A Randomized Phase II Study To Assess the Effectiveness of Fluid Therapy or Intensive Nutritional Support on Survival in Patients with Advanced Cancer Who Cannot be Nourished via Enteral Route

Abstract

Background:

Experts advise against parenteral nutrition (PN) for patients with advanced cancer at the end of life. But because many patients and families fear starvation, many physicians administer PN to patients with terminal cancer in Korea.

Objective:

We designed this study to investigate the effect of PN on survival in patients with terminal cancer at the end of life.

Design:

We planned a randomized phase II study enrolling 116 patients randomized to receive either fluid or PN.

Setting/Subjects:

Eligible patients are who could not tolerate enteral feeding and had short life expectancies (<3 months) due to progressive cancer. Patients with functioning bowels were excluded.

Measurements:

The primary end point was overall survival and the secondary end point was total administered calories.

Results:

We prospectively enrolled 31 consecutive patients and 16 patients were assigned to the PN group. The study ended early because many patients and families were extremely concerned about starvation. The baseline characteristics, including nutritional parameters, were not significantly different between the two groups. The mean administered calories was 374.7 (±71.7) kcal/d for the fluid group and 1286.8 (±108.3) kcal/d for the PN group (p<0.001). Median survival was 8 days (95% confidence interval [CI], 5.7–10.3 days) in the fluid group and 13 days (95% CI: 3.1–22.9 days) in the PN group, and this difference was not statistically significant (p=0.982 by Log-rank test).

Conclusions:

This study did not conclusively determine the role of PN for patients with advanced cancer, however, PN support failed to significantly prolong survival in these patients compared to similar patients receiving only fluid.

Introduction

Experts advise against the use of parenteral nutrition (PN) for patients with advanced cancer at the end of life. Most treatment guidelines from Western countries do not recommend PN for terminally ill patients with cancer, while some guidelines even prohibit PN in these cases.^1–3 These recommendations, however, are based primarily on retrospective data as prospective controlled trials are rare. In Korea, many people believe that the inability to consume any food kills the patient sooner than expected. Therefore, in clinical practice, many physicians administer PN to patients with advanced cancer until death.⁴

Some studies suggest that PN may prolong survival of patients with advanced cancer who cannot tolerate enteral nutrition (EN), but data with regards to PN at the end of life are insufficient. Thus, we designed this study to investigate the effect of PN on prolonging survival at the end of life in patients with terminal cancer.

Design and Methods

Patient eligibility

Study patients were consecutively enrolled at the Seoul Medical Center from June 2011 to December 2011. The Institutional Review Board approved this study. Eligibility criteria included: (1) a diagnosis of advanced cancer with no further plans for anticancer treatment; (2) inability to feed via an enteral route; (3) age18 years or older; (4) life expectancy 12 weeks or less; (5) Eastern Cooperative Oncology Group (ECOG) performance status of 3 or 4; (6) the presence of a venous access device for administration of fluid or intravenous nutrients; (7) admission to the hospital for a minimum of 1 day for the initial period of this study; and (8) written informed consent. Patients were excluded if any one of the following criteria was met: (1) cardiac or renal disease that restricted the administration of fluid; (2) an electrolyte imbalance that required immediate correction; (3) poorly controlled diabetes (HbA_1c greater than 8% despite therapy); or (4) an indication of unsuitability for participating in the clinical trial as determined by the attending physician.

Study design and assessment

Eligible patients were randomly assigned to either the control arm or the PN arm at a 1:1 ratio. The permuted-block randomization method was used for patient assignment. The block size was four. Random allocation was made by research staff of Seoul Medical Center Research Institute. Allocated group was announced to investigators at the time of assignment of each patient by telephone call. Before randomization, all eligible participants received tests for anthropometric data, complete blood count, and blood chemistry relevant to nutritional status.

Treatment

In the control arm, participants were provided intravenous fluid. The total amount of fluid was determined by the attending physician with a maximum of 30 mL/kg per day in addition to replacement of abnormal losses from the previous day to meet the physiologic fluid requirement of healthy adults. The fluids were normal saline, half saline, or dextrose water. Decision of total administering calories was made by attending physician, but limited to under the 20 kcal/kg per day, which is the minimum energy requirement of a bedridden patient.⁵ In the PN arm, patients were prescribed PN. The Nutritional Support Team determined the PN composition during initial periods of study treatment. All type of marketed intravenous amino acid and fat emulsions were allowed, including ready-to-use products. Treatment was continued from randomization to patient's death or withdrawal of consent.

Statistical analyses

The primary end point was overall survival. Survival was defined as the time from randomization to death or to withdrawal from the study. The secondary end point was total administered calories. At randomization, 53 patients were required to be in each arm to achieve 90% statistical power with a two-sided α error of 0.05 based on the Simon selection design, assuming expected survival of 40 days in the control group and 60 days in the PN group with 1 year of accrual time. This assumption was based on the literature showing that patients with malignant bowel obstruction have a median survival of around 40 days.¹³ Based on these parameters, the planned sample size was 116 patients, which took into account a 10% dropout rate. The Student's t test was used for comparison of mean values. The Kaplan-Meier method was used to analyze survival, and the log-rank test was used to compare the differences in survival between the two groups. All values were analyzed using IBM SPSS statistics version 20 (Armonk, NY). All statistical tests were two-sided, and a p value of<0.05 was considered significant.

Results

Patients

From June 2011 to December 2011, a total of 31 patients were randomly assigned to each treatment arm. Fifteen patients were allocated to the control arm and 16 patients were in the PN arm. All patients' data were included in the analysis. The median age was 59 years (range, 40–83 years). The study was closed early because of poor patient accrual, and it was expected that the target population could not be achieved within the study time frame. Many patients and families were repulsed by the idea of the study due to their concerns about starving to death if the patient was allocated to the control arm.

The baseline demographics, anthropometric and laboratory data, and clinical characteristics of participants according to study groups are summarized in Table 1. Because we did not stratify patients according to diagnosis, primary site of malignancy is not evenly distributed. Table 2 shows the distribution of body mass index (BMI) in both study groups.

Table 1.

Baseline Demographic and Nutritional Parameters Across Study Groups

	Control group (n=15)	PN group (n=16)	p
Gender
Male	9	10	0.589^a
Performance status			0.106^a
ECOG 3	6	11
ECOG 4	9	5
Type of cancer
Hepatobiliary and pancreas	8	2	0.023^b
Colon	3	4
Stomach	0	4
Breast	2	1
Neuroendocrine	0	2
Lung	0	1
Prostate	0	1
Melanoma	1	0
Salivary gland	0	1
Leukemia	1	0
Reason for PN
MBO by peritoneal seeding	4	7	0.458^b
MBO by mass	4	5
Subocclusion by massive liver metastases	4	3
Subocclusion by massive LN metastases	1	1
Bowel rest due to bleeding	2	0
Age (years)	59.1±9.6^c	60.4±12.6	0.749
Hemoglobin (g/dL)	9.6±2.4	8.3±1.9	0.126
Total lymphocyte (per μL)	1457.3±866.6	973.0±844.7	0.126
Albumin (g/dL)	2.7±0.4	2.6±0.4	0.379
Total cholesterol (mg/dL)	132.0±69.4	146.1±36.5	0.646

Calculated by χ² test.

Calculated by Fisher's exact test. This two p values were presented only for items with biggest difference, all unwritten values are bigger than 0.05.

Continuous values are expressed as mean±SD. The p values were calculated by t test.

PN, parenteral nutrition; ECOG, Eastern Cooperative Oncology Group; MBO, malignant bowel obstruction; LN, intra-abdominal lymph nodes.

Table 2.

Distribution of Body Mass Index According to World Health Organization Classification in Both Study Groups

		Frequency (%)
Category	BMI (kg/m²)	Control group	PN group	p
Underweight	<18.5	1 (6.7)	4 (25)	0.333
Normal	18.5–24.9	9 (60)	10 (62.5)	0.886^a
Overweight	25.0–29.9	3 (20)	2 (12.5)	0.654
Obese	≥30.0	2 (13.3)	0 (0)	0.226
Average BMI		23.2±4.1	21.1±3.4	0.126^b

Calculated by χ² test.

Student's t test; other p values were calculated by Fisher's exact test.

BMI, body mass index.

Survival

For all patients, the median survival was 9 days (95% confidence interval [CI], 3.7–14.3 days). All patients except one were dead at the time of analysis. One patient returned to her home country during the study period. The median survival of the control group was 8 days (95% CI, 5.7–10.3 days), while the median survival of the PN group was 13 days (95% CI, 3.1–22.9 days). This difference was not statistically significant (p=0.982 by Log-rank test; Fig. 1). Survival time was not affected by baseline characteristics or by the assigned study group.

FIG. 1.

Survival of both study groups are analyzed by Kaplan-Meier methods. There was no significant statistical difference between two study groups.

Administered calories

The mean number of calories administered to patients in the control group was 374.7±71.7 kcal/d and 1286.8±108.3 kcal/d for the PN group. This difference was statistically significant according to the Student's t test (p<0.001).

Discussion

In this study, PN did not prolong survival in patients with terminal cancer. Despite the fact that the PN group received 3.4 times the calories compared to the control group, it was not associated with an increase in overall survival. The observed nonsignificant difference in survival may be rooted in the incompleteness of the study. However, we learned that patients and their families exhibit considerable resistance to withdrawal or withholding of nutritional support even in terminal patients.

Palliative use of PN has long been a controversial issue because malnutrition and cachexia are common and important issues. Cachexia is a chronic wasting syndrome that involves multiple cytokines and abnormal metabolism, which results in consequent uncoupling of tissue anabolism from nutrients.^6,7 Therefore, Western guidelines recommend against provision of nutrients by the parenteral route in patients receiving palliative care.^1–3 These recommendations are based on retrospective studies from the 1980s, and these study results are summarized in an outstanding meta-analysis.⁸ In that report, PN had detrimental effects, such as an increased infection rate (odds ratio 4.1) and inferior survival (relative risk 0.81). The increased risk of infection persisted even when catheter-related sepsis was excluded, suggesting that PN itself may increase susceptibility to infection. Several recent prospective studies demonstrated that PN may actually provide a survival benefit (Table 3) in patients with advanced cancer patients,^9–11 but the patients included in these studies exhibited relatively longer survival times than the patients in our study, implying that PN may be beneficial only in patients who have survival expectancy of more than 3–4 months.

Table 3.

Prospective Studies Describing (home) Parenteral Nutrition in Patients with Advanced Patients

Citation	Design	Comparator	n	Intervention	Results	Comments
Bozzetti et al. (2013)¹²	Prospective uncontrolled	NoneSingle arm	69	Home PN; advanced cancer	Median duration of home PN was 4 months.	No survival data
Lundholm et al. (2004)¹⁰	Prospectiverandomized controlled	Prospective cohort	309	Cyclooxygenase inhibitor±EN/PN; cachexia	TPN in 50% of patentsEN/PN prolonged survival.Survival>6 months in both groups	No survival data regarding patients receiving TPN
Violante et al. (2005)¹¹	Prospective uncontrolled	NoneObservational study	159	PPN or TPN according to patients' condition	TPN in 98 patientsMedian duration of PN was 81 days.	No survival data
This study	Prospectiverandomized controlled	Prospective cohort	31	Hydration only versus TPN	Survival 8 days in hydration group and 13 days in PN group (not significant difference)	Premature study closure

TPN, total parenteral nutrition; PPN, partial parenteral nutrition (with enteral nutrition); EN, enteral nutrition.

Nutrition support in terminally ill patients with incurable cancer is still widely debated. This issue appears to be not only a medical issue but also a cultural topic, although a specific cultural point of view is not yet universally accepted.^12,13 In Korea, more than 80% of patients and families request provision of PN if the patient cannot eat or drink.^14–16 Therefore, it is worth continuing to investigate the benefit and possible harm of PN despite the recommendations by many authorities against PN for terminally ill patients. Clearly, we require more evidence to identify the most suitable patient candidates for benefit from PN.

This study has several shortcomings. First, poor patient enrollment probably stemmed from inadequate explanation of the advantages and disadvantages of hydration or PN. Second, outcomes other than overall survival and administered calories were not included in the investigation. Third, there is no observation without hydration for use as a control arm, probably as withholding parenteral hydration is perceived as unethical by the general public in Korea.

In summary, nutrition support via the parenteral route proved to neither prolong nor reduce survival of terminally ill patients with advanced cancer in this study. In the future, meticulous education for patients and families must be developed before designing another study to investigate the effects of nutrition in terminally ill patients.

Footnotes

Acknowledgments

This study received 2011 grant of Seoul Medical Center Research Institute.

This study was presented as poster at the MASCC/ISOO International Symposium on Supportive Care in Cancer, June 28–30, 2012, New York, NY.

Author Disclosure Statement

No competing financial interests exist.

References

Arends

, Zuercher

, Dossett

, Fietkau

, Hug

, Schmid

, Shang

, Zander

: Non-surgical oncology—Guidelines on Parenteral Nutrition, Chapter 19. Ger Med Sci, 2009; 7:Doc09.

August

, Huhmann

: A.S.P.E.N. clinical guidelines: Nutrition support therapy during adult anticancer treatment and in hematopoietic cell transplantation. J Parenter Enteral Nutr, 2009; 33:472–500.

Bozzetti

, Arends

, Lundholm

, Micklewright

, Zurcher

, Muscaritoli

: ESPEN Guidelines on Parenteral Nutrition: Non-surgical oncology. Clin Nutr, 2009; 28:445–454.

Kim do

, Lee

, Kim

, Lee

: An evaluation of nutrition support for terminal cancer patients at teaching hospitals in Korea. Cancer Res Treat, 2006; 38:214–217.

Bistrian

, Driscoll

: Enteral and parenteral nutrition therapy In: Fauci

, Braunwald

, Kasper

(eds): Harrison's Principles of Internal Medicine, 17th ed., Vol. 1. New York: McGraw-Hill Companies, Inc., 2008, pp. 455–462.

Haehling

, Anker

: Cachexia as a major underestimated and unmet medical need: Facts and numbers. J Cachexia Sarcopenia Muscle, 2010; 1:1–5.

Kim

, Kim

, Yun

, Kim

, Lee

S-C

, Bae

, Kim

, Lee

, Park

, Won

, Park

, Hong

: Pathophysiological role of hormones and cytokines in cancer cachexia. J Korean Med Sci, 2012; 27:128–134.

McGeer

, Detsky

, O'Rourke

: Parenteral nutrition in cancer patients undergoing chemotherapy: A meta-analysis. Nutrition, 1990; 6:233–240.

Bozzetti

, Cozzaglio

, Biganzoli

, Chiavenna

, De Cicco

, Donati

, Gilli

, Percolla

, Pironi

: Quality of life and length of survival in advanced cancer patients on home parenteral nutrition. Clin Nutr, 2002; 21:281–288.

10.

Lundholm

, Daneryd

, Bosaeus

, Korner

, Lindholm

: Palliative nutritional intervention in addition to cyclooxygenase and erythropoietin treatment for patients with malignant disease: Effects on survival, metabolism, and function. Cancer, 2004; 100:1967–1977.

11.

Violante

, Alfonsi

, Santarpia

, Cillis

, Negro

, De Caprio

, Russo

, Contaldo

, Pasanisi

: Adult home parenteral nutrition: A clinical evaluation after a 3-year experience in a Southern European centre. Eur J Clin Nutr, 2005; 60:58–61.

12.

Bozzetti

: Nutritional support of the oncology patient. Crit Rev Oncol Hematol, 2013; 87:172–200.

13.

Chiu

T-Y

, Hu

W-Y

, Chuang

R-B

, Cheng

Y-R

, Chen

C-Y

, Wakai

: Terminal cancer patients' wishes and influencing factors toward the provision of artificial nutrition and hydration in Taiwan. J Pain Symptom Manage, 2004; 27:206–214.

14.

Kwon

, Im

, Cho

, Yoon

S-J

, Oh

: Most advance directives written by patients with advanced cancer or their proxies request only minimally invasive treatments during end-of-life care. Am J Hosp Palliat Care, 2012; 29:622–626.

15.

Yoon

, Choi

, Hyun

: Current situation on signing advance medical directives and actual life-sustaining treatment given at a university hospital [in Korean]. Korean J Hosp Palliat Care, 2011; 14:91–100.

16.

Yang

, Yong

: Perception of artificial hydration for terminally Ill cancer patients: Patients, families and general public [in Korean]. Korean J Hosp Palliat Care, 2009; 12:220–227.