To the Editor:
Subacute rehabilitation (SAR), which is generally provided in skilled nursing facilities, involves less rigorous rehabilitation than inpatient (“acute”) programs,
1
and does not routinely incorporate specialized pain management or palliative care. If a patient's prognosis is poor, especially if there are significant symptoms such as debilitating pain or exhaustion that make participation in physical therapy difficult, a hospice program may be more clinically appropriate. Nonetheless, hospitalized patients and/or their families may sometimes decline a recommendation for a hospice discharge plan for a multitude of reasons, including a hope for further productive anticancer treatment.Much complexity lies in the fact that oncology patients are often not fully aware of the intent of their treatment and their overall prognosis.
2
, 3
Reasons for this may include varying degrees of failure on the part of clinicians to accurately explain treatment intent and prognosis.4
In one survey, 39% of rehabilitation physicians, compared with 61% of oncologists, believed that patients with advanced cancer actively participating in rehabilitation services genuinely understood their overall prognosis.5
Although it is possible that patients with advanced cancer may benefit from SAR, it is also possible that either no desired benefit is conferred, or that actual harm or detriment is incurred.We performed a retrospective review of a consecutive series of patients with advanced metastatic gastrointestinal (GI) cancer who were discharged from our inpatient service to an SAR and analyzed their median survival and whether these patients subsequently received systemic chemotherapy after hospital discharge.
Methods
An institutional electronic database was queried to identify patients of the GI Medical Oncology Service at Memorial Sloan Kettering Cancer Center (MSK) who were admitted between September 1, 2008 and December 31, 2014, and were discharged to SAR.
Excluded were patients with: non-metastatic cancer (21 patients), low-grade neuroendocrine tumors (five patients), those receiving or awaiting first-line chemotherapy (32 patients), or those lost to follow-up (five patients). Also excluded were patients with a history of fully resected metastatic cancer in the past but off treatment for several years and under observation (three patients), as well as patients admitted because of a reversible toxicity of an otherwise effective chemotherapy that was planned to be resumed after recovery (five patients). Reversible treatment-related toxicities included diarrhea (two patients), nausea/vomiting (one patient), abscess (one patient), and myelitis (one patient).
Primary outcomes included the median overall survival from the time of hospital discharge and the percentage of patients who received further chemotherapy after hospital discharge. Secondary outcomes included the percentages of patients who died at SAR, those who died within three months and six months of hospital discharge, as well as the percentages of patients who were transitioned to hospice within three months of hospital discharge, who were re-admitted within one month of hospital discharge, and who followed up with their oncologists for an outpatient appointment after hospital discharge.
Two authors (A. V. D. and L. E. C.) performed independent in-depth chart reviews of the electronic medical records of all included patients. All study authors reviewed the final data set. Descriptive statistics were generated for all included patients. The institutional review board of MSK approved this study.
Results
Patient Characteristics
Twenty-two patients met eligibility criteria. These patients had progressive metastatic GI malignancies despite receiving at least two lines of chemotherapy (18 patients) or being ineligible for chemotherapy because of performance status (four patients). Twelve patients (55%) were female. The median age was 74 years (range 45–90 years). Malignancies included pancreatic adenocarcinoma (seven patients), esophageal adenocarcinoma (four patients), colorectal adenocarcinoma (six patients), gastric adenocarcinoma (two patients), cholangiocarcinoma (two patients), and cancer of unknown primary (one patient). The most common metastatic sites included liver, lungs, peritoneum, lymph nodes, and malignant pleural effusions. The most common admitting diagnoses included failure to thrive, sepsis, acute kidney injury, and deep venous thrombosis. A Do-Not-Resuscitate order was present in 10 patients (45%). Table 1 summarizes these patients' characteristics.
Table 1Patient Characteristics
| Patient Number | Gender | Age, yrs | Cancer Type | Number of Lines of Prior Cancer Therapies | Metastatic Sites at Time of Admission | Primary Admitting Diagnosis | Code Status |
|---|---|---|---|---|---|---|---|
| 1 | Female | 77 | Gallbladder adenocarcinoma | 2 | Liver, lungs, peritoneum, lymph nodes | Failure to thrive | FULL |
| 2 | Female | 79 | Pancreatic adenocarcinoma | 2 | Liver | Right arm cellulitis | DNR |
| 3 | Male | 62 | Esophageal adenocarcinoma | 4 | Lymph nodes | Acute kidney injury | DNR |
| 4 | Female | 59 | Rectal adenocarcinoma | 3 | Liver, lung, adrenal gland | Peri-rectal abscess | FULL |
| 5 | Male | 45 | Esophageal adenocarcinoma | 2 | Brain | Lung abscess | FULL |
| 6 | Male | 68 | Pancreatic adenocarcinoma | 2 | Liver, peritoneum | Failure to thrive | DNR |
| 7 | Male | 56 | Esophageal adenocarcinoma | 2 | Pleural effusion | Failure to thrive | DNR |
| 8 | Female | 65 | Pancreatic adenocarcinoma | 3 | Liver, lymph nodes | Biliary obstruction | FULL |
| 9 | Female | 74 | Colon adenocarcinoma | 2 | Liver, lymph nodes | Spinal compression and pubic ramus fractures | FULL |
| 10 | Male | 67 | Rectal adenocarcinoma | 2 | Peritoneum, pleural effusion, perineal masses | Urinary tract infection | FULL |
| 11 | Female | 83 | Pancreatic adenocarcinoma | 4 | Liver, lungs, lymph nodes | Deep venous thrombosis | FULL |
| 12 | Female | 90 | Gastric adenocarcinoma | 3 | Lymph nodes, liver, lung, peritoneum | Refractory malignant ascites | DNR |
| 13 | Male | 77 | Esophageal adenocarcinoma | 3 | Lung, pleural effusion | Electrolyte derangements and cardiac arrhythmia | DNR |
| 14 | Female | 71 | Pancreatic adenocarcinoma | None (ineligible because of performance status) | Liver | Deep venous thrombosis | DNR |
| 15 | Female | 51 | Gastric adenocarcinoma | None (ineligible because of performance status) | Bone, lung, lymph nodes | Endocarditis | FULL |
| 16 | Female | 73 | Cancer of unknown enteric primary | None (ineligible because of performance status) | Left uterine wall, rectum | Failure to thrive and urinary tract infection | FULL |
| 17 | Male | 84 | Cholangiocarcinoma | 2 | Liver | Chronic subdural hematoma | FULL |
| 18 | Male | 78 | Colon adenocarcinoma | 4 | Liver | Infectious colitis and pneumonia | FULL |
| 19 | Male | 76 | Pancreatic adenocarcinoma | 2 | Liver, spine | Spinal cord compression | FULL |
| 20 | Female | 59 | Colon adenocarcinoma | 2 | Ovaries, lungs, liver, peritoneum, spleen, adrenal gland | Small bowel obstruction and urinary tract infection | DNR |
| 21 | Female | 81 | Colon adenocarcinoma | 5 | Lungs, lymph nodes, liver | Pneumonia, urinary tract infection and acute kidney injury | DNR |
| 22 | Male | 78 | Pancreatic adenocarcinoma | None (ineligible because of performance status) | Liver | Endocarditis | DNR |
DNR = Do-Not-Resuscitate.
Patient Outcomes
The median overall survival from the time of hospital discharge was 24 days (range 6–156 days). Seven patients (32%) died at the SAR to which they were discharged. Neither overall survival nor death at SAR varied across cancer types or hospital admission diagnoses. Only three patients (14%) lived beyond three months and none were alive at six months after discharge. Nine patients (41%) were transitioned to hospice within three months of discharge. Eight patients (36%) were readmitted within one month of discharge. Two patients (9%) followed up with their oncologists for outpatient appointments after discharge. None of the patients received further chemotherapy after discharge.
Comment
Our data show that in patients with metastatic GI cancer who had been extensively treated with chemotherapy, or who were ineligible for chemotherapy based on their performance status, and whose admitting diagnosis did not represent a reversible chemotherapy-related complication, discharges to SAR were associated with short survival and no further chemotherapy receipt. Given that SAR facilities are not typically staffed with palliative care and pain control expertise, these necessary services likely were not available to patients in our study when needed. As such, there may have been a detriment incurred by an SAR discharge that a hospice discharge may have avoided.
This study is the only one, to our knowledge, that specifically addresses the clinical question of what happens to inpatients with progressive and/or debilitating cancers after discharge to SAR. These data, therefore, can help establish goals of care in advanced GI cancer patients and help further guide discussions regarding appropriate discharge planning for this patient population. Although our sample size of 22 may seem small, the clinical setting is one where not uncommonly, a hope is voiced by oncologists, nurses, patients, and/or families, that “if you/I could only just get stronger with rehabilitation, then you/I could get more chemotherapy.” Although a hope for more cancer treatment in the face of a life-threatening illness is certainly understandable, our data indicate that SAR does not achieve this goal of further chemotherapy in such progressive GI malignancies.
Limited data from elsewhere agree with ours; in one study of a tertiary cancer center's quality improvement program aimed to improve the appropriateness of which inpatients are transferred to SAR, only one of 25 patients (4%) returned for further cancer-directed therapy.
6
The authors concluded that earlier and more in-depth discussions of care goals and hospice were warranted for such a sick and vulnerable patient population.6
Our study was limited in that it was retrospective and included a single service at a tertiary care center, potentially limiting the generalizability of the findings. Nonetheless, our patient sample did encompass a wide range of GI malignancies, which are a common and heterogeneous group of cancers. Future studies might investigate whether these conclusions hold true for other malignancies. It is also possible that SAR may have benefited some patients (e.g., by providing services that hospice could not, or maintaining hope in a way hospice seemed to not be able to) in ways we were unable to measure.
In conclusion, in GI cancer inpatients, a discharge to SAR could be more harmful than beneficial for patients and families with advancing or debilitating disease, and our study offers data to suggest that this question is worthy of further study. Advances in research, education, and health care policy (including advances in palliative care delivery) are needed
7
, 8
to improve the quality of the ways in which clinicians, particularly primary oncology teams, dually inform and empathize with9
these patients with serious illness who are often struggling at the end of life. Communicating how SAR might not help these particular patients and how it might potentially do harm, and instead, indicating how other avenues for end-of-life care could be pursued, represents one important strategy for facilitating more informed goals of care discussions.Disclosures and Acknowledgments
This research was funded in part through the NIH/NCI Cancer Center Support Grant P30 CA008748. The authors declare no conflicts of interest.
This work was presented in abstract form at the annual Palliative Care in Oncology Symposium of the American Society of Clinical Oncology (ASCO), September 10, 2016.
References
- Acute and subacute rehabilitation for stroke: a comparison.Arch Phys Med Rehabil. 1995; 76: 495-500
- Discussions of life expectancy and changes in illness understanding in patients with advanced cancer.J Clin Oncol. 2016; 34: 2398-2403
- Patients' expectations about effects of chemotherapy for advanced cancer.N Engl J Med. 2012; 367: 1616-1625
- Reasons why physicians do not have discussions about poor prognosis, why it matters, and what can be improved.J Clin Oncol. 2012; 30: 2715-2717
- Oncologists' and physiatrists' attitudes regarding rehabilitation for patients with advanced cancer.PM R. 2012; 4: 96-108
- A quality improvement initiative for improving appropriateness of referrals from a cancer center to subacute rehabilitation.J Pain Symptom Manage. 2014; 48: 127-131
- Primary palliative care education in specialty oncology training: more work is needed.JAMA Oncol. 2016; 2: 858-859
- Generalist plus specialist palliative care — creating a more sustainable model.N Engl J Med. 2013; 368: 1173-1175
- Current State of the art and science of patient-clinician communication in progressive disease: patients' need to know and need to feel known.J Clin Oncol. 2014; 32: 3474-3478
Article info
Publication history
Published online: April 21, 2017
Identification
Copyright
© 2017 American Academy of Hospice and Palliative Medicine. Published by Elsevier Inc.
User license
Elsevier user license | How you can reuse 
Elsevier's open access license policy
Elsevier user license
Permitted
For non-commercial purposes:
- Read, print & download
- Text & data mine
- Translate the article
Not Permitted
- Reuse portions or extracts from the article in other works
- Redistribute or republish the final article
- Sell or re-use for commercial purposes
Elsevier's open access license policy
ScienceDirect
Access this article on ScienceDirectLinked Article
- Subacute Rehabilitation Does Have Benefits for Patients With Advanced CancerJournal of Pain and Symptom ManagementVol. 55Issue 1
- PreviewIt is with concern that we read a recent Letter to the Editor by Desai et al.,1 that stated “people with gastrointestinal cancer did not benefit from the admission to subacute rehabilitation,” concluding that inpatient rehabilitation could do more harm than good. People with advanced cancer may not always achieve physical gains after rehabilitation but to state there is no benefit is an oversimplification. This was a very small sample size (n = 22), one-third of whom had metastatic pancreas cancer where rapid functional decline is seen in the last weeks of life.
- Full-Text
- Preview
