Non-compliance with end-of-life parenteral nutrition prescription recommendations: retrospective study of 1,260 cancer patients

Introduction

Adult cancer patients account for a high percentage, sometimes even the majority, of subjects receiving parenteral nutrition (PN) (1). Patients with advanced cancer and a good prognosis should receive adequate nutritional counselling and support including, if required, PN but only after considering together patients’ prognosis and the expected benefit on quality of life, as well as the burden associated with nutritional care (2). Additionally, iatrogenic deterioration of nutritional status should be prevented by adequate prophylactic or symptomatic supportive care comprising “permissive” nutritional support in patients undergoing palliative anti-cancer treatment (2). There remain many controversial issues for and against PN, with proponents emphasizing the social and cultural value of nutrition and patients’ fear of starvation, and opponents pointing to very little evidence to support its use due to the lack of demonstrated benefit on quality of life and overall survival in end-of-life patients (3-5). Similarly, PN has not been proven effective in advanced cancer (3) and malignant bowel obstruction (MBO) (6). Moreover, blood stream infections (BSI) are more frequent in patients with advanced cancer and PN (7-9). Overall, there is an agreement that unconditional artificial nutrition in all patients undergoing curative or palliative anticancer therapy is associated with more harm (hyperglycemia, hypertriglyceridemia, refeeding syndrome, oedema, PN-associated liver disease) than benefit, therefore, such practice should not be used (2). For dying patients, European Society of Parenteral and Enteral Nutrition (ESPEN) guidelines outline the ineffectiveness of artificial nutritional support to improve comfort and quality of life, thereby excluding most indications for nutritional support (2). American Society of Clinical Oncology (ASCO) and National Comprehensive Cancer Network (NCCN) guidelines also state that discontinuation of previously initiated PN near the end of life (EOL) is appropriate (10,11), and ethically permissible as nutritional support cannot reverse weight loss in these patients (11). However, the impact of guidelines has been limited due to the frequently moderate interest of oncologists in nutritional aspects of cancer care (2). This lack of interest still results in an overuse of PN near EOL in cancer patients (12).

This study aims to assess the application of international guidelines by evaluating the use of PN near EOL in patients with advanced cancer. Our secondary objective was to identify factors associated with the use of PN, and the incidence of BSI when using PN in this population. We present the following article in accordance with the STROBE reporting checklist (available at https://apm.amegroups.com/article/view/10.21037/apm-22-499/rc) (13).

Methods

Study design and cohort selection

We included all patients who were diagnosed with advanced or metastatic cancer and died during their hospitalization in 2013, 2015, 2017 and 2019.

The inclusion center is a French cancer center caring for more than 5,000 new patients yearly. Data were extracted from our hospital register named Programme de Médicalisation des Systèmes d’Information (PMSI). This register collects healthcare claims for every patient’s admission. All data were anonymized. The study was conducted in accordance with the Declaration of Helsinki (as revised in 2013). At their first consultation, all patients were informed that unless they disagree, their clinical data can be used. The Ethics Committee of our institute (Avignon-Provence Cancer Institute) approved this project as part of our clinical practice self-assessment (first part: chemotherapy in end-of-life cancer patients, second part: PN in end-of-life cancer patients). According to the Jardé decree (November 17th 2016 in France), written informed consent and registration were not required. First, we analyzed only previously and routinely collected information. Individual patient data were documented anonymously. Secondly, patients are informed that anonymized data can be analyzed and collected unless they are opposed.

Main outcome measure and definitions

The main outcome was the use of PN during the hospital stay. Factor associated with the use of PN were analyzed. The rate of BSI when using PN at the EOL of patients with advanced cancer was also evaluated.

PN was defined as an intravenous administration through a central venous access device (port-a-cath or peripherally inserted central catheter), of nutrients using mixtures including at least aminoacids, vitamins electrolytes and minerals (14,15). Parenteral hydration without the above-mentioned elements was not considered as a form of PN. Malnutrition was classified in adult cancer patients and those over 70 years of age as severe or moderate, according to the French guidelines (16).

The incidence of BSI was confirmed by microbiological analyses. Blood samples were drawn in case of clinical signs suggestive of infection. Because of the difficulty to draw at the same time one blood sample from a percutaneous vein and from the catheter tip in EOL cancer patients, we chose to study BSI and not central line associated BSI (17). BSI was defined as the presence of viable bacteria or fungi in the blood (18). PN and BSI were considered in the last hospital stay before the patient’s death.

Clinical factors potentially associated with PN use were evaluated. Especially, the diagnosis of MBO was established based on history (metastatic intra-abdominal cancer) and physical examination and was confirmed, when possible, with imaging studies (19).

Data collection

We collected the following data from our computerized hospital systems: (I) demographical patient data as age, sex, length of the hospitalization where death occurred, body mass index (BMI) and Eastern Cooperative Oncology Group performance status (ECOG PS) at onset of hospitalization; (II) previous antitumor treatments, referent physician’s specialty (medical or radiation oncology, referent physicians were not specifically trained in nutrition but a dietician was systematically involved in the treatment decisions and could seek help from a nutritionist, and professional experience); (III) follow-up by a palliative care team (PCT); (IV) cancer type: digestive, urinary, respiratory tract, gynecological, head and neck or male cancers, miscellaneous tumors, malignant hemopathy and a separate group of other tumors (sarcoma, melanoma, neuroendocrine tumor and primary central nervous system cancer) for which treatment decisions are discussed in regional specialized multidisciplinary tumor boards; (V) PN, complete/incomplete MBO, and BSI.

Statistical analysis

Descriptive statistics are presented as means [standard deviation (SD)] and medians (ranges) for continuous variables. Discrete variables are reported as counts (percentages). A non-parametric Mann-Whitney test was applied to compare the distribution of continuous variables, and χ² test (or Fisher’s exact test when appropriate) was used to test the association of categorical variables. Continuous variables were then transformed into categorical variables with the median or a predetermined threshold. Performance status was separated into ECOG PS 0–2 and ECOG PS 3–4.

Potential factors associated with PN were compared on univariate analysis first with χ² test or Fisher’s exact test when appropriate. A multivariate logistic model based on selected parameters from the univariate analysis (the level of significance was set at a P value of <0.20 for selection) was then performed. In addition, three procedures for selecting variables (forward, backward and stepwise) by minimization of Akaike’s information criterion were performed to obtain the most appropriate logistic regression model. Among the variables introduced into the logistic regression model, multiple imputation was performed on variables with missing data (complete/incomplete bowel obstruction and BMI <22.14 kg/m²), since missing values were less than 5%.

A sensitivity analysis was carried out using a regression model without multiple imputation, which found the same statistically significant variables. A P value of <0.05 was considered to be statistically significant, and all statistical tests were two-sided. Statistical analysis was performed using R software system (version 3.6.1, Vienna, Austria) version 4.0.3.

Results

Patients’ characteristics (Figure 1, Table 1)

Figure 1 Flow chart. ATT, anti-tumor treatment; MBO, malignant bowel obstruction; ECOG PS, Eastern Cooperative Oncology Group performance status; BMI, body mass index.

In 2013, 2015, 2017 and 2019, 1,283 patients died, 1,260 patients were included, and 23 patients were excluded because they had a non-locally advanced or a non-metastatic cancer. There was no change in the rate of PN between 2013 and 2019; 80% of the patients had a poor performance status (ECOG PS >2). The mean BMI was 22.8±4.9 kg/m² (median: 22.14 kg/m²). Most patients (89%) were severely malnourished. The median duration of hospitalization was 15 days (range, 1–80 days). One hundred and sixty-two (13%) patients had a complete/incomplete bowel obstruction. PN was prescribed in 574 (45.6%) patients, mean duration of PN and mean survival from start PN were 10±9.7 and 14.7±11.8 days. BSI was diagnosed in 113 patients (9%).

Factors associated with PN

In univariate analysis (Table 1), MBO (19% vs. 7.9%), digestive tract cancers (31% vs. 24%), a BMI <22.14 kg/m² (59% vs. 43%), severe malnutrition (93% vs. 85%), previous antitumor treatment (81% vs. 73%) and a length of hospitalization >12 days (63% vs. 43%) were associated with a higher rate of PN at the EOL. The rate of PN was lower among patients with a urinary tract cancer (6.6% vs. 10%). In multivariate analysis (Table 2), patients with severe malnutrition (OR =2.36; 95% CI: 1.42–4.02), a MBO (OR =2.25; 95% CI: 1.44–3.56), a length of hospitalization >12 days (OR =2.21; 95% CI: 1.67–2.94), a BMI <22.14 kg/m² (OR =2.02; 95% CI: 1.52–2.67) were more frequently prescribed a PN whereas patients with a urinary tract cancer received less PN (OR =0.59; 95% CI: 0.39–0.99). The rate of BSI was also higher in patients receiving a PN (13% vs. 6%), this association remaining statistically significant in multivariate analysis (OR =2.01; 95% CI: 1.18–3.54).

Discussion

Nowadays, around 25,000 patients are estimated to be on home PN in the United States (20). In a retrospective study on patients with advanced cancer, home artificial nutrition was prescribed only to 333 of 29,348 cancer patients (1.1%) (5). Conversely, in the French settings, PN has been found to be much higher and reached up to 13.1% in a study that included 1,301 patients (21). Focusing on EOL cancer patients, the frequency of PN has been estimated to range from 2% to 32% within the last week of life in various countries (12). The higher rates in our study (45.6% of PN near the EOL), are consistent with an overuse of PN in the French settings.

This high figure can easily be explained by the symbolic value of nutrition in cancer patients and ethical, emotional, and cultural aspects heavily influence the decision-making process (22). Indeed, most patients, families, and healthcare professionals believe that increasing food and liquid intake is essential to stave off physical deterioration (22), which is not supported by the available evidence (3). Furthermore, the majority of patients and families express the fear of starvation (22). There is also a strong cultural attachment to food in France. In fact, PN neither improves quality of life nor survival in patients with advanced cancer as demonstrated by two randomized controlled trials. In the first one, 309 cancer patients received erythropoietin and a cyclo-oxygenase inhibitor without (n=170) or with a nutritional support (n=109, about half of patients with PN) (23). No quality-of-life (QoL) assessment was performed but there was no statistically significant difference in survival between intervention and control patients (23). The second trial randomized 70 patients in the PN group and 78 in the oral feeding group (24). The main findings were the lack of improvement in QoL and a negative trend for overall survival among patients treated with PN (median survival 2 vs. 3 months) (24). We assume that the French cultural settings might influence the use of PN resulting in the overuse of PN found in the studies on French populations.

In a nationwide study including 32 palliative care units and 1,083 patients, nausea, vomiting, and gastrointestinal obstruction were the most common indications for PN in palliative patients (25). We showed MBO to be another important factor associated with end-of-life PN (OR =2.25; 95% CI: 1.44–3.56). According to the ESPEN guidelines, total inability to eat (malignant bowel failure or complete obstruction) can require timely implementation of artificial nutrition to avoid starvation (2). The position statement of the British Association for Parenteral and Enteral Nutrition (BAPEN) on intestinal failure in patients with malignancy indicates that all patients who may be considered for PN should have an early multidisciplinary team review (nutrition, oncology and PCTs) (26). BAPEN also states that the plans for withdrawal, either at the EOL or any time after commencing, should be discussed (26). The potential benefits of PN in MBO have never been formally established; the Cochrane Library was unable to perform a metanalysis, but data synthesized via a narrative summary did not highlight evidence of the benefit of PN in terms of survival or quality of life (6).

Because of the very low amount of data on this subject, we did not investigate whether the rate of complications was associated with some components of PN. In our study, BMI <22.14 kg/m² (OR =2.02; 95% CI: 1.52–2.67) and severe malnutrition (OR =2.36; 95% CI: 1.42–4.02) were found to be independently associated with EOL PN. Malnutrition (low BMI +/− weight loss) (16), or cachexia (weight loss +/− low BMI +/− sarcopenia) (27) occurs in 50–80% of patients with advanced cancer (27,28). The main goal of guidelines is treatment of malnutrition and metabolic derangements in cancer patients (2,10,11). So, ESPEN recommendations leave the possibility of a “permissive” nutritional support in patients with advanced cancer (2). Due to their life expectancy (mean survival of 15 days from the start of hospitalization), most patients we included in our study had a refractory cachexia (27). At this stage of the disease trajectory, cancer patients do not respond to artificial nutrition aimed at reversing cachexia (27,29). In any case, in hospitalized patients with a normal digestive tract, algorithms of good clinical practice recommend beginning with an oral nutrition support, to try, if unsuccessful, enteral nutrition and to prescribe PN only as a “last resort” (2,30,31). But as PN is an easier and more medical response to nutritional care requests, and as there is an overestimation of cancer patients’ survival (32), it is difficult to stop PN (22). We assume that it might contribute to PN sustained at EOL.

Most patients with body weight loss have three or more other uncontrolled symptoms, the most common being early satiety, constipation, nausea or vomiting, mood changes, abnormal taste and smell, abdominal pain (33). Furthermore, cancer patients often experience at EOL a symptom burden that can have an impact on food intake (34). These data advocate for an integrative care intervention and the results of a recent review suggest that a team-based approach, involving palliative care physicians, nurses, dietitians, and occupational therapists, was associated with significant improvement in food intake, nutritional indices, and distress, concurrent with successful management of pain and symptoms (35). Palliative and supportive care teams typically do not yet offer psychosocial interventions for cachexia-related distress. These interventions, tailored to patients’ prognosis and cachexia severity, may help patients and family members to cope with cachexia that can render active management of body weight loss no longer possible or appropriate.

Finally, the incidence of BSI was 9%. The rate of BSI has been estimated between 0.27–2.78 episodes/1,000 days in patients with advanced cancer (36). Our results are higher but severe malnutrition and metastatic disease, as seen in end-of-life cancer patients we included, double the risk of developing BSI (37). The other factor to consider is probably the high rate of PN in EOL cancer patients in our study as we highlighted the correlation between PN and BSI (OR =2.01; 95% CI: 1.18–3.54). Total PN administered via central venous catheter has been identified as an independent risk factor of developing BSI in two large studies: (OR =2.65; 95% CI: 2.20–3.19) among 38,674 hospitalized patients with central line, (OR =4.33; 95% CI: 2.50–7.48) in a survey of 4,840 patients in a large tertiary care hospital in New York City who received PN (8,9). A higher incidence of BSI was also demonstrated in cancer patients with PN when compared to those with enteral nutrition [risk ratio (RR) =1.11; 95% CI: 1.04–1.19] in a recent meta-analysis of 43 studies and 2,827 patients (7).

Our study did not allow us to have a precise estimate of BSI. Due to their clinical condition, many patients had incomplete blood samplings. Most clinical records did not provide body weight curve follow-up or PN indications and we did not look at the dose given, which may explain in part the high percentage of PN we noted. We also did not include patients who had died outside of our cancer center, which would have resulted most likely in a decrease of the rate of BSI, because patients with infectious complications are prone to be hospitalized more often. Our study is therefore not representative of PN in all cancer patients at the EOL. Another limit of our study is its retrospective design do not necessarily produce an accurate portrait of care provided to patients who are dying (38).

Conclusions

We showed that PN near the EOL of patients with advanced cancer was overused. MBO, BMI, severe malnutrition and length of hospitalization were correlated with end-of-life PN in cancer patients, and we also demonstrated a higher rate of BSI in patients with PN. These findings argue for the use of a survival estimation tool when considering PN and for a multidisciplinary integrative care intervention in order to discuss and re-assess the benefit-risk ratio of PN in EOL cancer patients.

Acknowledgments

Funding: None.

Footnote

Reporting Checklist: The authors have completed the PRISMA reporting checklist. Available at https://apm.amegroups.com/article/view/10.21037/apm-22-499/rc

Data Sharing Statement: Available at https://apm.amegroups.com/article/view/10.21037/apm-22-499/dss

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://apm.amegroups.com/article/view/10.21037/apm-22-499/coif). PD reports personal fees from SANOFI and non-financial support from Bayer Pharma and LVL. The other authors have no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. The study was conducted in accordance with the Declaration of Helsinki (as revised in 2013). At their first consultation, all patients were informed that unless they disagree, their clinical data can be used. The Ethics Committee of our institute (Avignon-Provence Cancer Institute) approved this project as part of our clinical practice self-assessment (first part: chemotherapy in end-of-life cancer patients, second part: PN in end-of-life cancer patients).

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Bakker H, Bozzetti F, Staun M, et al. Home parenteral nutrition in adults: a uropean multicentre survey in 1997. ESPEN-Home Artificial Nutrition Working Group. Clin Nutr 1999;18:135-40. [Crossref] [PubMed]
2. Arends J, Bachmann P, Baracos V, et al. ESPEN guidelines on nutrition in cancer patients. Clin Nutr 2017;36:11-48. [Crossref] [PubMed]
3. Good P, Richard R, Syrmis W, et al. Medically assisted nutrition for adult palliative care patients. Cochrane Database Syst Rev 2014;CD006274. [Crossref] [PubMed]
4. Baumstarck K, Boyer L, Pauly V, et al. Use of artificial nutrition near the end of life: Results from a French national population-based study of hospitalized cancer patients. Cancer Med 2020;9:530-40. [Crossref] [PubMed]
5. Ruggeri E, Agostini F, Fettucciari L, et al. Home artificial nutrition in advanced cancer patients. Tumori 2013;99:218-24. [Crossref] [PubMed]
6. Sowerbutts AM, Lal S, Sremanakova J, et al. Home parenteral nutrition for people with inoperable malignant bowel obstruction. Cochrane Database Syst Rev 2018;8:CD012812. [PubMed]
7. Chow R, Bruera E, Arends J, et al. Enteral and parenteral nutrition in cancer patients, a comparison of complication rates: an updated systematic review and (cumulative) meta-analysis. Support Care Cancer 2020;28:979-1010. [Crossref] [PubMed]
8. Fonseca G, Burgermaster M, Larson E, et al. The Relationship Between Parenteral Nutrition and Central Line-Associated Bloodstream Infections: 2009-2014. JPEN J Parenter Enteral Nutr 2018;42:171-5. [PubMed]
9. Ippolito P, Larson EL, Furuya EY, et al. Utility of Electronic Medical Records to Assess the Relationship Between Parenteral Nutrition and Central Line-Associated Bloodstream Infections in Adult Hospitalized Patients. JPEN J Parenter Enteral Nutr 2015;39:929-34. [Crossref] [PubMed]
10. Roeland EJ, Bohlke K, Baracos VE, et al. Management of Cancer Cachexia: ASCO Guideline. J Clin Oncol 2020;38:2438-53. [Crossref] [PubMed]
11. National Comprehensive Cancer Network. Guidelines for Palliative Care (Version 2.2020). Available online: https://www.nccn.org/store/login/login.aspx?ReturnURL=https://www.nccn.org/professionals/physician_gls/pdf/palliative.pdf
12. Raijmakers NJH, van Zuylen L, Costantini M, et al. Artificial nutrition and hydration in the last week of life in cancer patients. A systematic literature review of practices and effects. Ann Oncol 2011;22:1478-86. [Crossref] [PubMed]
13. von Elm E, Altman DG, Egger M, et al. Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studies. BMJ 2007;335:806-8. [Crossref] [PubMed]
14. Hamdan M, Puckett Y. Total Parenteral Nutrition. Treasure Island, FL, USA: Stat Pearls Publishing, 2022.
15. Kovacevich DS, Corrigan M, Ross VM, et al. American Society for Parenteral and Enteral Nutrition Guidelines for the Selection and Care of Central Venous Access Devices for Adult Home Parenteral Nutrition Administration. JPEN J Parenter Enteral Nutr 2019;43:15-31. [Crossref] [PubMed]
16. Haute Autorité de Santé. Malnourishment diagnosis in adults and children. 2020. Available online: https://www.has-sante.fr/jcms/p_3165946/fr/diagnostic-de-la-denutrition-de-la-personne-agee-note-de-cadrage
17. Mermel LA, Allon M, Bouza E, et al. Clinical practice guidelines for the diagnosis and management of intravascular catheter-related infection: 2009 Update by the Infectious Diseases Society of America. Clin Infect Dis 2009;49:1-45. [Crossref] [PubMed]
18. Magadia RR, Weinstein MP. Laboratory diagnosis of bacteremia and fungemia. Infect Dis Clin North Am 2001;15:1009-24. [Crossref] [PubMed]
19. Ripamonti CI, Easson AM, Gerdes H. Management of malignant bowel obstruction. Eur J Cancer 2008;44:1105-15. [Crossref] [PubMed]
20. Mundi MS, Pattinson A, McMahon MT, et al. Prevalence of Home Parenteral and Enteral Nutrition in the United States. Nutr Clin Pract 2017;32:799-805. [Crossref] [PubMed]
21. Hébuterne X, Lemarié E, Michallet M, et al. Prevalence of malnutrition and current use of nutrition support in patients with cancer. JPEN J Parenter Enteral Nutr 2014;38:196-204. [Crossref] [PubMed]
22. Del Río MI, Shand B, Bonati P, et al. Hydration and nutrition at the end of life: a systematic review of emotional impact, perceptions, and decision-making among patients, family, and health care staff. Psychooncology 2012;21:913-21. [Crossref] [PubMed]
23. Lundholm K, Daneryd P, Bosaeus I, et al. Palliative nutritional intervention in addition to cyclooxygenase and erythropoietin treatment for patients with malignant disease: Effects on survival, metabolism, and function. Cancer 2004;100:1967-77. [Crossref] [PubMed]
24. Bouleuc C, Anota A, Cornet C, et al. Multicenter randomized controlled trial for advanced cancer patients receiving parenteral nutrition (PN) versus oral feeding (OF): Results of AlimK study. J Clin Oncol 2018;36:10029. [Crossref]
25. Orrevall Y, Tishelman C, Permert J, et al. A national observational study of the prevalence and use of enteral tube feeding, parenteral nutrition and intravenous glucose in cancer patients enrolled in specialized palliative care. Nutrients 2013;5:267-82. [Crossref] [PubMed]
26. Naghibi M, Woodward J, Neild P, et al. The British Intestinal Failure Alliance (BIFA) Position Statement Palliative parenteral nutrition (HPN) for patients with malignancy. 2020. Available online: https://www.bapen.org.uk/pdfs/bifa/position-statements/position-statement-on-palliative-hpn-for-patients-with-malignancy-dec-2020.pdf
27. Fearon K, Strasser F, Anker SD, et al. Definition and classification of cancer cachexia: an international consensus. Lancet Oncol 2011;12:489-95. [Crossref] [PubMed]
28. Fearon KC, Glass DJ, Guttridge DC. Cancer cachexia: mediators, signaling, and metabolic pathways. Cell Metab 2012;16:153-66. [Crossref] [PubMed]
29. Cotogni P, Stragliotto S, Ossola M, et al. The Role of Nutritional Support for Cancer Patients in Palliative Care. Nutrients 2021;13:306. [Crossref] [PubMed]
30. Bounoure L, Gomes F, Stanga Z, et al. Detection and treatment of medical inpatients with or at-risk of malnutrition: Suggested procedures based on validated guidelines. Nutrition 2016;32:790-8. [Crossref] [PubMed]
31. Schuetz P, Seres D, Lobo DN, et al. Management of disease-related malnutrition for patients being treated in hospital. Lancet 2021;398:1927-38. [Crossref] [PubMed]
32. Glare P, Virik K, Jones M, et al. A systematic review of physicians’ survival predictions in terminally ill cancer patients. BMJ 2003;327:195-8. [Crossref] [PubMed]
33. Del Fabbro E, Hui D, Dalal S, et al. Clinical outcomes and contributors to weight loss in a cancer cachexia clinic. J Palliat Med 2011;14:1004-8. [Crossref] [PubMed]
34. Amano K, Morita T, Miyamoto J, et al. Perception of need for nutritional support in advanced cancer patients with cachexia: a survey in palliative care settings. Support Care Cancer 2018;26:2793-9. [Crossref] [PubMed]
35. Amano K, Baracos VE, Hopkinson JB. Integration of palliative, supportive, and nutritional care to alleviate eating-related distress among advanced cancer patients with cachexia and their family members. Crit Rev Oncol Hematol 2019;143:117-23. [Crossref] [PubMed]
36. O’Hanlon FJ, Fragkos KC, Fini L, et al. Home Parenteral Nutrition in Patients with Advanced Cancer: A Systematic Review and Meta-Analysis. Nutr Cancer 2021;73:943-55. [Crossref] [PubMed]
37. Gudiol C, Aguado JM, Carratalà J. Bloodstream infections in patients with solid tumors. Virulence 2016;7:298-308. [Crossref] [PubMed]
38. Bach PB, Schrag D, Begg CB. Resurrecting treatment histories of dead patients: a study design that should be laid to rest. JAMA 2004;292:2765-70. [Crossref] [PubMed]