5.11 Routes and Timing of Non-Oral Nutritional Interventions
5.11.1 Routes of Nutrient Administration
In the early stages of recovery a significant percentage of patients will be comatose and mechanically ventilated; precluding oral feeding. While enteral feeding is the preferred route of nutrient administration, feeding intolerance due to gastroparesis and ileus are common. Enteral feeding has been associated with a decrease in bacterial translocation and a reduced incidence of infection.
Enteral feeding intolerance may be related to increase intracranial pressure (Ott et al., 1990). Medications may also play a role in delayed gastric emptying. Although the placement of feeding tubes into the small bowel may theoretically improve tolerance, placement can be difficult and empirical evidence of superiority is lacking. If intolerance is prolonged, parenteral feeding may be indicated (Cerra et al., 1997) although the risk of hyperglycemia and cerebral edema are increased (see Table 5.20).
Individual Studies
Table 5.20 Enteral Nutrition vs. Total Parenteral Nutrition
|
Author/ Year/ Country/ Study design/D&B and PEDro Score |
Methods |
Outcome |
|
Rapp et al., (1983) |
N=38 Head-injured patients were randomized to receive total parenteral nutrition (TPN) or standard enteral nutrition (SEN). The mean admission GCS Score was 7.2 for the SEN group and 7.7 for the TPN group. Nutritional and clinical data was obtained until either death or 18 days hospitalization. Patients’ functional recovery was monitored for up to one year. |
Within an 18-day period, 8 of the 18 patients died in the SEN group compared to no patient deaths in the TPN group (p<0.0001). The TPN group had a significantly greater intake in nitrogen then the SEN group (0.002). Therefore the overall nitrogen balance was significantly different for the TPN and SEN group (p=0.002). |
|
Hadley et al., (1986) |
N=45 Head trauma patients with a mean admitting GCS of 5.8 were randomly assigned to receive either total parenteral nutrition (TPN) or enteral nutrition (NG). Patients received high nitrogen and calorie feedings for a 14-day period of the study to try to obtain a positive nitrogen and calorie balance. Nitrogen loss was measured every other day |
Patients who received TPN achieved significantly higher mean daily nitrogen intakes (p<0.01) and losses (p<0.001) compared to the patients who received NG. However, there was no significant between-group difference in nitrogen balance. |
|
Young et al., (1987) |
N=96 Severely head injured patients with a GSC score between 3-10 were randomized to receive either parenteral nutrition (TPN) or enteral feeding (EN) and the study lasted from hospital admission to 18 days following injury. |
Intracranial pressure was >20 mmHG in 75% of the TPN patients and 73% of the EN patients. Standard therapy failed to control elevated ICP in 36% of the TPN group and 38% of the EN group. There were no significant between-group differences in Serum osmolality. Although mean serum glucose levels were higher in the TPN group for the first 13 days following injury than EN group who had increased mean serum glucose content after 13 days. |
|
Borzotta et al., (1994) |
N=59 Prospective, randomized non-blinded clinical comparison of adult head injury patients with GCS < 8 and coma persisting for > 24 hours (severe ABI patients). 23 patients were treated with early parenteral nutrition which at day 5 began conversion to gastric feeding with tapering of TPN while 21 patients were treated with enteral feeding through a J-tube. |
Two groups were comparable at baseline. No significant differences noted for nitrogen excretion or balance, energy expenditures, meeting nutritional goals and infections. Patient charges were greater for TPN, hyperglycemia was more common in the TPN group (p<.05) as was the number of patients with diarrhea (p<.05). There were no differences in mortality at the end of follow up. |
|
Ott et al., (1999) |
N=57 Retrospective comparative study of patients with severe head injuries; GCS of <11. 30 patients received enteral nutrition while 27 received total parental nutrition (TPN). |
Overall cost of enteral nutrition was $170 per patient per day and parental nutrition was $308. |
D&B = Downs and Black (1998) quality assessment scale score.
PEDro = Physiotherapy Evidence Database rating scale score (Moseley et al., (2002).
Discussion
A Cochrane review authored by Yanagawa et al. (2002) identified six trials, which compared parenteral versus enteral nutrition. Parental feeding was associated with protection from both death and the combined outcome of death and disability, although the result was not statistically significant. The relative risk for mortality at the end of the follow-up period was 0.66 (0.41-1.07) while two trials noted the relative risk of death and disability as 0.69 (0.40-1.15). Young et al. (1992) noted that both parenteral and enteral feeding methods of nutrition administration safely and effectively reduce mortality and improve outcome following traumatic head injury.
Conclusions
Based on one RCT there is Level 2 evidence that parenteral nutrition is more costly compared to enteral nutrition.
There is conflicting evidence that choice of feeding route does not affect nitrogen balance or mortality in post-ABI patients.
Based on a single RCT, there is Level 2 evidence that TPN can safely be administered without causing serum hypersomolality or influencing intracranial (ICP) pressure levels or ICP therapy in post-ABI patients.
|
Parenteral feeds are safe; however, enteral feeds are less expensive and associated with fewer complications than parenteral feeds. Further research is needed to investigate the affect of both feeding routes on nitrogen balance and mortality.
|
5.11.2 Enhanced Enteral Nutrition
Enteral feeding solutions enriched with immune-enhancing nutrients may decrease the occurrence of sepsis and reduce the inflammatory response. Theoretically, glutamine may improve the nutrition of both the gut mucosa and immune cells, while probiotic bacteria could favourably alter the intraluminal environment, competing for nutrients and adhesion sites with pathogenic bacteria. These co-operative actions may reduce the rate of bacterial translocation and, thus, decrease both the incidence of infection and the length of hospitalization (Falcao de Arruda & Aguilar-Nascimento 2004).
Individual Studies
Table 5.21 Enhanced Feeding Solutions
|
Author/ Year/ Country/ Study design/ D&B and PEDro Score
|
Methods |
Outcomes |
|
Falcao de Arruda and Aguilar-Nascimento (2004) |
N=23 Prospective RCT of consecutive patients with brain trauma requiring ICU admission and enteral feeding. Treatment group (n=10) received early enteral feeding with glutamine and probiotics added. Control group (n=10) received only enteral feeds. |
Infection rate higher in control than treatment group (p=.03). In control group critical care unit stay significantly longer (p<.01) as was mechanical ventilation (p<.04). |
PEDro = Physiotherapy Evidence Database rating scale score (Moseley et al., 2002).
D&B = Downs and Black (1998) quality assessment scale score.
Conclusions
There is Level 1 evidence based on a single RCT that enhanced enteral nutrition can reduce the incidence of infection, and reduce both the ventilator dependency period and ICU stay.
|
Enhanced enteral feeds improve a number of outcomes. |
5.11.3 Timing of Enteral Nutrition
Early enteral feeding is desirable as a means to prevent intestinal mucosal atrophy and to preserve gut integrity, although, as previously noted, feeding intolerance occurs frequently. Three studies examined the effects of early vs. delayed enteral feeding (Table 5.22).
Individual Studies
Table 5.22 Timing of Enteral Feeding
|
Author/ Year/ Country/ Study design/ D&B and PEDro score |
Methods |
Outcomes |
|
Minard et al., (2000) |
N=30 Prospective randomized controlled study of adult patients < 65 with severe closed-head injuries (GCS >3 and <11 within 6 hrs of injury) admitted to a trauma center. Treatment group (n=12) received early enteral feeding (within 60 hrs of injury) while control group (n=15) received delayed enteral feeding. |
No significant differences between groups with regard to mortality, length of hospital or ICU stay, ventilator days, number of infections per patient or patients with pneumonia. |
|
Taylor and Fettes |
N=82 Head-injured patients with GCS above 3 were randomly assigned to receive either conventional enteral nutrition (EN) or early EN. Data was collected on nutritional intake, causes of interruption to EN, nitrogen balance, volume of gastric residuals and incidence of pneumonia. |
Patients receiving early EN had a greater median percentage of energy and nitrogen delivered compared to conventional EN patients over the initial week following brain injury (p<0.02). Overall, patients received EN during 57% of the potential feeding time. Intervention patients received a higher fluid volume (p<0.02) but did not have the higher incidence of pneumonia. |
|
Taylor et al., (1999) |
N=82 Prospective randomized controlled trial of head injury patients requiring mechanical ventilation who were randomized to a control group of standard enteral nutrition gradually increasing from 15 ml/hr up to the estimated energy and nitrogen requirements vs. enhanced enteral nutrition (started at a feeding rate that met estimated energy and nitrogen requirements from day 1). |
The intervention patients had a significantly higher mean percentage of energy (p=.0008) and nitrogen (<.0001) requirements met in the 1st week after injury when compared to the control group. This was achieved mostly by improved NG feeding as only 14 intervention patients (34%) had intestinal tubes successfully placed. The median percentage of energy and nitrogen requirements delivered in the control patient < 60% even by day 7 after injury. Neurologic outcome at 6 months was similar between the two groups, but there was a tendency for more intervention patients to have a good neurologic outcome at 3 months than control patients (61% vs. 39%; p=.08). Intervention patients had fewer infection complications (61% vs. 85%; p=.02) and earlier discharge. |
D&B = Downs and Black (1998) quality assessment scale score.
PEDro = Physiotherapy Evidence Database rating scale score (Moseley et al., 2002).
Discussion
A Cochrane review authored by Yanagawa et al. (2002) identified six RCTs, which addressed the timing to initiation of feeding and assessed mortality as an outcome. The relative risk for death associated with early nutritional support was 0.71 (95% CI 0.43-1.16). The pooled RR from three trials, which also assessed death and disability for early feeding was 0.75 (0.50-1.11). Although the results were not statistically significant, the authors concluded that early feeding may be associated with a trend towards better outcomes in terms of survival and disability.
Conclusions
There is Level 2 evidence based on a single RCT that initiating enteral feeding at goal rate will increase the percentage of prescribed energy and protein actually received.
|
Beginning enteral feeding at final rate increases the percentage of prescribed energy and protein actually received.
|
5.11.4 Timing of Parenteral Nutrition
Early parenteral nutrition support provided directly following injury could assist in the maintenance of immunocompetence and help reduce the frequency of the incidence of infection following acquired brain injury (Sacks et al., 1995). One study examined the outcome of early versus delayed parenteral feeding (Table 5.23).
Individual Studies
Table 5.23 Timing of Parenteral Feeding
|
Author/ Year/ Country/ Study design/ D&B and PEDro Score |
Methods |
Outcomes |
|
Sacks et al., (1995) |
N=9 Patients with a GCS of 3-12 were prospectively randomized to receive either early parenteral nutrition (PN) (n=4) at day 1 or delayed PN (n=5) at day 5. All patients received ON through a central venous catheter with a nutrient goal of 2 g protein/kg per day & 40 nonprotein kcal/kg per day for at least the initial 14 days of hospitalization. |
There was a significant increase in total CD4 cell counts and CD4% for early PN versus delayed PN at day 14 (p<0.05). From Baseline to Day 14, following Con A stimulation, an improved lymphocyte response was demonstrated in the early PN group (p<0.05). The CD4-CD8 ratio significantly increased from baseline to day 12 in the early PN group (p<0.05). |
D&B = Downs and Black (1998) quality assessment scale score.
PEDro = Physiotherapy Evidence Database rating scale score (Moseley et al., 2002).
Conclusions
There is Level 2 evidence that early parenteral nutrition support of closed head-injury patients appears to modify immunologic function by increasing CD4 cells, CD4-CD8 ratios, and T-lymphocyte responsiveness to Con A.
|
Early parenteral nutrition support of ABI patients appears to modify immunologic function.
|
5.11.5 Types of Enteral Feeding Tubes
Early enteral feeding has been associated with improved outcome. However, the effectiveness of the intervention may vary depending on the mode of feeding. Nasogastric feeding tubes have been associated with increased incidence of pneumonia, while, theoretically feeding tubes placed more remotely decrease the risk. Gastronomies are proved to be a safe and dependable process used to provide enteral access for meeting nutritional needs of ABI patients and delivering essential medications (Harbrecht et al.,1998).
Individual Studies
Table 5.24 Early Gastrostomy
|
Author/ Year/ Country/ Study design/ D&B and PEDro Score |
Methods |
Outcomes |
|
Kostadima et al., (2005) |
N=41 Acutely ill, ventilator dependent patients with a diagnosis of either stroke (n=25) or head injury (n=16) were randomized to receive a gastrostomy or to NG tube for enteral feeding. Tubes were inserted within 24 hours of intubation. Patients were followed for 3 weeks and the incidence of pneumonia was noted and compared between groups. A diagnosis of pneumonia was established using previously validated criteria. |
At the end of weeks 2 and 3 the cumulative incidence of pneumonia was significantly higher in the NG compared to the gastrostomy group (p<0.05). At the end of the first week the incidence of pneumonia was higher in the gastrostomy group although the result was not statistically significant.
Gastrostomy NG |
|
Grahm et al., (1989)
|
N=32 Brain injury patients with a GSC <10 were randomized to receive either nasojejunal (NJ) feeding tubes placed fluoroscopically (treatment) or gastric feeding when bowel sounds returned (control). |
There were no significant between-group differences for metabolic parameters. Daily caloric intake was higher on days 2-6 was higher among patients in the NJ group compared to the control group (p<0.05). On days 2-4, Nitrogen intake and daily nitrogen balance significantly improved improves for the treatment group versus the control (p<0.001). The incidence of bacterial infections and days of intensive care unit hospitalization were significantly reduced in both groups (p<0.05). |
D&B = Downs and Black (1998) quality assessment scale score.
PEDro = Physiotherapy Evidence Database rating scale score (Moseley et al., 2002).
Conclusions
There is Level 1 evidence that the risk of developing pneumonia is higher among ventilated patients fed by a naso-gastric tube compared with a gastrostomy tube.
There is Level 2 evidence that early naso jejunal hyperalimentation improves caloric intake, nitrogen intake, nitrogen balance, bacterial infection and days of stay in the intensive care unit in post-ABI patients.
|
There is an increased risk of developing pneumonia for ventilated patients fed by a naso-gastric versus a gastrostomy tube. Early jejunal hyperalimentation improves a number of outcomes in post ABI patients.
|
5.11.6 Metoclopramide and Enternal Feeding
Individuals who sustain a severe traumatic brain injury often show signs of gastroparesis. It has been noted that many with a severe ABI, their energy requirements may reach 60% more than predicted. Metoclopramide has been used and continues to be used despite the inconsistent findings supporting its use (Nursal et al., 2007). To enhance the effectiveness of enteral nutrition metoclopramide has been used with limited success (Nursal et al., 2007).
Individual Study
Table 5.25 Metoclopramide and Enteral Nutrition
| Author/ Year/ Country/ Study design/ D&B and PEDro Score |
Methods |
Outcomes |
|
Nursal et al., (2007) |
N=19 head-injured patients with GSC scores ranging from 4-11 were included in the RCT. Patients were administered either metoclopramide or a saline solution. The treatment group received 10 mg intravenously 3 x day for 5 days, while the control group was given a saline solution. |
Oral/enteral calories in relation to the total number of calories received during the first 5 days was higher for those in the control group. There was no significant difference between the groups when looking at the results of the paracetamol absorption test. When looking at absorption parameters, those in the treatment group had levels that were slightly more pronounced than those in the control group. |
D&B = Downs and Black (1998) quality assessment scale score.
PEDro = Physiotherapy Evidence Database rating scale score (Moseley et al., 2002).
Discussion
In one RCT, conducted by Nursal et al. (2007) the result of the paracetamol absorption test showed statistically significant differences between the control group and the treatment groups which had been receiving 10 mg of metocopramide. When looking at the absorption parameters of the two groups a small non-significant difference was found. This difference was more pronounced in the treatment group. Although this study demonstrated slightly favorable results, there has been some controversy over the benefits of using metoclopramide for this purpose (Altmayer et al., 1996).
Conclusion
There is Level 1 evidence indicating that metoclopramide is not effective as an aid to gastric emptying.
|
The therapeutic benefits of using metoclopramide to aid in gastric emptying are minimal.
|