Anda di halaman 1dari 5


NCPXXX10.1177/0884533614522833Nutrition in Clinical PracticeEdakkanambeth Asfaw et al

Clinical Observations
Nutrition in Clinical Practice
Volume 29 Number 2
How Hyperalimentation May Be Necessary to Reverse April 2014 229­–233
© 2014 American Society
Severe Malnutrition in Selected Patients Receiving Home for Parenteral and Enteral Nutrition

Parenteral Nutrition DOI: 10.1177/0884533614522833
hosted at

Jithinraj Edakkanambeth Varayil, MD1,2; Ryan T. Hurt, MD, PhD1,3; and

Darlene G. Kelly, MD, PhD4,5

Standard predictive equations may under- or overestimate caloric requirements in disease states such as obesity or in patients with a low
body mass index (BMI). Although this principle is common knowledge among nutrition specialists, it is often not prioritized with other
clinicians outside the intensive care unit (ICU). Indirect calorimetry (IC) is often used in the ICU to estimate caloric requirements. This
article outlines a very complicated case of a cachectic man with an enterocutaneous fistula who had lost more than 50% of body weight
over 2 years. In rehabilitating this patient, we found that the most common formulas of basal needs greatly underestimated the calories
required to prepare him for restorative surgery. Key learning points are that in malnourished ambulatory patients, predictive equations
may not adequately estimate caloric needs and IC may be required. (Nutr Clin Pract. 2014;29:229-233)

home nutrition support; nutritional support; calorimetry; parenteral nutrition; enteral nutrition; nutrition assessment

Background [BMI], 13.2 kg/m2), blood pressure of 92/61 mm Hg (sitting),

and pulse of 100 bpm. His abdomen was scaphoid with a large
Over the past year, we encountered an extremely malnourished enterocutaneous fistula that had associated skin breakdown, as
patient in whom the various equations for calculating caloric well as multiple healed surgical scars (Figure 1A). He had a
needs and usual ways of evaluating laboratory findings were significantly decreased muscle mass and fat store as shown in
very misleading. This case is described below. the computed tomography (CT) scan (Figure 2A).
A 62-year-old man presented to our clinic for evaluation of a
longstanding jejunocutaneous fistula and weight loss. He had a
history of gastric cancer that was resected with Billroth II gas- Investigations
trectomy in 2004 with subsequent chemotherapy and radiation His laboratory studies at the time of admission showed a nor-
therapy. In 2010, he had proximal bowel ischemia and gangrene, mal complete blood count, except that the mean corpuscular
leading to several reconstructive surgeries of the gastrojejunos- volume (MCV) was borderline low with an elevated red blood
tomy anastomosis, resulting in conversion of his Billroth II into cell (RBC) distribution width (18.7%), suggestive of iron defi-
a Roux-en Y gastrojejunostomy. These repetitive abdominal sur- ciency (iron, 25 mcg/dL; normal, 50–150 mcg/dL). Hemoglobin
geries led to skin breakdown and a fistula. He stated that he had decreased from 13.3 g/dL to 8.6 g/dL in the first 3 days as a
lost 45.3 kg from a baseline of 81 kg, amounting to 56% weight
loss in the past 3 years. Since then, he has been hospitalized mul-
tiple times elsewhere with sepsis, aspiration pneumonia, and
other complications. These events as well as his decreased appe- From 1Department of Internal Medicine, Mayo Clinic, Rochester,
tite and extreme fluid losses from the fistula led to his poor nutri- Minnesota; 2Second Department of Internal Medicine, Charles University
tion status. He had received enteral nutrition (EN) support at Faculty of Medicine, Hradec Kralove, Czech Republic; 3Department
times (details of the volume and calories unavailable), but he of Internal Medicine, University of Louisville, Kentucky; 4Division of
continued to lose weight. The patient described some dysphagia Gastroenterology and Hepatology, Mayo Clinic, Rochester, Minnesota;
and 5Oley Foundation, Albany, New York.
but denied nausea, vomiting, or abdominal pain. He reported an
anxious mood for the past 2 years. He had no history of liver or Financial disclosure: None declared.
kidney disease. He was a smoker with a 55-pack-year history but This article originally appeared online on February 25, 2014.
denied heavy alcohol use. He received parenteral nutrition (PN)
Corresponding Author:
elsewhere for 1 week prior to coming to our clinic. Ryan T. Hurt, Division of General Internal Medicine, Mayo Clinic, 200
On initial examination, he appeared weak and fatigued. He First Street SW, Rochester, MN 55905, USA.
weighed 35.7 kg with a height of 165.2 cm (body mass index Email:
230 Nutrition in Clinical Practice 29(2)

Figure 1.  (A) Abdomen at the start of home parenteral nutrition (HPN) shows extreme cachexia with absence of subcutaneous fat and
jejunocutaneous fistula with a feeding tube near the costal margin. (B) Abdomen after 6 months of HPN shows partial resolution of
cachexia with restoration of some subcutaneous fat but with persistence of the jejunocutaneous fistula and peristomal erythema.

Figure 2.  (A) Abdominal computed tomography (CT) at the beginning of home parenteral nutrition treatment showing a clear lack
of subcutaneous fat and muscle tissue (arrow). (B) Abdominal CT 6 months later clearly showing restoration of subcutaneous fat and
muscle tissue (arrow).

result of correction of dehydration. The abnormal electrolyte Treatment

levels included sodium, 131 mmol/L; potassium, 5.5 mmol/L;
creatinine, 1.1 mg/dL; and serum urea nitrogen, 61 mg/dL An upper endoscopy and placement of a Surgisis (Cook
(normal, 8–24 mg/dL). Other parameters of nutrition status Surgical, Bloomington, IN) fistula plug with both abdominal
that were abnormal included serum albumin, 2.3 g/dL (normal, wall suturing and intraluminal endoscopic suturing was per-
3.5–5.0 g/dL); phosphate, 2.0 mcg/mL (normal, 2.5–4.5 mcg/ formed in an attempt to close the fistula. In addition, a nasoje-
mL); zinc, 0.51 mcg/mL (normal, 0.66–1.10 mcg/mL); and junal feeding tube and a stent across his gastrojejunostomy was
magnesium, 1.4 mg/dL (normal, 1.7–2.3 mg/dL). placed. The patient tolerated nasojejunal feeding with a goal of
Resting energy expenditure (REE) was measured to esti- 60 mL/h × 24 hours using Osmolite 1.0 (Abbott Nutrition,
mate caloric expenditure for basal requirements given that his Columbus, OH).
BMI was well below the normal range. Using the Harris- He also received nasogastric (NG) decompression, and the
Benedict (HB) equation, we estimated his basal caloric require- patient was discharged home with a suction device. He reported
ments to be 939 kcal/d. Using indirect calorimetry, the REE diarrhea (from 1–3 stools per day to 8–13 per day) with his
was measured to be 1270 kcal/d, with VCO2 of 167 mL/min, continuous nasojejunal enteral feedings and blocked NG
VO2 of 177 mL/min, and respiratory quotient (RQ) of 0.95. decompression tube. He was readmitted 3 days later with
Edakkanambeth Varayil et al 231

Table 1.  Various Validated Tools for Calculating BEE, Predictions in Our Patient, and Comparison of REE Determined With Indirect

Variation From Indirect

Equation Formula Calculated BEE/REE, kcal/d Calorimetry, %
Harris-Benedict equation8 66.5 + 13.8(wt-kg) + 5(ht-cm) – 6.8(age) 939 –26
Mifflin equation9 5 + (10 × Act BW) + (6.25 × Ht) − (5 × age) 1070.7 –16
Ireton-Jones equation10 629 – 11 × age (y) + 25 × W (kg) – 609 × 0 839.5 –34
ACCP guidelines11 (BMI <25): Act BW × 25 892.5 –30
Owen equation1 879 + (10.2 × Act BW) 1243 –2
WHO/FAO/UNU2 31–60 y: 11.6 × W (kg) + 879 1293 +2

ACCP, American College of Clinical Pharmacology; BEE, basal energy expenditure; BMI, body mass index; BW, body weight; REE, resting energy
expenditure; WHO/FAO/UNU, World Health Organization/Food and Agriculture Organization/United Nations University.

increased NG output. On readmission, nasogastric decompres- stabilized, further decisions could be made to wean off the
sion was accomplished with a new pump and frequent saline infusions.
flushing to prevent obstruction. However, an extended upper
gastrointestinal (GI) study with water-soluble contrast showed
high-grade small bowel obstruction and most of the contrast
Discussion and Clinical Lessons
refluxing through the gastrojejunostomy into the stomach. His case posed a unique challenge, and a combined team
After this, PN was started, and he stayed nil per os. He was sent approach with experts from multiple fields has been required
home with an average daily intake of 1787 kcal (HB + 90%, for successful management. The use of typical guidelines
but measured REE + 40%) with fat emulsion 5 days per week would have presented several potential opportunities for clini-
(50.1 kcal/kg/d). This level was chosen to provide about 500 cal errors.
kcal in excess of REE (1270 kcal), which would be expected to
increase the weight by about 1 pound per week. 1. Equations for estimation of basal metabolic rate/REE may
fail in some patients.  This extremely malnourished patient’s
actual basal caloric expenditure was markedly underestimated
Outcome and Follow-Up by 4 of the frequently applied formulas (Table 1). In fact, these
During the following months, the patient’s nutrition status equations underestimated his REE by 16%–34%. This was
improved remarkably. At the 1-month follow up, he had gained demonstrated by the fact that the tube feedings and even prior
4 kg with total calories of 1644 kcal/d (HB + 75%/REE + PN formulas calculated using predictive equations were asso-
29%), lipids 3 d/wk. His serum albumin level was 2.6 g/dL, ciated with continued weight loss. Only 2 equations, the Owen
1.8 g/dL, and 2.9 g/dL after 1 month, 3 months, and 6 months equation1 and the World Health Organization/Food and Agri-
of home PN (HPN), respectively. Because of the improvement culture Organization/United Nations University (WHO/FAO/
in the subcutaneous fat as evidenced by his abdominal wall UNU),2 approximated the measured value (–2% and +2%,
improvement (Figure 1B and Figure 2B), the patient was able respectively). While it is generally understood that such equa-
to undergo placement of a tunneled single-lumen Hickman tions are typically designed to meet needs of most patients, the
catheter line at 6 months. Eight months after initial HPN evalu- cachectic and markedly obese patients are frequent exceptions
ation, the patient had gained almost 12.8 kg. The indirect calo- and may require indirect calorimetry to determine actual
rimetry was repeated after treating the infection during the caloric requirements.3 This is largely related to the fact that
6-month visit. At that time, the results were the following: cachectic patients are universally nearly devoid of fat mass,
measured REE was 1041 kcal/24 h, VCO2 was 146 mL/min, making the muscle mass (albeit depleted) a higher percentage
VO2 was 132 mL/min, and RQ was 0.91. The estimated basal of actual body weight. In such individuals, the ratio of muscle/
energy expenditure (BEE), based on the Harris-Benedict equa- fat determines the basal metabolic rate (BMR). Since muscle is
tion at that time, was 1082 kcal, within 4% of the REE a more metabolically active tissue than fat, the caloric expendi-
measurement. ture is higher than expected solely by height and weight. The
After 11 months, a surgical procedure was performed to reverse is true in the markedly obese.
take down the complex jejunocutaneous fistulae and recon- Several studies have reported increased caloric require-
struct the complicated post–Roux-en-Y abdomen, with cre- ments in severely malnourished patients: 42.3 ± 4.4 kcal/kg/d
ation of a duodenojejunostomy, gastrojejunostomy, and by Matarese et al4 and 50 kcal/kg/d by Hill et al.5 Our patient
complex abdominal wall closure. Two months following the received 50.1 kcal/kg/d at the beginning of HPN, after we had
procedure, the patient was still receiving HPN. Once been assured that refeeding syndrome was not occurring. After
232 Nutrition in Clinical Practice 29(2)

Table 2.  Weight, Calories in Tube Feeding, and HPN Formula and Laboratory Results During Treatment.

After 6 mo With Catheter Removal

Initial +2 Days of Staphylococcus + 5 Days of
Characteristic Evaluation Hydration Start of HPN After 1 mo After 3 mo aureus Antibiotics
Weight, kg 35.7 37.4 35.9 35.6 43.6 44.8 —
Serum albumin — 2.3 2.6 — 1.8 2.9 2.3
(3.5–5.0 g/dL)a
Amount of calories — 1440 1787 1644 1885 (at 4.5 mo) 1742 —
provided (= REE + 13%)
Alkaline phosphatase — — 77 480 146 406 155
(45–155 ULN)a (0.6 X ULN) (4.2 X ULN) (0.9 X ULN) (3.5 X ULN) (1.3 X ULN)
AST (8–48 U/L)a — — 21 — 31 55 17
ALT (7–55 U/L)a — — 9 — 17 59 8

AST, aspartate aminotransferase; ALT, alanine aminotransferase; HPN, home parenteral nutrition; REE, resting energy expenditure; ULN, upper limit of
Normal reference range in parentheses.

his weight loss had been reversed, the Harris-Benedict equa- 5. In patients with an enterocutaneous fistula, care must be
tion was a much better estimate of his requirements determined used to determine whether tube feedings are a component of
by indirect calorimetry. effluent volume. This must be considered in trying to correct
nutrition status. This patient had fistula outputs that were
2. Initially obtained labs of the malnourished will be erroneous directly related to infusion of his tube feeding. The output
in the dehydrated.  Most laboratory values are heavily affected ranged from 1858 mL at one point during enteral feedings
by dehydration, which results in hemoconcentration. In evalu- down to as low as 25 mL when feedings were stopped. Subse-
ating the clinical status of patients, these may be misleadingly quent radiological studies demonstrated that there was a tight
reassuring. For example, in the first 2 days, this patient’s stricture in the small bowel distal to the tube and that contrast
hemoglobin dropped by 4.7 g/dL, with correction of (thus enteral feedings) refluxed from the gastrojejunostomy
dehydration. back into the gastric remnant. This precluded enteral feedings
for this patient, which might have been suspected based on the
3. Laboratory studies may not be what they appear. This extremely large effluent volumes.
patient had a major elevation in his liver enzymes on 2 differ-
ent occasions. Often, these can be associated with a decreased
serum albumin. The typical response from clinicians who are
not extensively trained in nutrition is that liver dysfunction and There are certain nutrition clinical scenarios in which standard
malnutrition are the clinical problems. However, this can be a predictive equations may under- or overestimate caloric
warning sign that there is a bloodstream infection. In this requirements (eg, obesity or malnourished). Nutrition support
patient, the laboratory studies demonstrate this nicely. When teams will often recommend indirect calorimetry for such
he returned after 6 months of HPN, his serum albumin had actu- patients in the intensive care unit, but its use in the ambulatory
ally dropped to 2.3 g/dL, and he had gained only 0.34 kg/wk setting is often limited. This case demonstrates that what is
(Table 2). Blood tests at the time of his 6-month return identi- expected to be normal in the general population may not be
fied that the alkaline phosphatase and transaminases were both normal in those who are severely malnourished. Furthermore,
markedly elevated. Blood cultures were positive for Staphylo- it often requires interaction between multiple disciplines to
coccus aureus. Over the next 5 days, his catheter was removed successfully manage such patients.
and he was treated with antibiotics. The liver enzymes were
nearly completely corrected after effective antibiotic treat- References
ment. This was demonstrated and presented in a prior study.6 1. Owen OE, Holup JL, D’Alessio DA, et al. A reappraisal of the caloric
requirements of men. Am J Clin Nutr. 1987;46(6):875-885.
4. Monitoring patients on nutrition support must include more 2. Energy and protein requirements: report of a joint FAO/WHO/UNU
than laboratory studies.  Body temperature, urine output, and Expert Consultation. World Health Organ Tech Rep Ser. 1985;724:
weight trends are all critical parameters to be monitored, par-
3. Neelemaat F, van Bokhorst-de van der Schueren MA, Thijs A, Seidell JC,
ticularly in those receiving long-term support. In this patient, Weijs PJ. Resting energy expenditure in malnourished older patients at
his continued weight loss despite nutrition support could have hospital admission and three months after discharge: predictive equations
identified compromised caloric intake earlier in his course.7 versus measurements. Clin Nutr. 2012;31(6):958-966.
Edakkanambeth Varayil et al 233

4. Matarese LE, Steiger E, Seidner DL, Richmond B. Body composition 8. Harris JA, Benedict FG. A biometric study of human basal metabolism.
changes in cachectic patients receiving home parenteral nutrition. JPEN Proc Natl Acad Sci U S A. 1918;4(12):370-373.
J Parenter Enteral Nutr. 2002;26(6):366-371. 9. Mifflin MD, St Jeor ST, Hill LA, Scott BJ, Daugherty SA, Koh YO. A
5. Hill GL, Bradley JA, Smith RC, et al. Changes in body weight and body protein new predictive equation for resting energy expenditure in healthy indi-
with intravenous nutrition. JPEN J Parenter Enteral Nutr. 1979;3(4):215-218. viduals. Am J Clin Nutr. 1990;51(2):241-247.
6. Van Gossum AM, Arvanitakis M, Ballarin A, Lievin V. Long-term follow- 10. Ireton-Jones C, Jones JD. Improved equations for predicting energy
up of cholestasis in home parenteral nutrition (HPN) patients: prevalence, expenditure in patients: the Ireton-Jones equations. Nutr Clin Pract.
severity, factors, outcome. Gastroenterology. 2009;136(5):A538-A538. 2002;17(1):29-31.
7. Hess RA, Welch KB, Brown PI, Teitelbaum DH. Survival outcomes of pedi- 11. Uehara M, Plank LD, Hill GL. Components of energy expenditure in
atric intestinal failure patients: analysis of factors contributing to improved patients with severe sepsis and major trauma: a basis for clinical care. Crit
survival over the past two decades. J Surg Res. 2011;170(1):27-31. Care Med. 1999;27(7):1295-1302.