h. After 30 min, the rack is shaken again to promote uniform hydrolysis. After the 1-hour incubation, the caps are removes, and 5 mL of distilled water is added to each tube. The caps are then screwed back on, and the samples are boiled at 99.9 °C in a water bath for one hour. The rack is shaken midway through boiling. After the full hour, the samples are removed and allowed to cool to room temperature. Sample transfer and reagent addition The contents are then transferred to new screw-cap glass tubes, avoiding any undissolved material that may accumulate at the surface. Using a new pipette tip for each sample, 200 µL is transferred to a clean tube. The following reagents are then added to each tube. •200 µL of 4 M sodium hydroxide 200 µL of 4 M sodium hydroxide •400 µL of 3.5-dinitrosalicylic acid reagent 400 µL of 3.5-dinitrosalicylic acid reagent •20 µL of glucose solution 20 µL of glucose solution See Supplementary Material for reagent preparation. Preparation of calibration standards A five-point calibration series is prepared in soda glass tubes, starting with a blank and followed by four dilutions of increasing sugar concentrations. The standards are first prepared in soda glass test tubes before being transferred into new screw-cap glass tubes. 1.Blank: Contains 2 M sulfuric acid only. Blank: Contains 2 M sulfuric acid only. 2.Standard 1 (highest concentration): 5000 µL of 2.4 M sulfuric acid + 1000 µL sugar mix. Standard 1 (highest concentration): 5000 µL of 2.4 M sulfuric acid + 1000 µL sugar mix. 3.Standards 2–4: Prepared by serial dilution—1000 µL of the previous standard is added to 2 M sulfuric acid, halving the concentration at each step. Standards 2–4: Prepared by serial dilution—1000 µL of the previous standard is added to 2 M sulfuric acid, halving the concentration at each step. From each standard, 200 µL is transferred to new tubes, followed by the same reagent additions as for the samples. Color development and measurement Caps are screwed onto all glass tubes, which are then boiled for 15 min at 99.9 °C in a water bath. After boiling, the tubes are cooled to room temperature. A progressive color change should be visible in the standards, from light yellow (low concentration) to deep red (high concentration). Samples and standards are measured spectrophotometrically at 540 nm [12]. A calibration factor is established from the standards by dividing the known concentration by the average absorbance (minus the absorbance of the blank). The mean of these individual factors is then used to calculate the concentration of reducing sugars in the samples. The carbohydrate concentration in the samples is calculated by dividing the calibration factor by the sample weight (mg). The result is then multiplied by the measured absorbance (ext), the dry matter content (g/day), and the final volume of 6 mL. A deviation of up to 20% between duplicate measurements is considered acceptable. If the deviation exceeds this threshold, the analysis should be repeated. Determination of protein content The determination of protein content in the samples is performed using the Kjeldahl method, which measures total nitrogen as an indirect estimate of protein concentration [13]. In this method, organic nitrogen in the sample is converted to ammonium ions by boiling with concentrated sulfuric acid, which breaks down proteins into amino acids [13], [14]. After digestion, excess sodium hydroxide is added, converting ammonium to ammonia, which is then distilled and captured in an Erlenmeyer flask containing boric acid. The collected ammonia is quantified by back-titration with hydrochloric acid, and the volume of acid consumed corresponds to the nitrogen content of the sample [15]. Sample preparation and digestion For each analysis, 100 mg of freeze-dried material is weighed and transferred into 100 mL Kjeldahl digestion tubes. Quality control samples are included in each run, consisting of alanine in aqueous solution (positive control) and distilled water (blank), as detailed in the Supplementary Material. To each digestion tube, two Kjeltabs (containing potassium sulfate and selenium) [16] and 10 mL of concentrated sulfuric acid are added. Kjeltabs function as catalysts, increasing the boiling point and enhancing oxidation to promote complete digestion of the sample [17]. Using a Digestor Auto Heating Block and Scrubber (FOSS, Denmark), samples are boiled for 1.5 h at 420 °C, until turning completely clear. Residual color indicates incomplete digestion and requires further boiling [17]. Distillation and ammonia capture After cooling, each sample is individually distilled using the Kjeltec™ 8100 system (FOSS, Denmark). This device automatically dispenses sodium hydroxide and distilled water into the digestion tube. During distillation, ammonia is released and condensed into an Erlenmeyer flask containing 25 mL of 4% boric acid solution, with methyl red and bromocresol green as indicators (see Supplementary Material). The presence of ammonia causes a visible color change in the boric acid solution, ranging from grey-purple to green depending on ammonia concentration. Each analytical run begins and ends with a blank and a quality control (ammonium chloride in aqueous solution), both of which have not undergone digestion. By comparing results from digested and undigested controls, the performance of both the digestion system (heating block) and the distillation unit is validated. Titration and calculations The ammonia-containing solution is back-titrated with 0.1 M hydrochloric acid. The titration endpoint is reached when the solution changes to a bright pink/red color, indicating complete neutralization and correlating with the nitrogen content in the sample. Nitrogen concentration is calculated by subtracting the mean titration volume of the blank samples from that of each test sample. The resulting volume is then multiplied by the atomic weight of nitrogen in mmol/mL (1.401), multiplied by the dry matter content, and finally divided by the original sample weight to determine the nitrogen content. To convert the content of nitrogen to total protein content, the result is multiplied by a factor of 6.25 [18]. A deviation of up to 20% between duplicate measurements is considered acceptable. If the deviation exceeds this threshold, the analysis should be repeated. Determination of electrolyte content Electrolyte concentrations are quantified using a Flame Atomic Absorption Spectrometer (e.g., PinAAcle 500, PerkinElmer Inc., Denmark). At the reference center, calcium, magnesium, potassium, and sodium are routinely measured as part of electrolyte analysis. Principle of Detection Liquid samples are aspirated and nebulized into the flame, where solvent evaporation and thermal dissociation produce free atoms. A hollow cathode lamp, specific to each element, emits light at its characteristic wavelength. The atoms in the flame absorb this radiation, and the remaining light is filtered by a monochromator and quantified by a detector. The absorbance signal is directly proportional to the element’s concentration in the sample [19], [20]. Sample material and overview Analyses are performed on standards, blanks, quality control materials, and biological samples (feceal/stoma meals & beverages, and urine) in solution form. For this method, wet homogenized material stored at –20 °C is used. Preparation of standard solutions Standard stock solutions are prepared using Titrisol® concentrated ampoules (Merck KGaA, Darmstadt, Germany). Sodium and potassium standards are prepared as a combined solution, while calcium and magnesium are prepared individually. •Calcium/Magnesium: The contents of each specific ampoule are mixed with 50 mL of 1 M hydrochloric acid and diluted with distilled water in a volumetric flask to a final volume of 500 mL [21]. Calcium/Magnesium: The contents of each specific ampoule are mixed with 50 mL of 1 M hydrochloric acid and diluted with distilled water in a volumetric flask to a final volume of 500 mL [21]. •Sodium/Potassium (combined): The contents of both ampoules are mixed with 25 mL of 1 M hydrochloric acid and diluted with distilled water to a final volume of 250 mL [21]. Sodium/Potassium (combined): The contents of both ampoules are mixed with 25 mL of 1 M hydrochloric acid and diluted with distilled water to a final volume of 250 mL [21]. Serial dilutions are then prepared from the stock solutions to achieve the desired concentrations (see Supplementary Materials). Stock solutions and dilutions can be stored under refrigeration. A blank solution is prepared by mixing 50 mL of 1 M hydrochloric acid with distilled water to a total volume of 500 mL. A fluid infusion bag (e.g., Plasmalyte) containing known electrolyte concentrations is used as an additional quality control to verify analytical accuracy. If no single bag contains all desired electrolytes, more than one bag may be used. Sample preparation Samples are prepared in soda glass test tubes as follows: •For urine, standards, blanks, and controls: 500 µL is pipetted into each tube. For urine, standards, blanks, and controls: 500 µL is pipetted into each tube. •For fecal and meal/beverage samples: 500 mg of thawed material is weighed directly into the tubes. For fecal and meal/beverage samples: 500 mg of thawed material is weighed directly into the tubes. Step 1: Acid digestion Each tube receives 6 mL of 1 M hydrochloric acid, after which the tubes are sealed and stored overnight at 4 °C. This step facilitates decomposition of organic material and releases electrolytes. Step 2: Solvent extraction The next day, 6 mL of a chloroform/methanol mixture (ratio 2:1) is added to each tube to extract soluble components. After thorough mixing, samples are centrifuged at 600 g for 5 min, resulting in a phase separation (chloroform bottom phase, methanol top phase). Step 3: Element-specific sample processing The supernatant is used for element-specific dilutions: •Calcium: 500 µL supernatant + 5 mL lanthanum-strontium solution Calcium: 500 µL supernatant + 5 mL lanthanum-strontium solution •Magnesium: 500 µL supernatant + 10 mL of 1% hydrochloric Magnesium: 500 µL supernatant + 10 mL of 1% hydrochloric •Potassium: 50 µL supernatant + 10 mL of 1% hydrochloric acid Potassium: 50 µL supernatant + 10 mL of 1% hydrochloric acid •Sodium: 100 µL supernatant + 12.5 mL of 1% hydrochloric acid Sodium: 100 µL supernatant + 12.5 mL of 1% hydrochloric acid Flame atomic absorption analysis Once the instrument is turned on, the element of interest is selected, and the acetylene-oxygen flame is ignited. A calibration curve is generated from standards of increasing concentrations and is considered valid if the correlation coefficient (R²) exceeds 0.95. Measurement sequence 1.Calibration standards Calibration standards 2.First quality control (duplicate) First quality control (duplicate) 3.Sample duplicates Sample duplicates 4.Final quality control Final quality control Acceptance criteria for standards, controls, and duplicates are provided in the Supplementary Materials. Calculation of electrolyte concentrations To correct for dilution effects in fecal samples, the final concentration of each electrolyte is calculated by dividing the mean measured concentration (from duplicate measurements) by the dilution factor. The dilution factor is defined as the weight of the feces divided by the total weight of the homogenate. The calculation for the meals & beverages and urine is based on the mean value of the duplicate measurements, provided the material is not diluted. To express the result in mmol/day, the concentration in mmol/L is multiplied by the weight of the material divided by 1000. Discussion Until recently, full metabolic balance studies had only been conducted at a single reference center; Rigshospitalet in Copenhagen. This procedure provides a unique and detailed insight into the intestines ability to absorb nutrients and can help determine whether the patient is absorbing the nutrients they consume. MBS differs significantly from simpler methods such as 24-hour fecal collections, food diaries and food frequency questionnaires. The latter often underestimate actual oral intake [22]. Instead, MBS applies a direct and quantitative approach by analyzing the balance between what is ingested and what is excreted. This allows to evaluate whether nutritional needs of the patient are met, whether enteral/parental nutrition is required, or if an existing treatment needs adjustment. Although the method provides important insight into the intestinal function, limitations remain. Digestion and absorption are complex biological processes involving multiple factors such as enzymatic breakdown, nutrient transport and utilization as well as microbial fermentation, which cannot be fully quantified. Therefore, despite its strengths, MBS should always be interpreted within a broader clinical context that considers these physiological complexities. Further analyses, such as metabolomics [23] and stable isotope testing [22] could provide valuable insights into the metabolic pathways and nutrient utilization, helping to refine our understanding of intestinal function. The most important element in metabolic balance studies is the accuracy of sample collection, and it is essential that the patient follows the instructions carefully. The collection of sample material forms the foundation of the entire study, and any errors can compromise the validity of the result. The collection of beverages and meals must be performed with great precision. If the patient eats without first weighing the food and collects a ”visually identical portion” for reference, the recorded intake can either be over- or underestimated. If the reference portion is transferred to the bucket immediately, but the patient does not finish their own portion, the intake will be overestimated. Accurate and separate collection of urine and feces is essential to ensure valid results. The material must be collected separately ensuring no material is accidentally discarded in the toilet, as even small losses can impact calculations of excretion and thereby distort the balance between intake and output. Since the study relies heavily on patient compliance, there is also a risk of intentional or unintentional deviations from the protocol. For instance, the patient may misreport meal intake or fail to collect all urine and feces, which can falsely suggest either improved or worsened health status. Therefore, it is important that study staff remain observant and maintain clear communication with the patient throughout the study period. For the same reason, it is not advisable to perform the MBS at the home of the patients. Full MBS require a four-day inpatient admission and impose a non-negligible burden on patients, including intensive sample collection, dietary duplication, and temporary restrictions on work, normal daily activities, and social and family life. From an ethical perspective, their use is justified only when the anticipated diagnostic or therapeutic benefit clearly outweighs patient inconvenience and resource utilization, particularly in complex clinical cases and research settings where precise quantification of absorptive capacity is expected to influence specific management decisions, including the evaluation and regulatory approval of new therapies. Acceptability is generally high in experienced centers when thorough pre-study education and continuous staff support are provided but may be lower in frail individuals or in patients with limited self-care capacity, mobility, or cognitive ability to adhere to detailed instructions. Consequently, MBS should be regarded as a highly specialized investigational tool, applied selectively and preferably complemented by simpler assessments (e.g. wet-weight balance and fecal fat measurements), rather than used as a routine diagnostic procedure. A collaborative effort between the reference center at Rigshospitalet, Denmark, and an inexperienced IF center in Leuven, Belgium, was initiated with the aim of implementing MBS at the Belgian center. The process was time-consuming and labor-intensive, taking four years and involving close cooperation and peer training between the centers. This collaboration resulted in a transferability study demonstrating that, with thorough training and close collaboration, MBS can be successfully implemented in new centers without prior experience [24]. Institutional costs are substantial, reflecting personnel time, dedicated laboratory infrastructure, and specialized analytical equipment, thereby restricting feasibility to high-volume referral centers. Establishing a fully equipped laboratory capable of performing all analyses required for a full MBS represents a major financial investment. In the transferability study, the cost of laboratory equipment alone exceeded EUR 200,000. In addition to this investment, substantial time and dedication are required to implement the analytical setup, including extensive staff training. Several years may be needed before personnel achieve the level of proficiency required to perform the analyses reliably and reproducibly. The execution of MBS is also resource intensive. Each study period requires up to four days of hospitalization, resulting in costs associated with bed occupancy and clinical support. This is followed by approximately one additional week of laboratory work dedicated to sample processing and analysis, requiring personnel whose time is specifically allocated to the study. Salary expenses and hospitalization costs may vary considerably between institutions and countries. The current article aims to reduce the need for such extensive peer training by providing a detailed and structured guide to support more independent implementation. By consolidating knowledge, practical considerations, and analytical tools into a single resource, this practice guideline offers a solid foundation that may minimize the requirement for prolonged external support and help streamline the setup process in new centers. Conclusion This article provides all the necessary information for implementing MBS in practice. It includes a comprehensive description of the method, associated analyses, and practical considerations. In the supplementary material, details on the required utensils, reagents, recipes for solutions, as well as a ready-to-use Excel spreadsheet containing the relevant equations are provided. The methodology has already been successfully transferred to a new, previously unexperienced center as a part of a transferability study, demonstrating that MBS can be implemented outside the original reference center. Using this protocol as a framework, it is possible to establish MBS and obtain reliable measurements of intestinal function, and thereby contributing to both improved research and enhanced patient care. CRediT authorship contribution statement Mark Hvistendahl: Conceptualization. Tim Vanuytsel: Conceptualization, Supervision, Writing – review & editing. Palle Bekker Jeppesen: Conceptualization, Supervision, Writing – original draft, Writing – review & editing. Astrid Verbiest: Writing – review & editing, Writing – original draft, Visualization, Validation, Supervision, Project administration, Methodology, Investigation, Formal analysis, Data curation, Conceptualization. Julie Hvid Møller Andersen: Writing – review & editing, Writing – original draft, Visualization, Validation, Project administration, Methodology, Investigation, Formal analysis, Conceptualization. Ethical statement N/A Declaration of Generative AI and AI-assisted technologies in the writing process During the preparation of this work the authors used ChatGPT in order to assist with language and phrasing. After using this tool/service, the authors reviewed and edited the content as needed and take full responsibility for the content of the published article. Funding N/A. Patient’s / Guardians’s consent N/A. Declaration of Competing Interest The authors declare no conflict of interest. Acknowledgement TV is supported by a senior clinical research fellowship of the Flanders Research Foundation (FWO; 1830522N) Footnotes Supplementary data associated with this article can be found in the online version at doi:10.1016/j.intf.2026.100373. Supplementary material Supplementary material Supplementary material References Associated Data Supplementary Materials Supplementary material Supplementary material