- Email: [email protected]
Is early enteral feeding safe in patients who have suffered spinal cord injury?
Injury, Int. J. Care Injured (2004) 35, 238—242
Is early enteral feeding safe in patients who have suffered spinal cord injury? C.J. Rowana,*, L.K. Gillandersa, R.L. Paiceb, J.A. Judsonc a
Nutrition Services, Auckland Hospital, Auckland, New Zealand Trauma Services, Auckland Hospital, Auckland, New Zealand c Department of Critical Care Medicine, Auckland Hospital, Auckland, New Zealand b
Accepted 19 April 2003
KEYWORDS Enteral feeding; Spinal cord injury; Gastric aspirates
Summary Objective: To examine whether enteral feeding is a safe technique to use in the acute stage of spinal cord injury. Methods: We searched the departmental computerised patient database and clinical records for all patients with spinal cord injuries admitted to the Auckland Hospital Intensive Care Unit (ICU), known as the Department of Critical Care Medicine (DCCM), between January 1988 and December 2000. Patients were included in the study if they had suffered complete spinal cord transection resulting in either paraplegia or quadriplegia. Data was collected for the following variables: length of time to commence enteral feeding, type of enteral feeding, duration of enteral feeding and reasons for interrupting the feed. Results: Thirty-three patients were found and were included in the study. Twenty-seven (82%) of the patients commenced enteral feeding in the DCCM, 25 by nasogastric (NG) and 2 by nasojejunal (NJ) tube. Feeding was commenced a median of 2 days after admission and the median length of enteral feeding was 7.7 days. The main feeding complications that resulted in interrupting the feed were high gastric aspirates. One patient commenced on enteral feeding developed medical complications that prevented continuation. Two patients on NG feeding converted to NJ feeding. Conclusion: No major complications associated with enteral feeding were seen in this study. This would indicate that enteral feeding can be safely administered in the acute stage of spinal cord injury provided complications are monitored for daily. ß 2003 Elsevier Ltd. All rights reserved.
Introduction Spinal cord injury is a devastating event with farreaching functional and psychological effects on the survivor. Optimal medical and rehabilitative care is important for patients with such injuries to regain a sense of control over their life and maximise their independence and quality of life.14 Nutrition support is an important component of *Corresponding author. Tel.: þ64-9-638-6654; fax: þ64-9-263-8709. E-mail address: [email protected] (C.J. Rowan).
therapy especially in the acute stages following injury. For these patients early nutrition intervention is important to attenuate the protein loss that occurs in such injury. Many spinal cord injured patients have suffered other trauma that demands additional energy and protein requirements. For patients that require nutrition support, enteral feeding is preferred over total parental nutrition (TPN) because it maintains gastrointestinal function, results in fewer infectious and metabolic complications and is cheaper than TPN.3,4 Despite the importance of enteral nutrition, feeding is often withheld in spinal cord injured
0020–1383/$ — see front matter ß 2003 Elsevier Ltd. All rights reserved. doi:10.1016/S0020-1383(03)00203-1
Safety of early enteral feeding in spinal cord injury patients
patients for concerns about ileus and other complications.2 Some patients remain ‘nil by mouth’ with intravenous fluids until oral food intake is resumed after days or weeks. The purpose of this study was to examine whether enteral feeding is a safe technique to use in the acute stage of spinal cord injury.
Population and methods Auckland is a city of approximately 1.08 million people, divided into three health areas (Northwest Auckland, Central Auckland and South Auckland). Two trauma centres with well defined catchment areas serve these areas. Auckland Hospital, a 538 bed university affiliated tertiary referral hospital, covers Central Auckland and Northwest Auckland areas, and Middlemore Hospital covers the South Auckland area. At the 1996 census, a total of 739,926 people resided permanently within the Auckland Hospital trauma catchment area.12 The Auckland Hospital Intensive Care Unit (ICU), known as the Department of Critical Care Medicine (DCCM) was the only ICU providing care for trauma patients aged 15 years and over for the entire period of this study. Its computerised patient registry has included trauma variables since 1988. We searched the departmental computerised patient database and clinical records for all patients with spinal cord injuries admitted to the DCCM, in the 13 years from 1 January 1988 to 31 December 2000. All patients who had suffered clinical spinal cord transection resulting in either paraplegia or quadriplegia were included in the study. Date of birth, sex, date admitted to the DCCM, date transferred from the DCCM and level of injury were recorded for each patient from the clinical record. Feeding practices performed on Table 1
239
each patient during their DCCM stay were obtained from 24 h record sheets. These contain a record of all clinical interventions performed over a 24 h period and include the feeding regimen. Data was collected manually on a spreadsheet for the following variables: time to commence enteral feeding, type of enteral feeding started, either nasogastric (NG) or nasojejunal (NJ) and reasons for interrupting the feed. All patients were gravity fed and received a standard, polymeric feed with a nutrient density of 1 to 1.2 kcal/ml. Feeding was started at 3 h intervals and progressed to a continuous 24 h regimen. A retrospective, descriptive analysis was then performed. A literature review was conducted using Medline. The key words ‘spinal injuries’ and ‘enteral nutrition’ were used to source all relevant literature from 1 January 1990 to 31 December 2000.
Results Thirty-three patients with paraplegia or quadraplegia were identified. As shown in Table 1, they consisted primarily of men with a median age of 27 years, reflecting the high incidence of spinal cord injury in a young male population. Twenty-seven (82%) of the 33 patients were commenced on enteral feeding. Enteral feeding was commenced a median of 2 days after admission to the DCCM and patients were fed for a median of 7.7 days during their DCCM stay. Twenty-five patients were commenced on NG feeding and two of them were later converted to NJ feeding. Two patients were commenced on NJ feeding from the outset because of high gastric aspirates. Twenty-three out of 27 patients were transferred from DCCM on enteral feeding.
Patient demographics
Number of patients Age (median years (range)) Sex (female/male) Length of DCCM stay (median days (range)) Patients with paraplegia Patients with quadriplegia
33 27 (15—47) 2/31 8.8 (2.8—36.2) 20 13
Number of patients commenced enteral feeding Nasogastric (NG) feeding Nasojejunal (NJ) feeding from the outset
27 (82%) 25a 2
Time to commence enteral feeding (median days (range)) Number of times enteral feeding interrupted per patient (median days (range)) Duration of enteral feeding in DCCM (median days (range)) Number of patients transferred from DCCM on enteral feeding
2 (0.5—4.8) 2 (0—20) 7.7 (0.2—34.6) 23
a
Two of these changed to nasojejunal (NJ) feeding.
240
C.J. Rowan et al.
Table 2 Reasons for not commencing and stopping enteral feeding Number of patients Reason for not commencing feeding Commenced oral diet 4 Ileus 1 Patient transferred to another hospital 1 Reason for stopping feeding Death Tear in bowel mesentery Commenced oral diet
2 1 1
As shown in Table 2, six patients were not started on enteral feeding in the DCCM. Of these patients, four commenced an oral diet, one developed an ileus and one was transferred to another hospital. The patient who developed an ileus had a NG tube in situ for drainage and was transferred to another hospital after 2.8 days. Whether enteral feeding was commenced in this patient is unknown. Of these patients two died, one sustained a tear in the bowel mesentery and was started on TPN, and one was able to meet nutrient requirements orally. In the two patients that died the enteral feed was stopped at day 1 and day 5 of feeding and for the other two patients at day 1 and day 2, respectively. The median number of times for interrupting enteral feeding was twice with a range of 0—20 times (Table 1). As shown in Fig. 1, most patients had their feed interrupted fewer than four times. As shown in Fig. 2, the main reasons for interrupting the feed were high gastric aspirates, attempting an oral diet, and diagnostic or radiological procedures.
Figure 2
Figure 1 Number of times per patient enteral feeding interrupted.
In one patient, enteral feeding was interrupted because of ileus and failure to absorb the feed. In this patient, the NG tube was changed for an NJ tube 4 days after the onset of ileus. The feed was not interrupted again after the NJ tube was placed. Similarly, one other patient commenced NG feeding and was unable to absorb the feed. In this patient, an NJ tube was placed approximately 1.5 days after NG feeding was started, and there were no further complications. As shown in Fig. 3, in 10 patients (37%) enteral feeding was not interrupted for high gastric aspirates and 10 (37%) had their feed interrupted only one or two times. Only three patients had their feed interrupted greater than four times for high gastric aspirates. For patients admitted during the earlier years of the study, feeding was interrupted if gastric aspirates were greater than 100 ml. The current policy of the DCCM is to withhold the feed if gastric aspirates are greater than 200 ml. All patients who had their feed interrupted for high gastric aspirates
Reasons for interrupting the enteral feed.
Safety of early enteral feeding in spinal cord injury patients
Figure 3 Number of times per patient enteral feeding interrupted for high gastric aspirates.
were enterally fed by NG or NJ tube until they were transferred from DCCM, commenced an oral diet or died.
Discussion Studies looking at nutrition in spinal cord injured patients have mostly focused on the long-term consequences of injury rather than the acute stage. Chronic paraplegia and quadriplegia are marked by a reduction in energy expenditure of up to 67% and progressive loss of lean body mass.10 Few studies have reported on the acute metabolic response of patients with spinal cord transection. Metabolic changes that occur in the acute stage of spinal cord injury may be unique because of the muscle atrophy that results from paralysis. Kaufman et al. studied the nutritional status of eight patients who had suffered spinal cord transection during their first 10—14 days in hospital. Results demonstrated an ongoing negative nitrogen balance that persisted at least 10 days post-injury. Nutritional status deteriorated partly due to an inadequate supply of protein and caloric needs during the study.6 Laven et al. assessed the nutritional status of 51 spinal cord injured patients at 2, 4 and 8 weeks after injury. Nutrient abnormalities such as albumin, carotene, transferrin, ascorbate, thiamine, folate and copper were documented most frequently in the initial period following injury.8 Rodriguez et al. compared 10 patients with spinal cord injuries with 20 controls who had suffered multisystem trauma but no spinal cord injury. In both groups, total nutrition support including enteral feeding and TPN was delivered within 72 h of admission. No spinal cord injured patient established positive nitrogen balance during the 7-week period following injury despite an average delivery of 2.4 g of protein/kg ideal body weight. In contrast 17 of the 20 controls achieved the nitrogen balance within 3
241
weeks of admission. The authors concluded that negative nitrogen balance is obligatory in patients who have suffered spinal injuries.10 Protein loss and rapid depletion of energy stores are well-documented responses following stress and injury in other critically ill patients. Early enteral feeding is recommended to blunt these responses. Gastric dysfunction is believed to be more common in spinal cord injury and this may cause some reluctance to initiate early enteral feeding if the patient cannot take food orally.2 In 1970, Watkin described ‘spinal ileus’ as a ‘‘functional intestinal obstruction complicating a lesion in the spine’’. He reported eight cases where patients had developed an intestinal pseudo obstruction believed to be caused by their spinal cord injury.13 Since this time literature has been scant on this topic. Studies reporting on the acute metabolic response of spinal cord injury did not record the number of complications associated with enteral feeding. This study has attempted to look at the safety of administering enteral feeding in a group of ICU patients who have suffered complete spinal cord transection. In the study group 82% were commenced on enteral feeding. Most patients were started on NG feeds (93%) as the preferred method of enteral feeding. Feeding was commenced quickly in most patients. The median of 2 days after admission is at the fast end of the 2—6.2 range reported in the literature.3,5,11 Valid reasons were given for not starting enteral feeding, including commencing an oral diet, ileus and transfer to another hospital. Of the patients commenced on enteral feeding all but one remained on a feed until transfer from the DCCM, commencing an oral diet or death. The one patient who could not sustain enteral feeding had a clear contraindication. This patient was in a motor vehicle crash and sustained multiple injuries including a tear in the bowel mesentery. TPN was administered in this patient. High gastric aspirates were the most common reason for interrupting the feed, followed by diagnostic or radiological procedures and commencing an oral diet which are not complications of the enteral feeding protocol. In most patients the feed was interrupted for high gastric aspirates only twice or less and very few patients had four or more interruptions. Although this was a common reason for interrupting the feed it was not a contraindication for enteral feeding in any patient. Two patients had prolonged high gastric aspirates and the diagnosis of ileus was made. In these patients, the NG tube was placed on free drainage and a NJ tube was placed with no further complications. There is controversy over whether or not bowel sounds should be present before starting enteral
242
feeding. Studies have cited absence of bowel sounds and gut dysfunction as a common reason for not initiating early enteral feeding in trauma patients.1,4,7 Other investigators have challenged this belief. McDonald et al.9 evaluated the safety and efficacy of immediate enteral feeding in 106 patients who had suffered from burns greater than 20% of their body surface area. Enteral feeding was started immediately–—within 1 h of admission and within 6 h of injury. The incidence of complications related to enteral feeding was low. The most common complication observed was vomiting which occurred in 16 patients. Eighty-two percent of patients absorbed at least a portion of their enteral feeding on the day of injury.9 The Auckland Hospital DCCM has an enteral feeding protocol that outlines the steps to be taken if a patient develops gastric aspirates greater than 200 ml. The protocol enables the absorption of feed to be monitored and complications such as ileus detected so measures can be taken to prevent further complications. Such a protocol ensures safe administration of enteral feeding. By ensuring that nursing staff adhere to a protocol, enteral feeding can be safely administered in all trauma patients who require tube feeding. Those who have suffered spinal cord injuries are no exception. Results from this study show that in all the patients started on enteral feeding only one had complications that prevented continuation. This would indicate that enteral feeding can be safely administered in patients that have suffered spinal cord injuries. Research into the metabolic changes and nutritional support in the acute stage of spinal cord injury is in its infancy. Rodriquez et al. showed that protein loss may exceed that of patients suffering multi trauma without spinal injury.10 Whether gut dysfunction and ileus are more common in patients with spinal cord injuries is unknown. Historical research such as that reported by Watkin13 has provided a theory, which has not been backed up by good evidence. No major complications were seen in this study in patients who were enterally fed. However, this study was a small retrospective, descriptive analysis which lacks statistical power. Clearly, more research is needed in this area.
C.J. Rowan et al.
Conclusion No major complications associated with enteral feeding were seen in this study. This indicates that enteral feeding is a safe technique to use in the acute stage of spinal cord injury provided complications are monitored daily. Further research is needed to determine whether ileus and complications associated with enteral feeding are more common in patients who have suffered spinal cord injuries.
References 1. Adam S, Batson S. A study of problems associated with the delivery of enteral feed in critically ill patients in five ICUs in the UK. Int Care Med 1997;23:261—6. 2. Baxendale BR, Winter RJ. Critical care management of acute spinal injury. Crit Care Monit 1998;3(1):3—5. 3. Berger MM, Chiolero RL, Pannatier A, Cayeux MC, Tappy L. A 10-year survey of nutritional support in a surgical ICU: 1986—1995. Nutrition 1997;13:870—7. 4. Heyland DK, Cook DJ, Guyatt GH. Enteral nutrition in the critically ill patient: a critical review of the evidence. Int Care Med 1993;19:435—42. 5. Heyland D, Cook DJ, Winder B, Brylowski L, Van deMark H, Guyatt G. Enteral nutrition in the critically ill patient. A prospective survey. Crit Care Med 1995;23(6):1055—60. 6. Kaufman HH, Rowlands BJ, Stein DK, Kopaniky DR, Gildenberg PL. General metabolism in patients with acute paraplegia and quadriplegia. Neurosurgery 1985;16(3): 309—13. 7. Kemper M, Weissman C, Hyman AI. Caloric requirements and supply in critically ill surgical patients. Crit Care Med 1992;20(3):344—8. 8. Laven GT, Huang C, Deviro MJ, Stover ST, Kuhlemeier KV, Fine PR. Nutritional status during the acute stage of spinal cord injury. Arch Phys Med Rehabil 1989;70:277—82. 9. McDonald WS, Sharp CW, Deitch EA. Immediate enteral feeding in burn patients is safe and effective. Ann Surg 1991;213(2):177—83. 10. Rodriguez DJ, Clevenger FW, Osler TM, Demarest GB, Fry DE, Turner MO. Obligatory negative nitrogen balance following spinal cord injury. JPEN 1991;15(3):319—22. 11. Stechmiller J, Treloar DM, Derrico D, Yarandi H, Guin P. Interruption of enteral feedings in head injured patients. J Neurol Nurs 1994;26(4):224—9. 12. Walker R. People in the northern region: a demographic profile from the 1996 census. Auckland Health Funding Authority; 1998. 13. Watkin DF. Spinal Ileus. Brit J Surg 1970;57(2):142—7. 14. Yu D. A crash course in spinal cord injury. Medical care issues for primary care physicians. Postgrad Med 1998; 104(2):109—20.
Is early enteral feeding safe in patients who have suffered spinal cord injury? C.J. Rowana,*, L.K. Gillandersa, R.L. Paiceb, J.A. Judsonc a
Nutrition Services, Auckland Hospital, Auckland, New Zealand Trauma Services, Auckland Hospital, Auckland, New Zealand c Department of Critical Care Medicine, Auckland Hospital, Auckland, New Zealand b
Accepted 19 April 2003
KEYWORDS Enteral feeding; Spinal cord injury; Gastric aspirates
Summary Objective: To examine whether enteral feeding is a safe technique to use in the acute stage of spinal cord injury. Methods: We searched the departmental computerised patient database and clinical records for all patients with spinal cord injuries admitted to the Auckland Hospital Intensive Care Unit (ICU), known as the Department of Critical Care Medicine (DCCM), between January 1988 and December 2000. Patients were included in the study if they had suffered complete spinal cord transection resulting in either paraplegia or quadriplegia. Data was collected for the following variables: length of time to commence enteral feeding, type of enteral feeding, duration of enteral feeding and reasons for interrupting the feed. Results: Thirty-three patients were found and were included in the study. Twenty-seven (82%) of the patients commenced enteral feeding in the DCCM, 25 by nasogastric (NG) and 2 by nasojejunal (NJ) tube. Feeding was commenced a median of 2 days after admission and the median length of enteral feeding was 7.7 days. The main feeding complications that resulted in interrupting the feed were high gastric aspirates. One patient commenced on enteral feeding developed medical complications that prevented continuation. Two patients on NG feeding converted to NJ feeding. Conclusion: No major complications associated with enteral feeding were seen in this study. This would indicate that enteral feeding can be safely administered in the acute stage of spinal cord injury provided complications are monitored for daily. ß 2003 Elsevier Ltd. All rights reserved.
Introduction Spinal cord injury is a devastating event with farreaching functional and psychological effects on the survivor. Optimal medical and rehabilitative care is important for patients with such injuries to regain a sense of control over their life and maximise their independence and quality of life.14 Nutrition support is an important component of *Corresponding author. Tel.: þ64-9-638-6654; fax: þ64-9-263-8709. E-mail address: [email protected] (C.J. Rowan).
therapy especially in the acute stages following injury. For these patients early nutrition intervention is important to attenuate the protein loss that occurs in such injury. Many spinal cord injured patients have suffered other trauma that demands additional energy and protein requirements. For patients that require nutrition support, enteral feeding is preferred over total parental nutrition (TPN) because it maintains gastrointestinal function, results in fewer infectious and metabolic complications and is cheaper than TPN.3,4 Despite the importance of enteral nutrition, feeding is often withheld in spinal cord injured
0020–1383/$ — see front matter ß 2003 Elsevier Ltd. All rights reserved. doi:10.1016/S0020-1383(03)00203-1
Safety of early enteral feeding in spinal cord injury patients
patients for concerns about ileus and other complications.2 Some patients remain ‘nil by mouth’ with intravenous fluids until oral food intake is resumed after days or weeks. The purpose of this study was to examine whether enteral feeding is a safe technique to use in the acute stage of spinal cord injury.
Population and methods Auckland is a city of approximately 1.08 million people, divided into three health areas (Northwest Auckland, Central Auckland and South Auckland). Two trauma centres with well defined catchment areas serve these areas. Auckland Hospital, a 538 bed university affiliated tertiary referral hospital, covers Central Auckland and Northwest Auckland areas, and Middlemore Hospital covers the South Auckland area. At the 1996 census, a total of 739,926 people resided permanently within the Auckland Hospital trauma catchment area.12 The Auckland Hospital Intensive Care Unit (ICU), known as the Department of Critical Care Medicine (DCCM) was the only ICU providing care for trauma patients aged 15 years and over for the entire period of this study. Its computerised patient registry has included trauma variables since 1988. We searched the departmental computerised patient database and clinical records for all patients with spinal cord injuries admitted to the DCCM, in the 13 years from 1 January 1988 to 31 December 2000. All patients who had suffered clinical spinal cord transection resulting in either paraplegia or quadriplegia were included in the study. Date of birth, sex, date admitted to the DCCM, date transferred from the DCCM and level of injury were recorded for each patient from the clinical record. Feeding practices performed on Table 1
239
each patient during their DCCM stay were obtained from 24 h record sheets. These contain a record of all clinical interventions performed over a 24 h period and include the feeding regimen. Data was collected manually on a spreadsheet for the following variables: time to commence enteral feeding, type of enteral feeding started, either nasogastric (NG) or nasojejunal (NJ) and reasons for interrupting the feed. All patients were gravity fed and received a standard, polymeric feed with a nutrient density of 1 to 1.2 kcal/ml. Feeding was started at 3 h intervals and progressed to a continuous 24 h regimen. A retrospective, descriptive analysis was then performed. A literature review was conducted using Medline. The key words ‘spinal injuries’ and ‘enteral nutrition’ were used to source all relevant literature from 1 January 1990 to 31 December 2000.
Results Thirty-three patients with paraplegia or quadraplegia were identified. As shown in Table 1, they consisted primarily of men with a median age of 27 years, reflecting the high incidence of spinal cord injury in a young male population. Twenty-seven (82%) of the 33 patients were commenced on enteral feeding. Enteral feeding was commenced a median of 2 days after admission to the DCCM and patients were fed for a median of 7.7 days during their DCCM stay. Twenty-five patients were commenced on NG feeding and two of them were later converted to NJ feeding. Two patients were commenced on NJ feeding from the outset because of high gastric aspirates. Twenty-three out of 27 patients were transferred from DCCM on enteral feeding.
Patient demographics
Number of patients Age (median years (range)) Sex (female/male) Length of DCCM stay (median days (range)) Patients with paraplegia Patients with quadriplegia
33 27 (15—47) 2/31 8.8 (2.8—36.2) 20 13
Number of patients commenced enteral feeding Nasogastric (NG) feeding Nasojejunal (NJ) feeding from the outset
27 (82%) 25a 2
Time to commence enteral feeding (median days (range)) Number of times enteral feeding interrupted per patient (median days (range)) Duration of enteral feeding in DCCM (median days (range)) Number of patients transferred from DCCM on enteral feeding
2 (0.5—4.8) 2 (0—20) 7.7 (0.2—34.6) 23
a
Two of these changed to nasojejunal (NJ) feeding.
240
C.J. Rowan et al.
Table 2 Reasons for not commencing and stopping enteral feeding Number of patients Reason for not commencing feeding Commenced oral diet 4 Ileus 1 Patient transferred to another hospital 1 Reason for stopping feeding Death Tear in bowel mesentery Commenced oral diet
2 1 1
As shown in Table 2, six patients were not started on enteral feeding in the DCCM. Of these patients, four commenced an oral diet, one developed an ileus and one was transferred to another hospital. The patient who developed an ileus had a NG tube in situ for drainage and was transferred to another hospital after 2.8 days. Whether enteral feeding was commenced in this patient is unknown. Of these patients two died, one sustained a tear in the bowel mesentery and was started on TPN, and one was able to meet nutrient requirements orally. In the two patients that died the enteral feed was stopped at day 1 and day 5 of feeding and for the other two patients at day 1 and day 2, respectively. The median number of times for interrupting enteral feeding was twice with a range of 0—20 times (Table 1). As shown in Fig. 1, most patients had their feed interrupted fewer than four times. As shown in Fig. 2, the main reasons for interrupting the feed were high gastric aspirates, attempting an oral diet, and diagnostic or radiological procedures.
Figure 2
Figure 1 Number of times per patient enteral feeding interrupted.
In one patient, enteral feeding was interrupted because of ileus and failure to absorb the feed. In this patient, the NG tube was changed for an NJ tube 4 days after the onset of ileus. The feed was not interrupted again after the NJ tube was placed. Similarly, one other patient commenced NG feeding and was unable to absorb the feed. In this patient, an NJ tube was placed approximately 1.5 days after NG feeding was started, and there were no further complications. As shown in Fig. 3, in 10 patients (37%) enteral feeding was not interrupted for high gastric aspirates and 10 (37%) had their feed interrupted only one or two times. Only three patients had their feed interrupted greater than four times for high gastric aspirates. For patients admitted during the earlier years of the study, feeding was interrupted if gastric aspirates were greater than 100 ml. The current policy of the DCCM is to withhold the feed if gastric aspirates are greater than 200 ml. All patients who had their feed interrupted for high gastric aspirates
Reasons for interrupting the enteral feed.
Safety of early enteral feeding in spinal cord injury patients
Figure 3 Number of times per patient enteral feeding interrupted for high gastric aspirates.
were enterally fed by NG or NJ tube until they were transferred from DCCM, commenced an oral diet or died.
Discussion Studies looking at nutrition in spinal cord injured patients have mostly focused on the long-term consequences of injury rather than the acute stage. Chronic paraplegia and quadriplegia are marked by a reduction in energy expenditure of up to 67% and progressive loss of lean body mass.10 Few studies have reported on the acute metabolic response of patients with spinal cord transection. Metabolic changes that occur in the acute stage of spinal cord injury may be unique because of the muscle atrophy that results from paralysis. Kaufman et al. studied the nutritional status of eight patients who had suffered spinal cord transection during their first 10—14 days in hospital. Results demonstrated an ongoing negative nitrogen balance that persisted at least 10 days post-injury. Nutritional status deteriorated partly due to an inadequate supply of protein and caloric needs during the study.6 Laven et al. assessed the nutritional status of 51 spinal cord injured patients at 2, 4 and 8 weeks after injury. Nutrient abnormalities such as albumin, carotene, transferrin, ascorbate, thiamine, folate and copper were documented most frequently in the initial period following injury.8 Rodriguez et al. compared 10 patients with spinal cord injuries with 20 controls who had suffered multisystem trauma but no spinal cord injury. In both groups, total nutrition support including enteral feeding and TPN was delivered within 72 h of admission. No spinal cord injured patient established positive nitrogen balance during the 7-week period following injury despite an average delivery of 2.4 g of protein/kg ideal body weight. In contrast 17 of the 20 controls achieved the nitrogen balance within 3
241
weeks of admission. The authors concluded that negative nitrogen balance is obligatory in patients who have suffered spinal injuries.10 Protein loss and rapid depletion of energy stores are well-documented responses following stress and injury in other critically ill patients. Early enteral feeding is recommended to blunt these responses. Gastric dysfunction is believed to be more common in spinal cord injury and this may cause some reluctance to initiate early enteral feeding if the patient cannot take food orally.2 In 1970, Watkin described ‘spinal ileus’ as a ‘‘functional intestinal obstruction complicating a lesion in the spine’’. He reported eight cases where patients had developed an intestinal pseudo obstruction believed to be caused by their spinal cord injury.13 Since this time literature has been scant on this topic. Studies reporting on the acute metabolic response of spinal cord injury did not record the number of complications associated with enteral feeding. This study has attempted to look at the safety of administering enteral feeding in a group of ICU patients who have suffered complete spinal cord transection. In the study group 82% were commenced on enteral feeding. Most patients were started on NG feeds (93%) as the preferred method of enteral feeding. Feeding was commenced quickly in most patients. The median of 2 days after admission is at the fast end of the 2—6.2 range reported in the literature.3,5,11 Valid reasons were given for not starting enteral feeding, including commencing an oral diet, ileus and transfer to another hospital. Of the patients commenced on enteral feeding all but one remained on a feed until transfer from the DCCM, commencing an oral diet or death. The one patient who could not sustain enteral feeding had a clear contraindication. This patient was in a motor vehicle crash and sustained multiple injuries including a tear in the bowel mesentery. TPN was administered in this patient. High gastric aspirates were the most common reason for interrupting the feed, followed by diagnostic or radiological procedures and commencing an oral diet which are not complications of the enteral feeding protocol. In most patients the feed was interrupted for high gastric aspirates only twice or less and very few patients had four or more interruptions. Although this was a common reason for interrupting the feed it was not a contraindication for enteral feeding in any patient. Two patients had prolonged high gastric aspirates and the diagnosis of ileus was made. In these patients, the NG tube was placed on free drainage and a NJ tube was placed with no further complications. There is controversy over whether or not bowel sounds should be present before starting enteral
242
feeding. Studies have cited absence of bowel sounds and gut dysfunction as a common reason for not initiating early enteral feeding in trauma patients.1,4,7 Other investigators have challenged this belief. McDonald et al.9 evaluated the safety and efficacy of immediate enteral feeding in 106 patients who had suffered from burns greater than 20% of their body surface area. Enteral feeding was started immediately–—within 1 h of admission and within 6 h of injury. The incidence of complications related to enteral feeding was low. The most common complication observed was vomiting which occurred in 16 patients. Eighty-two percent of patients absorbed at least a portion of their enteral feeding on the day of injury.9 The Auckland Hospital DCCM has an enteral feeding protocol that outlines the steps to be taken if a patient develops gastric aspirates greater than 200 ml. The protocol enables the absorption of feed to be monitored and complications such as ileus detected so measures can be taken to prevent further complications. Such a protocol ensures safe administration of enteral feeding. By ensuring that nursing staff adhere to a protocol, enteral feeding can be safely administered in all trauma patients who require tube feeding. Those who have suffered spinal cord injuries are no exception. Results from this study show that in all the patients started on enteral feeding only one had complications that prevented continuation. This would indicate that enteral feeding can be safely administered in patients that have suffered spinal cord injuries. Research into the metabolic changes and nutritional support in the acute stage of spinal cord injury is in its infancy. Rodriquez et al. showed that protein loss may exceed that of patients suffering multi trauma without spinal injury.10 Whether gut dysfunction and ileus are more common in patients with spinal cord injuries is unknown. Historical research such as that reported by Watkin13 has provided a theory, which has not been backed up by good evidence. No major complications were seen in this study in patients who were enterally fed. However, this study was a small retrospective, descriptive analysis which lacks statistical power. Clearly, more research is needed in this area.
C.J. Rowan et al.
Conclusion No major complications associated with enteral feeding were seen in this study. This indicates that enteral feeding is a safe technique to use in the acute stage of spinal cord injury provided complications are monitored daily. Further research is needed to determine whether ileus and complications associated with enteral feeding are more common in patients who have suffered spinal cord injuries.
References 1. Adam S, Batson S. A study of problems associated with the delivery of enteral feed in critically ill patients in five ICUs in the UK. Int Care Med 1997;23:261—6. 2. Baxendale BR, Winter RJ. Critical care management of acute spinal injury. Crit Care Monit 1998;3(1):3—5. 3. Berger MM, Chiolero RL, Pannatier A, Cayeux MC, Tappy L. A 10-year survey of nutritional support in a surgical ICU: 1986—1995. Nutrition 1997;13:870—7. 4. Heyland DK, Cook DJ, Guyatt GH. Enteral nutrition in the critically ill patient: a critical review of the evidence. Int Care Med 1993;19:435—42. 5. Heyland D, Cook DJ, Winder B, Brylowski L, Van deMark H, Guyatt G. Enteral nutrition in the critically ill patient. A prospective survey. Crit Care Med 1995;23(6):1055—60. 6. Kaufman HH, Rowlands BJ, Stein DK, Kopaniky DR, Gildenberg PL. General metabolism in patients with acute paraplegia and quadriplegia. Neurosurgery 1985;16(3): 309—13. 7. Kemper M, Weissman C, Hyman AI. Caloric requirements and supply in critically ill surgical patients. Crit Care Med 1992;20(3):344—8. 8. Laven GT, Huang C, Deviro MJ, Stover ST, Kuhlemeier KV, Fine PR. Nutritional status during the acute stage of spinal cord injury. Arch Phys Med Rehabil 1989;70:277—82. 9. McDonald WS, Sharp CW, Deitch EA. Immediate enteral feeding in burn patients is safe and effective. Ann Surg 1991;213(2):177—83. 10. Rodriguez DJ, Clevenger FW, Osler TM, Demarest GB, Fry DE, Turner MO. Obligatory negative nitrogen balance following spinal cord injury. JPEN 1991;15(3):319—22. 11. Stechmiller J, Treloar DM, Derrico D, Yarandi H, Guin P. Interruption of enteral feedings in head injured patients. J Neurol Nurs 1994;26(4):224—9. 12. Walker R. People in the northern region: a demographic profile from the 1996 census. Auckland Health Funding Authority; 1998. 13. Watkin DF. Spinal Ileus. Brit J Surg 1970;57(2):142—7. 14. Yu D. A crash course in spinal cord injury. Medical care issues for primary care physicians. Postgrad Med 1998; 104(2):109—20.