Jane R. Cowan, MD, MPH

 

Clinical scenario

An 83-year-old man is admitted with a hemiplegic cerebrovascular accident. He is found to have dysphagia, and a nasogastric feeding tube is placed. Over the next several days, his strength begins to recover, and he tolerates his tube feeding well. Discharge to a skilled nursing facility (SNF) for subacute rehabilitation is planned. His swallowing is showing signs of recovery; it has not recovered adequately but is expected to continue to improve such that he is predicted to be independent of tube feeding within 7-14 days. None of the facilities in the region are willing to admit a patient with a nasal feeding tube, despite the anticipated short duration. The patient is medically ready for discharge but is refusing the feeding gastrostomy. “Why would I want a hole in my stomach, if I’m only going to need it for 1-2 weeks and this tube in my nose is working fine and is comfortable?” he pleads with tears in his eyes.

Over the next several days he and his family are subject to numerous pressured conversations about tube placement, with well-meaning house staff explaining that his recovery from the stroke is dependent on transfer to the SNF and – erroneously – that nasal tubes are inappropriate for outpatient use. He extremely reluctantly assents to the gastrostomy, is discharged to the SNF, and is eating within 2 weeks. Subsequently the gastrostomy was removed at an outpatient appointment, and the gastrocutaneous fistula required wound care until it closed.

Feeding dysphagic patients after stroke

Dysphagia, potentially leading to aspiration and/or pneumonia, is a common sequela of stroke – up to half of hospitalized patients are affected.¹ When oral intake is contraindicated, patients are often fed by nasogastric tube (NGT) or by surgically or endoscopically placed gastrostomy tube (GT). Without good justification based on outcomes, NGTs are traditionally used when the need for feeding is thought to be short term (<4 weeks) and GTs are used for long term (>4 weeks). However, in 2005, a large multicenter randomized control trial found that the majority of stroke patients with dysphagia that would resolve had resolution within 2-3 weeks. Moreover, outcomes were equivalent or better for patients fed with an NGT versus GT.

The authors concluded by recommending feeding via NGT for 2-3 weeks, after which conversion to GT can be considered if dysphagia persists.¹ Notably, the recommendation allows consideration, and no evidence-based guideline requires or recommends GT be placed based on duration of tube feed dependence. Currently, while nutrition and neurology authorities have adopted these recommendations,^2,3 many authors have noted poor adherence to this guideline, and many find that the median period between stroke and GT placement is 7 days rather than the recommended minimum of 14.^4,5,6 While ignorance can partially explain the lack of widespread compliance,⁶ the policies of posthospital facilities are another culprit. Increasingly, and for a variety of reasons unsupported by the literature, SNFs refuse NGT and require GT.^4,7,8,9

 

Ethical considerations

The four principles of medical ethics – autonomy, beneficence, nonmaleficence, and justice – can guide clinicians, patients, and family members in decision-making. In our case, by withholding needed and desired treatment (discharge to and treatment by a rehabilitation facility) the patient is being coerced to undergo a procedure he does not want, and clinicians participate in denying him autonomy. Further, given that the evidence, national guidelines, and in fact federal regulations indicate that his preferences are congruent with best practices, pressuring him to accept gastrostomy placement runs afoul of the principles of beneficence and nonmaleficence. Though the mechanism is unclear, early gastrostomy (<14-21 days) is associated with increased risk of death, worse functional outcomes, and a lower rate of return to oral feeding, as well as a significant procedure-specific complication rate.^1,10 By insisting on gastrostomy, we neither act in this patient’s best interests nor “do no harm.”

However, the medical system is complex. The clinician at the bedside can evaluate this scenario, review the national guidelines, discuss the procedure and risks with the patient and family, and conclude that the patient should be discharged with a nasal feeding tube. Nevertheless, if no facility is willing to accept him without a gastrostomy, our decision-making model – previously limited to our patient’s best interests alone – is forced to change. Despite our misgivings, we often conclude that the harm done by an early gastrostomy is outweighed by the harm of remaining unnecessarily in the acute hospital setting. We further worry about other patients lingering in the emergency department for lack of an inpatient bed and the possible – though unknowable – harm done to them.
 

Looking forward

It is an unfortunate fact that medical decision-making must often include factors unrelated to the patient’s best interests, with financial considerations and structural barriers frequently driving deviation from ideal care. Providers and patients navigate these decisions to their best abilities, making compromises when forced. However, with education and professional activism, providers can advocate for the elimination of barriers to providing medically sound and ethically appropriate care. In our experience, delay of gastrostomy placement, until discharge is imminent and planning for postdischarge care is initiated, has resulted in a decrease by half the fraction of patients with tracheostomies who had gastrostomies placed prior to discharge.¹¹ With aggressive outreach and education, we now have nursing homes willing to accept patients with NGTs.

Criteria for admission to discharge facilities can drive medical decision-making that is unethical and unsupported by evidence. Continued efforts to eliminate barriers to appropriate and ethical care have been successful and are encouraged.
 

Dr. Cowan is administrative chief resident in the department of surgery at Columbia University Irving Medical Center, New York. Dr. Seres is professor of medicine in the Institute of Human Nutrition and associate clinical ethicist at Columbia University Irving Medical Center. The authors have no conflicts of interest to disclose.

References

1. Dennis MS et al. Lancet. 2005 Feb 26-Mar 4;365(9461):764-72.

2. Powers W. et al. Stroke. 2018 Mar;49(3):e46-e110.

3. Burgos R et al. Clin Nutr. 2018 Feb;37(1):354-96.

4. Wilmskoetter J et al. J Stroke Cerebrovasc Dis. 2016 Nov;25(11):2694-700.

5. George BP et al. Stroke. 2017 Feb;48(2):420-7.

6. Fessler TA. et al. Surg Endosc. 2019 Dec;33(12):4089-97.

7. Burgermaster M et al. Nutr Clin Pract. 2016 Jun;31(3):342-8.

8. Moran C and O’Mahoney S. Curr Opin Gastroenterol. 2015 Mar;31(2):137-42.

9. Gomes CA et al. Cochrane Database Syst Rev. 2010 Nov 10;(11):CD008096.

10. Joundi RA et al. Neurology. 2018 Feb 13;90(7):e544-52.

11. Bothra A et al. J Parenter Enteral Nutr. 2018 Feb;42(2):491.

 

Clinical scenario

An 83-year-old man is admitted with a hemiplegic cerebrovascular accident. He is found to have dysphagia, and a nasogastric feeding tube is placed. Over the next several days, his strength begins to recover, and he tolerates his tube feeding well. Discharge to a skilled nursing facility (SNF) for subacute rehabilitation is planned. His swallowing is showing signs of recovery; it has not recovered adequately but is expected to continue to improve such that he is predicted to be independent of tube feeding within 7-14 days. None of the facilities in the region are willing to admit a patient with a nasal feeding tube, despite the anticipated short duration. The patient is medically ready for discharge but is refusing the feeding gastrostomy. “Why would I want a hole in my stomach, if I’m only going to need it for 1-2 weeks and this tube in my nose is working fine and is comfortable?” he pleads with tears in his eyes.

Over the next several days he and his family are subject to numerous pressured conversations about tube placement, with well-meaning house staff explaining that his recovery from the stroke is dependent on transfer to the SNF and – erroneously – that nasal tubes are inappropriate for outpatient use. He extremely reluctantly assents to the gastrostomy, is discharged to the SNF, and is eating within 2 weeks. Subsequently the gastrostomy was removed at an outpatient appointment, and the gastrocutaneous fistula required wound care until it closed.

Feeding dysphagic patients after stroke

Dysphagia, potentially leading to aspiration and/or pneumonia, is a common sequela of stroke – up to half of hospitalized patients are affected.¹ When oral intake is contraindicated, patients are often fed by nasogastric tube (NGT) or by surgically or endoscopically placed gastrostomy tube (GT). Without good justification based on outcomes, NGTs are traditionally used when the need for feeding is thought to be short term (<4 weeks) and GTs are used for long term (>4 weeks). However, in 2005, a large multicenter randomized control trial found that the majority of stroke patients with dysphagia that would resolve had resolution within 2-3 weeks. Moreover, outcomes were equivalent or better for patients fed with an NGT versus GT.

The authors concluded by recommending feeding via NGT for 2-3 weeks, after which conversion to GT can be considered if dysphagia persists.¹ Notably, the recommendation allows consideration, and no evidence-based guideline requires or recommends GT be placed based on duration of tube feed dependence. Currently, while nutrition and neurology authorities have adopted these recommendations,^2,3 many authors have noted poor adherence to this guideline, and many find that the median period between stroke and GT placement is 7 days rather than the recommended minimum of 14.^4,5,6 While ignorance can partially explain the lack of widespread compliance,⁶ the policies of posthospital facilities are another culprit. Increasingly, and for a variety of reasons unsupported by the literature, SNFs refuse NGT and require GT.^4,7,8,9

 

Ethical considerations

The four principles of medical ethics – autonomy, beneficence, nonmaleficence, and justice – can guide clinicians, patients, and family members in decision-making. In our case, by withholding needed and desired treatment (discharge to and treatment by a rehabilitation facility) the patient is being coerced to undergo a procedure he does not want, and clinicians participate in denying him autonomy. Further, given that the evidence, national guidelines, and in fact federal regulations indicate that his preferences are congruent with best practices, pressuring him to accept gastrostomy placement runs afoul of the principles of beneficence and nonmaleficence. Though the mechanism is unclear, early gastrostomy (<14-21 days) is associated with increased risk of death, worse functional outcomes, and a lower rate of return to oral feeding, as well as a significant procedure-specific complication rate.^1,10 By insisting on gastrostomy, we neither act in this patient’s best interests nor “do no harm.”

However, the medical system is complex. The clinician at the bedside can evaluate this scenario, review the national guidelines, discuss the procedure and risks with the patient and family, and conclude that the patient should be discharged with a nasal feeding tube. Nevertheless, if no facility is willing to accept him without a gastrostomy, our decision-making model – previously limited to our patient’s best interests alone – is forced to change. Despite our misgivings, we often conclude that the harm done by an early gastrostomy is outweighed by the harm of remaining unnecessarily in the acute hospital setting. We further worry about other patients lingering in the emergency department for lack of an inpatient bed and the possible – though unknowable – harm done to them.
 

Looking forward

It is an unfortunate fact that medical decision-making must often include factors unrelated to the patient’s best interests, with financial considerations and structural barriers frequently driving deviation from ideal care. Providers and patients navigate these decisions to their best abilities, making compromises when forced. However, with education and professional activism, providers can advocate for the elimination of barriers to providing medically sound and ethically appropriate care. In our experience, delay of gastrostomy placement, until discharge is imminent and planning for postdischarge care is initiated, has resulted in a decrease by half the fraction of patients with tracheostomies who had gastrostomies placed prior to discharge.¹¹ With aggressive outreach and education, we now have nursing homes willing to accept patients with NGTs.

Criteria for admission to discharge facilities can drive medical decision-making that is unethical and unsupported by evidence. Continued efforts to eliminate barriers to appropriate and ethical care have been successful and are encouraged.
 

Dr. Cowan is administrative chief resident in the department of surgery at Columbia University Irving Medical Center, New York. Dr. Seres is professor of medicine in the Institute of Human Nutrition and associate clinical ethicist at Columbia University Irving Medical Center. The authors have no conflicts of interest to disclose.

References

1. Dennis MS et al. Lancet. 2005 Feb 26-Mar 4;365(9461):764-72.

2. Powers W. et al. Stroke. 2018 Mar;49(3):e46-e110.

3. Burgos R et al. Clin Nutr. 2018 Feb;37(1):354-96.

4. Wilmskoetter J et al. J Stroke Cerebrovasc Dis. 2016 Nov;25(11):2694-700.

5. George BP et al. Stroke. 2017 Feb;48(2):420-7.

6. Fessler TA. et al. Surg Endosc. 2019 Dec;33(12):4089-97.

7. Burgermaster M et al. Nutr Clin Pract. 2016 Jun;31(3):342-8.

8. Moran C and O’Mahoney S. Curr Opin Gastroenterol. 2015 Mar;31(2):137-42.

9. Gomes CA et al. Cochrane Database Syst Rev. 2010 Nov 10;(11):CD008096.

10. Joundi RA et al. Neurology. 2018 Feb 13;90(7):e544-52.

11. Bothra A et al. J Parenter Enteral Nutr. 2018 Feb;42(2):491.

 

Clinical scenario

An 83-year-old man is admitted with a hemiplegic cerebrovascular accident. He is found to have dysphagia, and a nasogastric feeding tube is placed. Over the next several days, his strength begins to recover, and he tolerates his tube feeding well. Discharge to a skilled nursing facility (SNF) for subacute rehabilitation is planned. His swallowing is showing signs of recovery; it has not recovered adequately but is expected to continue to improve such that he is predicted to be independent of tube feeding within 7-14 days. None of the facilities in the region are willing to admit a patient with a nasal feeding tube, despite the anticipated short duration. The patient is medically ready for discharge but is refusing the feeding gastrostomy. “Why would I want a hole in my stomach, if I’m only going to need it for 1-2 weeks and this tube in my nose is working fine and is comfortable?” he pleads with tears in his eyes.

Over the next several days he and his family are subject to numerous pressured conversations about tube placement, with well-meaning house staff explaining that his recovery from the stroke is dependent on transfer to the SNF and – erroneously – that nasal tubes are inappropriate for outpatient use. He extremely reluctantly assents to the gastrostomy, is discharged to the SNF, and is eating within 2 weeks. Subsequently the gastrostomy was removed at an outpatient appointment, and the gastrocutaneous fistula required wound care until it closed.

Feeding dysphagic patients after stroke

Dysphagia, potentially leading to aspiration and/or pneumonia, is a common sequela of stroke – up to half of hospitalized patients are affected.¹ When oral intake is contraindicated, patients are often fed by nasogastric tube (NGT) or by surgically or endoscopically placed gastrostomy tube (GT). Without good justification based on outcomes, NGTs are traditionally used when the need for feeding is thought to be short term (<4 weeks) and GTs are used for long term (>4 weeks). However, in 2005, a large multicenter randomized control trial found that the majority of stroke patients with dysphagia that would resolve had resolution within 2-3 weeks. Moreover, outcomes were equivalent or better for patients fed with an NGT versus GT.

The authors concluded by recommending feeding via NGT for 2-3 weeks, after which conversion to GT can be considered if dysphagia persists.¹ Notably, the recommendation allows consideration, and no evidence-based guideline requires or recommends GT be placed based on duration of tube feed dependence. Currently, while nutrition and neurology authorities have adopted these recommendations,^2,3 many authors have noted poor adherence to this guideline, and many find that the median period between stroke and GT placement is 7 days rather than the recommended minimum of 14.^4,5,6 While ignorance can partially explain the lack of widespread compliance,⁶ the policies of posthospital facilities are another culprit. Increasingly, and for a variety of reasons unsupported by the literature, SNFs refuse NGT and require GT.^4,7,8,9

 

Ethical considerations

The four principles of medical ethics – autonomy, beneficence, nonmaleficence, and justice – can guide clinicians, patients, and family members in decision-making. In our case, by withholding needed and desired treatment (discharge to and treatment by a rehabilitation facility) the patient is being coerced to undergo a procedure he does not want, and clinicians participate in denying him autonomy. Further, given that the evidence, national guidelines, and in fact federal regulations indicate that his preferences are congruent with best practices, pressuring him to accept gastrostomy placement runs afoul of the principles of beneficence and nonmaleficence. Though the mechanism is unclear, early gastrostomy (<14-21 days) is associated with increased risk of death, worse functional outcomes, and a lower rate of return to oral feeding, as well as a significant procedure-specific complication rate.^1,10 By insisting on gastrostomy, we neither act in this patient’s best interests nor “do no harm.”

However, the medical system is complex. The clinician at the bedside can evaluate this scenario, review the national guidelines, discuss the procedure and risks with the patient and family, and conclude that the patient should be discharged with a nasal feeding tube. Nevertheless, if no facility is willing to accept him without a gastrostomy, our decision-making model – previously limited to our patient’s best interests alone – is forced to change. Despite our misgivings, we often conclude that the harm done by an early gastrostomy is outweighed by the harm of remaining unnecessarily in the acute hospital setting. We further worry about other patients lingering in the emergency department for lack of an inpatient bed and the possible – though unknowable – harm done to them.
 

Looking forward

It is an unfortunate fact that medical decision-making must often include factors unrelated to the patient’s best interests, with financial considerations and structural barriers frequently driving deviation from ideal care. Providers and patients navigate these decisions to their best abilities, making compromises when forced. However, with education and professional activism, providers can advocate for the elimination of barriers to providing medically sound and ethically appropriate care. In our experience, delay of gastrostomy placement, until discharge is imminent and planning for postdischarge care is initiated, has resulted in a decrease by half the fraction of patients with tracheostomies who had gastrostomies placed prior to discharge.¹¹ With aggressive outreach and education, we now have nursing homes willing to accept patients with NGTs.

Criteria for admission to discharge facilities can drive medical decision-making that is unethical and unsupported by evidence. Continued efforts to eliminate barriers to appropriate and ethical care have been successful and are encouraged.
 

Dr. Cowan is administrative chief resident in the department of surgery at Columbia University Irving Medical Center, New York. Dr. Seres is professor of medicine in the Institute of Human Nutrition and associate clinical ethicist at Columbia University Irving Medical Center. The authors have no conflicts of interest to disclose.

References

1. Dennis MS et al. Lancet. 2005 Feb 26-Mar 4;365(9461):764-72.

2. Powers W. et al. Stroke. 2018 Mar;49(3):e46-e110.

3. Burgos R et al. Clin Nutr. 2018 Feb;37(1):354-96.

4. Wilmskoetter J et al. J Stroke Cerebrovasc Dis. 2016 Nov;25(11):2694-700.

5. George BP et al. Stroke. 2017 Feb;48(2):420-7.

6. Fessler TA. et al. Surg Endosc. 2019 Dec;33(12):4089-97.

7. Burgermaster M et al. Nutr Clin Pract. 2016 Jun;31(3):342-8.

8. Moran C and O’Mahoney S. Curr Opin Gastroenterol. 2015 Mar;31(2):137-42.

9. Gomes CA et al. Cochrane Database Syst Rev. 2010 Nov 10;(11):CD008096.

10. Joundi RA et al. Neurology. 2018 Feb 13;90(7):e544-52.

11. Bothra A et al. J Parenter Enteral Nutr. 2018 Feb;42(2):491.