118
Journal of Pain and Symptom Management
Vol. 50 No. 1 July 2015
Brief Report
Quality of Life, Pain Perception, and Distress Correlated to Ultrasound-Guided Peripherally Inserted Central Venous Catheters in Palliative Care Patients in a Home or Hospice Setting Roberto Bortolussi, MD, Paola Zotti, PhD, Maria Conte, MD, Rita Marson, RN, Jerry Polesel, ScD, Annamaria Colussi, RN, Donatella Piazza, RN, Gianna Tabaro, BSc, and Simon Spazzapan, MD Palliative Care and Pain Therapy Unit (R.B.), Psycho-Oncology Unit (P.Z.), Epidemiology and Biostatistics Department (J.P.), and Clinical Trials Office (A.C., G.T., S.S.), CRO Aviano National Cancer Institute, Aviano; Hospice ‘‘Il Gabbiano’’ (M.C., D.P.), San Vito al Tagliamento; and Hospice Via di Natale ‘‘Franco Gallini’’ (R.M., S.S.), Aviano, Italy
Abstract Context. Intravenous fluid administration with peripherally inserted central venous catheters (PICCs) and midline catheters in palliative care. Objectives. To evaluate distress and pain perceived by patients during the positioning of a PICC or midline catheter, both in the home and hospice settings. Methods. This was a prospective observational study performed by the Palliative Care Network of Pordenone. In addition to evaluating distress and pain, we monitored patient quality of life and the devices used. Quality of life was measured with the European Organization for Research and Treatment of Cancer-Core 15-Palliative scale. Results. From May 2012 to July 2013, 48 patients were enrolled in the study. The level of distress during the procedure was null or very low in 95.8% of the patients and completely absent after one week. Pain during insertion was null or very little in 93.8% of the patients and zero after one week in 98% of the patients. Quality of life was significantly improved after one week for certain specific parameters and also globally. The number of catheter days monitored was 3097. The weekly monitoring of the devices revealed a series of minor complications. Only two catheters were removed for serious complications. Conclusion. Our results showed a low impact on pain and distress, a low level of local and systemic complications and a favorable impact on patients’ quality of life. However, other studies are necessary to evaluate the cost-effectiveness of the use of these devices and their role in palliative care. J Pain Symptom Manage 2015;50:118e123. Ó 2015 American Academy of Hospice and Palliative Medicine. Published by Elsevier Inc. All rights reserved. Key Words Central venous catheters, PICC, palliative care, quality of life
Introduction The role of artificial nutrition and hydration (ANH) in palliative care patients has not clearly been proven. Recent revisions of the literature suggest that, even in the absence of strong evidence of benefit for the majority of patients, in selected cases benefit can be obtained.1,2 Moreover, a moderate level of hydration may prevent delirium in patients with advanced
Address correspondence to: Roberto Bortolussi, MD, Division of Anesthesiology, National Cancer Institute, Centro di Riferimento Oncologico, Via Franco Gallini 2, 33081 Aviano, Italy. E-mail:
[email protected] Ó 2015 American Academy of Hospice and Palliative Medicine. Published by Elsevier Inc. All rights reserved.
cancer, as well as myoclonus, hallucinations, fatigue, and sedation.3 The method of choice for parenteral administration of liquids and drugs in palliative care treatment is often the subcutaneous route because it is simple to administer, does not cause trauma, and has few contraindications; limits include the composition and the volume to be infused. However, there are data that indicate that both patients and caregivers consider intravenous hydration to be more
Accepted for publication: February 14, 2015.
0885-3924/$ - see front matter http://dx.doi.org/10.1016/j.jpainsymman.2015.02.027
Vol. 50 No. 1 July 2015
PICC Insertion for Palliative Care in the Home Setting
efficacious.4 A disadvantage of this method is the difficulty in finding venous access, especially in patients with an advanced stage of disease. However, in the last few years, peripherally inserted central venous catheter devices (PICCs) have become available. Other, shorter midline catheters, which have their apex in the axillary vein, are even easier to position and are compatible with isotonic solutions. In preliminary studies, PICCs were considered safe, useful, and comfortable for terminal cancer patients, with a low percentage of complications both during placement and thereafter.5 The gold standard for PICC placement is through an ultrasound-guided technique. The goal of this study was to evaluate distress and pain perceived by patients during the positioning of a PICC or midline catheter, both in the home and hospice settings.
Methods After approval by the Ethics Committee of the Aviano National Cancer Institute, Aviano, Italy, a prospective, observational study was performed within the Palliative Care Network of Pordenone between May 2012 and July 2013. This corresponds to a geographical area with a population of 315,000 and includes a home care service and two hospices. Each patient was evaluated for diagnosis, performance status, the presence or absence of a permanent vascular access, and the indications for a program of hydration, parenteral nutrition, or drug administration via the intravenous route. Informed consent was then obtained, and patients were assessed for feasibility by a palliative care physician expert in the methodology of the procedure. For patients who needed a PICC, chest radiography or an intracavitary electrocardiogram control was subsequently performed to confirm the correct positioning of the catheter tip. The primary objectives of the study were 1) to evaluate the level of distress and pain correlated to the placement of the catheter via an ultrasound-guided method and 2) to evaluate the impact of the device placement on the level of distress and pain and on the quality of life of the patients one week after placement. The secondary objectives were 1) to record the time needed to perform the procedure and 2) to monitor any complications correlated to the use of PICC. Patients who satisfied the following eligibility criteria were consecutively enrolled in the study: 1) diagnosis of oncologic disease or degenerative nononcologic disease who were cared for by the Palliative Care team; 2) life expectancy of two weeks or more; 3) necessity of a PICC or midline catheter either in the hospice or home setting; 4) age 18 years or older; 5) adequate cognitive function to be able to
119
understand the procedure and to complete the questionnaires; and 6) written informed consent. The day before the procedure, a baseline assessment of the patient’s quality of life was performed using the European Organization for Research and Treatment of Cancer Quality of Life Questionnaire Core15-Palliative (EORTC QLQ-C15-PAL),6 and a week later, the same questionnaire was readministered to the patient. Within one hour after the procedure, the patient was asked to rate the level of distress and pain perceived using a four-point Likert scale (rated as: not at all, a little, quite a lot, very much). The patient interviews were conducted by one of the clinical nurses of the multidisciplinary team, all of whom, in the preliminary phase before study initiation, received specific training on how to conduct the interviews by the physician and the psychologist responsible for the study. All the data of each case were collected by the same clinical nurse. The physician who positioned the device completed a form with the details of the cannulated vein, the type of device used, and the time spent on performing the procedure. After seven days, the patient was asked to self-evaluate, through separate interviews, the level of distress and pain and quality of life correlated to the procedure. Each week, the team nurse who changed the adhesive dressing and flushed the catheter, monitored the state of the skin, the functionality of the venous access, and any eventual local or systemic complications.
Assessment Tools Distress and Pain Interviews. To measure the perception of distress and pain correlated to the application and use of the intravenous device, two structured interviews with specific questions were conducted by a trained clinical nurse, which the patient answered using a four-point Likert scale. Quality of Life. The validated Italian version of the EORTC QLQ-C15-PAL questionnaire was used.6 Answers were given on a four-point Likert scale. PICC Monitoring Form. This was completed weekly by a team nurse. It graded the state of the dressing and of the skin, the catheter position related to the venipuncture site and signaled any consequent systemic infection correlated to PICC, defined as the presence of body temperature $39 C during the use of the catheter or of persistent body temperature $38 C in the absence of other evident sources of infection.
Statistical Analysis Sample size was estimated using the optimal twostage design by the Simon method.7 With an expectancy p1-p0 ¼ 15% between the percentage of patients who express distress as null or very little in
120
Bortolussi et al.
the standard positioning (p0 ¼ 75%) and that of the ultrasound-guided positioning (p1 ¼ 90%), while fixing the probability error a ¼ 0.05 and b ¼ 0.20, the number of patients needed in the first stage was 13. The study would have terminated if, at the first stage less than 10 patients reported their distress level as null or very little. None of these patients reported pain or distress. The study then proceeded until 48 patients were enrolled. To evaluate the impact of PICC positioning on quality of life after one week using the EORTC QLQ-C15-PAL, the different scales of the questionnaire were analyzed using the procedure recommended by EORTC. The rough score was transformed to a linear scale from 0 to 100. For each scale, the impact of PICC positioning was evaluated in terms of the difference in the scores before and after the positioning of the device via the t-test for coupled samples (two-tailed).
Results Between May 2012 and July 2013, 371 new consecutive patients were assessed for enrollment in the study. Of these, 48 (12.9%) fulfilled the eligibility criteria and gave their informed consent to participate in the study. Fifty other patients who underwent ultrasound-guided PICC or midline catheter placement were excluded because they did not meet all the inclusion criteria. Among the 48 enrolled patients, the success rate of the procedure was 100%. Among the patients who were not enrolled, seven PICCs and 42 midline catheters were successfully positioned. In only one patient did the procedure fail and was not repeated (99.01% success rate). The indications for positioning the PICCs were hydration and/or parenteral nutrition in 37 cases, drug infusion in 38 cases, and blood sampling in 13 cases. The characteristics of the 48 patients who were enrolled in the study are described in Table 1. Fifteen catheters (31.3%) were inserted in a hospice, and the remaining 33 (68.8%) in the home setting. The position of the atriocaval junction of the apex of PICC was controlled in one case via an intracavitary electrocardiogram, whereas in all the other cases, a chest X-ray was performed. The time needed to insert PICC, measured from the preparation of the sterilized operative area until the application of the stat-lock dressing, was between 13 and 22 minutes (median time 16 minutes). Less time was needed to position the midline catheters (range 8e33 minutes; median 11 minutes). No differences in time were seen in relation to hospice vs. home setting (Table 2). The number of days of catheter permanence for PICCs was 13e462 days (median 102 days), whereas that of the midline catheters was 8e231 days (median 50.5
Vol. 50 No. 1 July 2015
Table 1 Baseline Characteristics of 48 Patients Undergoing Catheter Positioning Characteristics
n (%)
Gender Men Women Age (yrs) <65 65e74 $75 Disease Pancreatic cancer Stomach cancer Other miscellaneous cancer Non-neoplastic diseases Performance status (Karnofsky) 30 40 50 60 Type of device PICC Midline catheter Palliative care setting Home Hospice
30 (62.5) 18 (37.5) 9 (18.8) 21 (43.8) 18 (37.5) 8 8 30 2
(16.7) (16.7) (62.5) (4.2)
5 17 12 14
(10.4) (35.4) (25.0) (30.2)
13 (27.1) 35 (72.9) 33 (68.8) 15 (31.3)
PICC ¼ peripherally inserted central venous catheter.
days) for a total of 3097 catheter days. On the date of study closure (September 30, 2013), seven patients were still alive (two PICCs and five midline catheters). Complications observed during the weekly followup after device placement were 1) Secretion of blood serum from the catheter insertion point: one PICC at Day 7; one PICC-midline at Day 10; 2) Inflammation (reddening without secretion of serum or pus) at the catheter insertion point: two midline catheters at Day 8 and at 6.5 months; 3) Skin reaction to the stat-lock adhesive dressing: one PICC-midline at 71 days; 4) Catheter displacement: one PICC-midline at three months. Table 2 Characteristics of Positioning According to Device Characteristics Duration (min) Median (min-max) Vein Basilic Brachial Cephalic Side Right Left
PICC
Midline Catheter
All
n (%)
n (%)
n (%)
16 (13e22)
11 (8e33)
12 (8e33)
12 (92.3) 1 (7.7) 0 (0.0)
30 (85.7) 5 (14.3) 0 (0.0)
42 (87.5) 6 (12.5) 0 (0.0)
10 (76.9) 3 (23.1)
18 (51.4) 17 (48.6)
28 (58.3) 20 (41.7)
Vol. 50 No. 1 July 2015
PICC Insertion for Palliative Care in the Home Setting
The patients’ levels of distress and pain during the procedure and at follow-up after one week are described in Table 3. Most patients (95.8%) reported no or very little distress during the procedure, whereas after one week the level of distress was none or very little in 100% of the patients. Pain perceived during the procedure was zero or very little in 93.8% of the patients and zero or very little in 98% of the patients at the one-week follow-up. The results of the EORTC QLQ-C15-PAL functional and symptom scales are described in Table 4. A significant improvement of the global quality of life was reported. The functional scales showed an improvement in physical functioning but a worsening of the emotional state of the patients, but without any statistical significance. An improvement in all the physical symptoms was seen but was statistically significant only for nausea. No significant results were revealed in global quality of life, symptoms, and function in correlation to sex, age, home or hospice setting, and type of device used (Table 5).
Table 3 Perception of Distress and Pain During PICC Positioning and at Follow-Up Questionnaire Items
% (95% CI)
n
Was the PICC positioning a distressing experience? No 40 83.3 A little 6 12.5 Quite a lot 1 2.1 Very much 1 2.1 Did you experience pain during PICC positioning? No 37 77.1 A little 8 16.7 Quite a lot 3 6.3 Very much 0 0.0 Distress at follow-up No 41 85.4 A little 7 14.6 Quite a lot 0 0.0 Very much 0 0.0 Pain at follow-up No 45 93.8 A little 2 4.2 Quite a lot 1 2.1 Very much 0 0.0
121
(72.8e93.9) (3.1e21.9) (0.0e6.1) (0.0e6.1) (65.2e89.0) (6.1e27.2) (0.0e13.1) (d) (75.4e95.4) (4.6e24.6) (d) (d) (86.9e100) (0.0e9.8) (0.0e6.1) (d)
All the above complications were resolved and allowed the subsequent normal use of the catheter.
Discussion To our knowledge, this is the only study that has investigated the impact of PICCs in a palliative care patient population since the study of Yamada.5 The decision to initiate a program of ANH within a palliative care program is one that needs to be carefully discussed by the multidisciplinary team, especially because of the lack of clear evidence on the efficacy of this type of treatment. Over the years, a series of guidelines have been produced that may be of help to the team when deciding what is best for the patient.8,9 A recent review on the attitudes toward ANH at the end of life reported that one-third of public opinion considers that this increases the patients’
5) Correlated systemic infections: a suspected infection of one PICC at Day 19, treated systemically and resolved; the same patient presented signs of another suspected infection at Day 188, and blood cultures revealed a contamination by Serratia marcencens and Candida parapsilosis. PICC was subsequently removed and not replaced; 6) Catheter obstruction of one midline catheter at 102 days, which was resolved and then re-occurred on Day 139 in the same patient; the catheter was removed and subsequently replaced.
Table 4 EORTC QLQ-C15-PAL Functional and Symptom Scales at PICC Positioning (T0) and After One Week (T1) Mean Scale Value Scales (range)
n
T0
T1
Difference T1eT0
Global health status (0e100) Functional scales Physical functioning (3e12) Emotional functioning (2e8) Symptom scales Fatigue (2e8) Nausea (1e4) Pain (2e8) Dyspnea (1e4) Insomnia (1e4) Appetite loss (1e4) Constipation (1e4)
44
29.92
41.29
þ11.36
0.02
46 46
8.39 4.46
8.85 4.15
þ0.46 0.30
0.24 0.21
46 47 46 44 46 45 46
6.26 1.87 6.26 1.95 2.37 2.67 2.52
5.85 1.45 5.85 1.84 2.09 2.29 2.20
0.41 0.43 0.41 0.11 0.28 0.38 0.33
0.10 <0.01 0.10 0.40 0.12 0.06 0.06
EORTC QLQ-C15-PAL ¼ European Organization for Research and Treatment of Cancer-Core 15-Palliative scale. a t-Test for paired samples (two-tailed).
Pa
0.41 0.51 0.41 0.09 0.33 0.49 0.19 0.24 0.35 0.24 þ0.08 0.12 0.25 0.12
þ0.59 0.29
0.58 0.41 0.58 0.26 0.44 0.41 0.29
þ0.14 0.33
0.07 0.47 0.07 þ0.23 þ0.07 0.31 0.40
0.06 0.47
1.00 0.11 1.00 0.45 0.59 0.49 0.23
þ11.46
0.47 0.53
0.41 0.71 0.41 0.25 0.41 0.56 0.06
þ11.31
þ1.00 0.17
0.41 0.27 0.41 0.04 0.21 0.28 0.48
Global health status Functional scales Physical Emotional Symptom scales Fatigue Nausea Pain Dyspnea Insomnia Loss of appetite Constipation
EORTC QLQ-C15-PAL ¼ European Organization for Research and Treatment of Cancer-Core 15-Palliative scale; PICC ¼ peripherally inserted central venous catheter. a Analysis of variance for repeated measures.
1.07 0.73 1.07 þ0.29 0.43 0.64 0.67 0.10 0.28 0.10 0.30 0.22 0.26 0.16 0.61 0.37 0.61 0.89 0.86 0.18 0.04 0.33 0.50 0.33 0.13 0.30 0.53 0.55 0.62 0.23 0.62 0.08 0.23 þ0.09 þ0.23 0.34 0.83 0.34 0.10 0.20 0.82 0.77
þ13.33 þ7.14 0.99
P
$70 <70 P
Women Men Scales
Vol. 50 No. 1 July 2015
level of comfort; the majority also thought that patients can derive physical and psychological benefits from hydration therapy. These opinions are less prevalent among health care staff.10 The opportunity for ANH, however, narrows toward the final stages of life. A multicenter study on this issue concluded that hydrating patients in the last two weeks of life did not improve quality of life or symptoms correlated to dehydration.11
0.07 0.11 0.07 0.04 0.60 0.37 0.17
0.62 0.52 0.62 0.39 0.48 0.52 0.57
0.42 0.23 þ0.73 0.04 þ0.10 0.62 þ0.32 0.21 þ0.83 0.54 0.60 0.93
0.57 0.54
þ0.50 0.19
þ0.36 0.53
0.87 0.51
þ9.33 þ14.04 þ17.78 þ8.05 þ15.59 þ1.28 0.52
0.15
P
Hospice Home Midline PICC
P
0.31
50þ <50
Performance Status
a
Setting
a
Device
a
Age (yrs)
a
Gender
Table 5 EORTC QLQ-C15-PAL Functional and Symptom Scales at PICC Positioning (T0) and After One Week (T1) According to Gender, Age, Device Type, Setting, and Performance Status
0.61
Bortolussi et al.
Pa
122
Another important point, in our opinion, that underlines the significance of the availability of these new, low impact techniques, is the indication for intravenous drug infusion, requested for 38 patients, 79.1% of our cases. A daily dose of hundreds of milligrams of morphine chlorhydrate or an equivalent opioid is often needed to obtain pain relief. The opioid may be diluted in the same solution with other drugs. The stability of the solution is correlated to the drug interactions and to the dilution volume.12 If a large total volume is necessary for an adequate dilution, the hourly infusions may be elevated and may not be compatible with a comfortable infusion via the subcutaneous route. For 13 patients (27%), the request to position a PICC also was correlated to the need to perform blood draws. This is a practice that is usually not recommended when using these devices because of the increased risk of catheter occlusion, but by taking suitable precautions such as repeated catheter flushing after the blood draw, one can avoid the stress and pain caused by venipuncture. The positioning of a PICC or midline catheter is undoubtedly more costly when compared with that of placing a Teflon catheter in a peripheral vein. The estimated cost could be as high as 10e20 times more. However, one must remember that the duration of a Teflon venous catheter is three to four days, whereas that of a PICC or midline catheter is a few months. In addition, the procedures to medicate and maintain these catheters are easy and quick for the health care professional and comfortable for the patient who does not have to endure frequent venipunctures. The low incidence of complications observed during the monitoring period suggest, to date, the validity of these devices. Removal of the device was necessary in only two cases, after 188 and 139 days, respectively, and were the only serious complications observed in 3097 catheter days. The ultrasound-guided technique in our study guaranteed a very high percentage of success (99.01%). It also reduced the time needed to position the catheter, thereby guaranteeing a high level of comfort for the patient. The feasibility of the procedure, even in the home setting, was confirmed by the overlap of the median times of the procedure and the low incidence of distress and pain seen in the hospice setting. We did not perform post-mortem cultures of
Vol. 50 No. 1 July 2015
PICC Insertion for Palliative Care in the Home Setting
the catheter tips as in Yamada’s study because we felt that enough safety data relative to the use of PICCs are available.13,14 The choice between PICC and midline catheters (Yamada’s group used only PICCs) fell mostly on the latter (72.9%) in our cases. These devices are more manageable, equally stable over time, more economical, and quicker to place. All the catheters were positioned by expert palliative care physicians, but it is known that the procedure also can be performed by nurses who have undergone adequate training. Once the indications for the procedure have been identified, it is up to the professional to evaluate the feasibility conditions, which must aim at the highest level of comfort for the patient. Last but not least, we feel that the results of an improvement in quality of life of the patient after one week need to be interpreted cautiously; this could have been influenced by the simultaneous start of palliative care treatment. We also are aware of the small sample size (48 patients) and the lack of comparators, which do not allow us to draw any definitive conclusions in terms of change in quality of life. The sample size was originally determined to test the percentage of patients who express no or very little distress during PICC positioning. Therefore, the sample size may not be large enough for the secondary endpoints, that is, to evaluate the changes over time in quality of life subscales (EORTC QLQ-C15-PAL). Information bias also may have occurred. However, interviewers were centrally trained and the EORTC QLQ C15-PAL questionnaire was previously validated. We would like to emphasize that the central venous catheter was positioned in the home setting in 33 patients. This is the first study, to our knowledge, which has demonstrated the safety of this procedure in the natural palliative care setting, that is, the patient’s home.
Conclusions The encouraging results concerning the low level of stress and pain at the time of placement and during use of the PICCs and midline catheters, together with a low incidence of complications both in the hospice and the home settings, are, in our opinion, useful indicators to consider these devices relevant for palliative care patients who need venous access. It is also very important that, in relation to the unit costs of PICC and midline catheters and of the apparatus necessary for their placement, further research be done to study the cost-effectiveness of these devices and the best role they may have in different palliative care settings.
Disclosures and Acknowledgments This study was made possible thanks to the donation of a dedicated ultrasound apparatus by the Association
123
‘‘Amici dell’Hospice’’ of S. Vito al Tagliamento and to the participation of the medical and nursing personnel of the Pordenonese Palliative Care Network: Giorgio Simon, Tiziana Danielis, Graziella Ricci, Cinzia Cozzi, Antonella Rovedo, Giuliana Tassan Got, Marta Zanutel, and Marta Deforni.
References 1. Dev R, Dalal S, Bruera E. Is there a role for parenteral nutrition or hydration at the end of life? Curr Opin Support Palliat Care 2012;6:365e370. 2. Prevost V, Grach MC. Nutritional support and quality of life in cancer patients undergoing palliative care. Eur J Cancer Care 2012;21:581e590. 3. Lawlor PG. Delirium and dehydration: some fluid for thought? Support Care Cancer 2002;10:445e454. 4. Mercadante S, Ferrera P, Girelli D, Casuccio A. Patients and relatives’ perception about intravenous and subcutaneous hydration. J Pain Symptom Manage 2005;30:354e358. 5. Yamada R, Morita T, Yashiro E, et al. Patient reported usefulness of peripherally inserted central venous catheters in terminally ill cancer patients. J Pain Symptom Manage 2010;40:60e66. 6. Groenvold M, Petersen MA, Aaronson NK, et al. EORTC Quality of Life Group. The development of the EORTC QLQ-C15-PAL: a shortened questionnaire for cancer patients in palliative care. Eur J Cancer 2006;42:55e64. 7. Simon R. Optimal two-stage designs for phase II clinical trials. Control Clin Trials 1989;10:1e10. 8. Amadori D, Bruera E, Cozzaglio L, et al. Guidelines on artificial nutrition versus hydration in terminal cancer patients. Nutrition 1996;12:163e167. 9. National Council of Hospice and Specialist Palliative Care Services. Artificial hydration for people who are terminally ill. Eur J Palliat Care 1997;4:124. 10. Raijmakers NJ, Fradsham S, van Zuelen L, et al. OPCARE9. Variation in attitude towards artificial hydration at the end of life: a systematic literature review. Curr Opin Support Palliat Care 2011;5:265e272. 11. Bruera E, Dalal S, Torres-Vigil I, et al. Parenteral hydration in patients with advanced cancer: a multicenter, double blind, placebo-controlled randomized trial. J Clin Oncol 2013;31:111e118. 12. Destro M, Ottolini L, Vicentini L, Boschetti S. Physical compatibility of binary and ternary mixtures of morphine and methadone with other drugs for parenteral administration in palliative care. Support Care Cancer 2012;20:2501e2509. 13. Skiest DJ, Abbot M, Keiser P. Peripherally inserted central catheters in patients with AIDS are associated with a low infection rate. Clin Infect Dis 2000;30:946e952. 14. Cotogni P, Pittiruti M, Barbero C, Monge T, Palmo A, Boggio Bertinet D. Catheter related complications in cancer patients on home parenteral nutrition: a prospective study over 51000 catheter days. JPEN J Parenter Enteral Nutr 2013;37:375e383.