• To determine the clinical effectiveness of pre-operative nutrition in a homogenous group of patients
• To determine whether nutritional benefits were confined to severely malnourished patients

• Newly detected, histologically confirmed gastric or colorectal cancer requiring surgical treatment
• No previous treatment for other malignant tumors
• Presence of nutritional depletion
  • Determined by Nutritional Index (NI), where:
    • NI = (0.14 x ALB)+(0.03 x %IBW)+(0.73 x TLC)
    • ALB = albumin; IBW = Ideal Body Weight; TLC = Total Lymphocyte Count 
  • < 1.31 considered to be nutritionally depleted

• > 80 years
• Normal nutritional status

Recruitment:

• Not specifically mentioned; all patients were from a single treatment center
• Study protocol was reviewed and approved by the Human Studies Committee of the University Hospital Maastricht

Design:

• Subjects were stratified for percent weight loss (PWL) (<,> 15%), age (<,> 65 years), and tumor localization (gastric or colorectal)
• Following stratification, depleted patients were randomized to one of three groups:
  • Pre-operative parenteral nutrition (TPN group)
  • Pre-operative enteral nutrition (TEN group)
  • Surgery without delay (depleted control group)
• Non-depleted patients (NI > 1.31) served as a non-randomized reference group, and underwent surgery without delay

Blinding used (if applicable):

• Not mentioned

Intervention (if applicable):

Patients were randomized to one of three groups:

• TPN group: Received 150% basal energy expenditure (BEE) as non-protein kcals, calculated using the Harris-Benedict equation
  • Solution contained 7 g nitrogen/I (Synthamin 14) and 25% dextrose
  • Electrolytes were added according to individual patient's needs
  • 500 ml fat emulsion (Intralipid 20%) were administered 3x/week
  • Average pre-operative energy intake was was 1783 (+ 350) kcals/day; 11.9 (+ 2.5) g nitrogen/day; 9.4 (+ 5.7)% fat; N/kcal ratio 1:168 (+ 20)
  • Patients received 0.97 - 1.25 g/kg body weight protein/day
  • Pre-operative nutrition support was received for 11.8 (10-23) days
  • Post-operative parenteral nutrition support was continued post-operatively until patients were consuming 120% BEE per oral diet
• TEN group:
  • Received either Precitene or Isotein by either nasogastric tube or by mouth
  • Energy intake was 150% calculated BEE
  • Average pre-operative energy intake was 1458 (+ 444) kcals/day; 20.5 (+ 5.5) g nitrogen/day; 27.4 (+ 8.3)% fat; N/kcal ratio 1:102 (+ 26)
  • Patients received 1.90 - 2.09 g/kg body wt protein/day
  • Pre-operative nutrition support was received for 11.4 (10-15) days 
  • Post-operatively, patients in the TEN group were allowed increasing amounts of foods and liquids as tolerated
  • Parenteral nutrition support was started post-operatively only in the case of serious post-operative complications
• Control group
  • Received surgery without delay 
  • Received no preoperative nutrition support
  • Underwent surgery without delay
• In addition, a non-randomized, non-depleted group of patients served as a nutritionally "normal" reference group.
• Complications were monitored daily, and were evaluated separately.
  • Patients were categorized within each trial group as having:
    • No complications
    • At least one minor, but no major, complications
    • At least one major complication
  • Mortality was evaluated separately.

 Statistical Analysis:

• All data was analyzed using the intent-to-treat principle
• Data were described as means and standard deviations
• Chi-square and Mantel Haenszel Chi-square tests were used for complication frequency in the trial groups
• ANOVA was used to compare the distribution of risk factors between trial groups
• P < 0.05 was considered significant

Timing of Measurements:

• Data was measured as complications occurring during each patient's hospital stay.

Variables:

• Complications unrelated to nutrition support:
  • Minor complications:
    • Wound infection - Positive culture or need for surgical drainage
    • Urinary tract infection - Positive culture > 10E7 micro-organisms per ml urine
    • Pulmonary tract infection - Positive sputum culture and abnormal X-ray requiring treatment with antibiotics
  • Major complications:
    • Wound dehiscence - Dehiscence of wound requiring surgical closure
    • Anstomotic leakage - Clinical or radiologic diagnosis, necessitating re-operation or drainage
    • Fistula - Clinical diagnosis and proven by fistulography
    • Intra-abdominal abscess - Purulent fluid collection, demonstrated by ultrasound or CT-scan guided drainage or at re-operation
    • Sepsis - Positive blood culture, persisting temperature above 39 degrees C, associated with hypotension and hyperperfusion
    • Respiratory insufficiency - Vent assistance for >12 hours post-operative or renewed mechanical ventilation after detubation
    • Circulatory insufficiency - Clinical and radiologic diagnosis requiring cardiotonics and/or diuretics or MI confirmed of ECG and by serum enzyme elevations
    • Renal insufficiency - Plasma urea levels elevated by 50% or plasma creatinine elevation by 20%
• Complications related to nutrition support
• Mortality  

Independent Variables:

• Surgical treatment for gastric or colorectal carcinoma
• Nutrition support

Control Variables:

• Control group
• Non-depleted reference group

Initial N:

• N = 200
  • TPN = 51 ( 29M, 22F)
  • TEN = 50 ( 32M, 18F)
  • Control group = 50 ( 32M, 18F)
  • Non-depleted reference group = 49 (33M, 16F)

Attrition (final N):  All patient data included in final analysis

Age:

• TPN = 67.3 (+ 10.2)  (range 41-80)
• TEN = 65.7 (+ 9.3)  (range 39-78)
• Control group = 65.8 (+ 7.5)  (range 49-79)
• Nutritionally "normal" reference group = 61.3 (+ 9.5)  (range 39-77)

Ethnicity:

• Not mentioned

Other relevant demographics:

• All groups were comparable in blood loss during surgery, and of duration in surgical procedure

Anthropometrics:

• Trial groups were comparable in terms of age, % weight loss, ratio gastric/colorectal cancer, % ideal weight, % weight loss,  albumin, and total lymphocyte count
• The TPN group had a higher number of patients with advanced disease (stage IV cancer)
• The reference group had a higher % ideal weight, albumin, and total lymphocyte count, and a lower % weight loss; this was expected in light of the way this group was selected 

Location:

• Department of Surgery, University of Maastricht, the Netherlands

Complication rate in each of the study groups:

 Minor and major complications by group:

Number of complications occurring in each patient:

Note:  Fewer patients in the TPN and TEN groups had more than one complication when compared to the control group.  If the number of complications is weighted (that is, 1 complication = 1; 2 complications = 2; 3 or more complications = 3) the difference between groups is significant, and TPN/TEN seem to reduce the number of complications per patient compared to the depleted group.

Complication rate of septic complications in patients with > 10% loss of body weight

*p < 0.05 TPN vs TEN vs Control vs Non-depleted

^p < 0.05 TPN + TEN vs Control

Note: There was a significant decrease in intra-abdominal abscess in the TPN and TEN groups when compared with patients displaying severe depletion.

Complication rate of septic complications in patients > 500 ml blood loss during surgical procedure

*p<0.05 TPN vs TEN vs Control vs Non-depleted

^p<0.05 TPN + TEN vs Control

#p<0.05 Control vs Non-depleted

Note:  Differences between the trial groups and the control group were amplified in depleted patients suffering severe blood loss.

Other Findings

• Complications of nutrition support:
• TPN group:
• 1 arterial puncture and 1 pneumothorax related to catheter insertion procedure
• 4 clinically proven episodes of catheter sepsis, requiring removal of the catheter
• TEN group:
• 3 cases of diarrhea requiring treatment
• 3 patients with vomiting
• 2 patients with delayed gastric emptying
• There were no cases of ileus or aspiration

• Hospital stay was shorter in the non-depleted reference group than in the trial groups, as might be expected (p< 0.01).
• Total stay TPN group: 36.3 (+ 17.7) days
• Total stay TEN group: 33.3 (+ 20.2) days
• Total stay control group: 31.7 (+ 22.2) days
• Total stay Non-depleted group: 24.8 (+ 11.3) days
• Total stay in TPN and TEN groups was not significantly longer than the control group, despite longer pre-operative stays in the nutrition support groups.
• Post-operative stay was significantly prolonged in patients with minor or major complications (p<0.001)
• There was no benefit from preoperative nutrition on hospital stay for patients with minor and major complications as a group.
  • In patients with major complications, there appeared to be a shorter stay in the TPN and TEN groups, although this trend was not statistically significant.

The authors conclude that:

• Preoperative nutrition support reduces sepsis and intra-abdominal abscess complications in severely depleted patients undergoing surgery for gastric or colorectal cancer
• Only patients with signs of severe nutritional depletion, defined as weight loss > 10%, should receive nutritional support prior to surgery, and that the benefit of such support increases with the degree of initial depletion
• Reduction of complications in severely depleted patients is not related to increased morbidity from administration of nutrition support 
• Nutritional depletion increases the risk of perioperative complications, but not all depleted patients develop complications; it is, therefore, unreasonable to expect that all post-operative complications can be eradicated by nutrition support

In addition, the authors discuss limitations of their study:

• The "Nutritional Index" (NI) used to identify nutritional depletion uses serum albumin, total lymphocyte count, and percent ideal body weight to figure degree of depletion.
  • All of these measures can reflect disease process and fluid status, as well as nutritional status.
  • The NI may not be an accurate indicator of nutrition status.

The authors suggest that their findings may possibly extend to other forms of cancer treatment such as chemotherapy, radiotherapy, and surgery for other types of cancer.

This is one of the few trials to test the effects of preoperative nutrition support in depleted cancer patients that actually had a control group of depleted patients receiving immediate surgery, and that used standard parenteral and enteral feedings.

Useful data was obtained by stratifying patients for weight loss as well as blood loss during surgery.

It is unknown if 10 days is the optimal length of time to offer preoperative nutrition to depleted cancer patients.

Energy needs were estimated using the Harris-Benedict equation.

Depletion was defined by a "Nutritional Index" equation, using serum albumin, total lymphocyte count, and percent ideal weight as indicators of nutritional status, and may not accurately reflect actual nutritional status since these indicators may reflect disease progression as well.