Background: Gastrostomy tubes (G-tubes) are invaluable clinical tools that play a role in palliation and nutrition in patients with cancer. This study aimed to better understand the risks and benefits associated with the placement and maintenance of G-tubes. Methods: Patients who underwent placement of a G-tube for cancer from January 2013 through December 2017 at a tertiary care center were considered for inclusion. Clinical data were retrospectively collected from medical records. Results: A total of 242 patients with cancer, whose average age at diagnosis was 61 years (range, 21–94 years), underwent G-tube placement for nutrition (76.4%), decompression (22.7%), or both (0.8%). Successful insertion was achieved in 96.8%, but 8 patients required >1 attempted method of insertion. In the decompression group, minor postplacement complications were less common (23.6% vs 53.5%; P<.001) and survival was shorter (P<.001) compared with the nutrition group. For those with decompressive G-tubes, 45.5% had a palliative care consult; 56.4% were seen by social workers; and 46.3% went to hospice. The frequency of hospice discharge was higher in patients who had consults (53.7% vs 23.1%; P=.01). Conclusions: Half of the patients who received decompressive G-tubes presented with stage IV disease and died within 1 month of placement. Those with >1 consult were more likely to be discharged to hospice. Patients with G-tubes for nutrition saw no change in functionality, complication rate, or survival, regardless of adjunct chemotherapy status. These findings illustrate the need for a tool to allow a better multidisciplinary approach and interventional decision-making for patients with cancer.
Gabrielle Gauvin, Chi Chi Do-Nguyen, Johanna Lou, Eileen Anne O’Halloran, Leigh T. Selesner, Elizabeth Handorf, Molly E. Collins, and Jeffrey M. Farma
Gabrielle Gauvin, Leigh Selesner, Marcin Chwistek, Molly Collins, Crystal Denlinger, Mark Itzen, LCSW, Kenneth Patrick, and Jeffrey Farma
Background: Malignant bowel obstruction (MBO) is a frequent presentation of advanced abdominal malignancy. Despite the frequency of MBO and the availability of NCCN Guidelines for its management, clinicians struggle to create a personalized care plan accounting for a patient’s current condition, prognosis, and goals of care. The aim of this project is to develop a discussion tool to implement on admission to ensure realistic prognostication and appropriate involvement of the patient, caregivers, and multidisciplinary team in treatment decisions. Methods: At our tertiary care center, a multidisciplinary team composed of surgical oncologists, medical oncologists, palliative care specialists, hospitalists, social workers, and nutritionists developed an algorithm to implement when a patient is admitted with MBO. The algorithm was influenced by the NCCN Guidelines for palliative care but was designed as a single-page checklist that could be easily executed by all team members. Results: The MBO admission checklist is divided into 4 sections. The general assessment section includes information about the etiology, severity, and reversibility of the current and past MBO. Also, the patient’s resuscitation and functional status are reviewed. The cancer care coordination/communication section covers the patient’s oncological history and estimated lifespan, as well as the need for further coordination of care and a goals of care discussion. The subsequent portion covers the management plan, with specific decisions resulting from a goals of care discussion as well as the need for specialized consults. The medical, procedural, and operative interventions the team and patient discussed as options are also documented. A key portion is the risks and benefits discussion for each treatment modality. Lastly, the discussion about future considerations section explores the patient’s wishes for the next episode of MBO. Conclusions: The goal of this checklist is to ensure that a comprehensive discussion is held between the different services involved in a patient’s care at every admission for MBO in order to provide a personalized management plan and improve communication. The next step in our study is to assess the effectiveness of implementation of the checklist as well as its impact on patient satisfaction, quality of life, early hospice referral, and outcomes.
Leigh Selesner, Gabrielle Gauvin, Dorotea Mutabdzic, Eileen O’Halloran, Maxwell Kilcoyne, Kwan-Keat Ang, Jeffrey Farma, Elin Sigurdson, and Sanjay Reddy
Introduction: Cytoreductive surgery and hyperthermic intraperitoneal chemotherapy (CS/HIPEC) has led to improved survival in select patients with peritoneal surface malignancies. Predicting the volume of disease and any unresectable disease is important for determining CS candidacy. Computed tomography (CT) scan is the preoperative assessment of choice, and diagnostic laparoscopy (DL) is also supported in the literature but has not been widely adopted. In this study, we report our experience comparing and evaluating the role of imaging and DL in the preoperative assessment of patients being considered for CS/HIPEC. Methods: Patients considered for CS/HIPEC at our tertiary cancer center between January 2012 and December 2017 were included. Diagnostic modality sensitivity and specificity were calculated by comparing findings on CT scan and DL to findings at the time of laparotomy and on final pathology. Specificity and sensitivity of the 2 modalities were compared using the McNemar Chi-square test. Results: Our analysis included 71 patients (60.5% male, mean age of 54.9) seen in consultation for CS/HIPEC. Primary cancer diagnosis was 57.7% colorectal cancer, 25.4% pseudomyxoma peritonei, 8.5% mesothelioma, and 8.5% adenocarcinoma of unknown primary. DL was done in 42.3% of patients (median time of 30 days between CT and DL) and an open procedure was done directly after CT in 39.4% (median interval time of 39 days). Findings of DL identified 70% as being unresectable and hence ineligible for HIPEC. The median interval time between 2 operations was 29 days (range, 16–42). When comparing diagnostic modalities to open surgery and final pathology, CT had a sensitivity and specificity of 48.2% and 76.4% and DL, 68.2% and 88.9%, respectively. DL was significantly more sensitive and specific than CT (χ2=5.54; P<.0186) at predicting ascites, small bowel, omental, liver, and lymph node involvement. Conclusion: Our results support the recommendation for performing DL prior to open exploration in patients considered for CS/HIPEC. In our cohort, DL was significantly more sensitive and specific than CT in predicting disease volume and distribution. While there is obviously greater risk to an invasive modality compared to non-invasive CT scan, routine performance of DL can potentially avoid laparotomy without CS/HIPEC in a large proportion of patients. These results will be used to inform the next phase of our study: a prospective clinical trial.