Prevalence and Implications of Abnormal Laboratory Results in Patients in the Terminal Phase of Life

Abstract

Background:

Pathophysiological changes at the end of life may affect pharmacokinetics of drugs. However, caregivers typically do not extensively monitor patients' laboratory parameters at the end of life.

Objective:

Our aim was to describe laboratory parameters of hospice patients in the week before death.

Methods:

A cohort study was conducted on available laboratory results in the week before death, including clinical chemistry and hematology tests.

Results:

Laboratory data of 125 patients in a palliative care center were included, assessed at a median of 3 days before death. Eighty percent of patients had anemia and almost all had hypoalbuminemia (97%). Elevated levels of gamma-glutamyl transferase (gGT) were found in 75%, of alkaline phosphatase (ALP) in 60%, of aspartate aminotransferase (ASAT) in 60%, and of calcium (Ca) in 68%. Alanine aminotransferase (ALAT), bilirubin, sodium (Na), and potassium (K) were abnormal in from 8.8% to 36.0% of patients. A previous unknown poor kidney function was found in 60% of patients. Thirteen patients (22%) with a regular morphine prescription and one patient treated with a non-steroidal anti-inflammatory drug (NSAID) had severe kidney failure.

Conclusions:

Abnormal laboratory results were expected due to the pathophysiological changes that occur during the last phase of life. Remarkably, however, electrolytes (Na and K) were balanced even shortly before death. Estimated glomerular filtration rate (eGFR), reflecting the kidney function, seems the most clinically relevant laboratory parameter, because it may guide drug choice and dosing.

Introduction

In Europe most patients (>90%) who are cared for in a hospice suffer from an advanced malignancy.¹ In contrast, the proportion of cancer patients is notably smaller (40%) in hospices in the United States.² These patients often show a gradual decline in physical health and functional status, like in chronic illness.^3,4 This, combined with a lower oral intake may result in weight loss and cachexia, liver impairment, and kidney failure,^5,6 eventually leading to end-stage organ failure, affecting the heart, kidneys, lung, and gastrointestinal tract.^7,8 The most marked deterioration is clinically seen in the last few days. Several aspects of pharmacokinetics, including the absorption, distribution, metabolism, and elimination of drugs, may then change as well (Table 1). As these pharmacokinetic changes have hardly been systematically described in palliative patients,^7,9,10 they are extrapolated from other patient populations, such as cancer patients,^11,12 elderly patients,^13–19 and intensive care patients.^20–23

Table 1.

Potential Pharmacokinetic Changes at the End of Life

Pharmacokinetic parameter	Pathophysiological change or relevant disease state	Effect
Absorption	Malignancy of digestive organs Intestinal complications (vomiting or ileus) Subcutaneous route of administration	Diminished absorption of drugs
Distribution	Change in body composition Lowered serum albumin (both due to cachexia in advanced malignancies or chronic disease state)	Lowered volume of distribution of lipophilic drugs Greater portion unbound drug in serum
Metabolism	Hepatic disease (primary malignancy or secondary disease, such as metastasis)	Reduced hepatic function Increase in half-life of drugs metabolized by the liver Increased risk for drug-drug interactions
Elimination	Kidney disease (primary or secondary disease)	Decrease of GFR Increase in half-life of renally eliminated drugs

GFR, glomerular filtration rate.

The pharmacokinetic changes can result in a prolonged drug effect and accumulation of drugs. To prevent side effects it may be necessary to reduce the doses of sedatives and analgesics, which are often prescribed at the end of life,²⁴ or to choose another drug. The latter is recommended by the European Association of Palliative Care (EAPC)²⁵ and the Dutch palliative pain guideline²⁶ for the use of opioids in palliative patients with renal failure; both recommend to prescribe another opioid, such as fentanyl, buprenorphine, or hydromorphone, instead of morphine in patients with renal failure. The latter guideline proposes to adjust drug doses in the case of liver or kidney failure²⁶ to avoid a negative impact on the quality of dying.

Only a few studies describe actual laboratory results during the last month^27,28 or last week^29–31 of life. However, the primary aim of these studies was not to describe laboratory results in relation to pharmacokinetics, and in most cases only a few parameters were assessed, such as sodium, potassium, and albumin. Therefore, we aimed to describe the results of a more complete panel of laboratory results of hospice patients in the week before death. Further, we evaluated the relevance of the laboratory parameters for the pharmacokinetics of drugs often used in patients at the end of life. As pronounced pathophysiological changes occur prior to death, we hypothesized, that laboratory results of patients shortly before death maybe quite abnormal with consequences for the choice of drugs and the dosing of sedatives and analgesics.

Methods

Design and setting

This observational cohort study was performed in Laurens Cadenza in Rotterdam, the Netherlands. This is the largest palliative care center in the Netherlands, with 20 beds for end-of-life care and symptom management; 200 to 250 patients are admitted annually. A multi-disciplinary team of health care professionals, including specialized nurses and elderly care physicians specialized in palliative care, is available 24 hours per day. Dutch palliative guidelines form the core for clinical practice, including the abstinence, in principle, of artificial fluids or nutrition in the last few days of life.²⁶

Data collection

All available laboratory results, including clinical chemistry and hematology tests, from April 2009 to September 2013, were extracted from the electronic medical records. Subsequently, only the last results of patients in their last week before death were selected. All patients died in the palliative care center. There is no policy for routine testing at admission or during the admission. In practice, laboratory testing, including clinical chemistry and hematology tests, is often done in the first 1 or 2 weeks after admission and as follow-up testing if needed. In a number of patients (48/125, 38.4%) laboratory tests were requested in the context of a drug study, and in the other patients laboratory testing was done at the physician's discretion and for clinical purposes only. Common indications for laboratory testing were monitoring a patient's health status, guiding medication use, finding causes of symptoms, or guiding symptom management, such as weighing the need for a blood transfusion in the case of severe anemia.

Laboratory parameters

The following serum parameters were available for the majority of patients in their final week: hemoglobin (Hb), alanine aminotransferase (ALAT), aspartate aminotransferase (ASAT), gamma-glutamyl transferase (gGT), alkaline phosphatase (ALP), total bilirubin (Bili), albumin, sodium (Na), potassium (K), calcium (Ca), and creatinine (from which the estimated glomerular filtration rate [eGFR]was calculated). The reference values of the hospital laboratory are given in Table 2.

Table 2.

Laboratory Results of Palliative Patients in the Final Week before Death (n = 125), Including the Corresponding Laboratory Reference Values

				Abnormal results
		Laboratory result Median; IQR		N (%)		Laboratory reference values
Parameter	N	Male	Female	Low	High	Male	Female
Hb in mmol/L (in g/dL)	86	6.3; 5.3 to 7.5 (10.1; 8.5 to 12.1)	6.4; 4.8 to 7.6 (10.3; 7.7 to 12.2)	69 (80.2)	-	8.5 to 11.0 (13.7 to 17.7)	7.5 to 10.0 (12.1 to 16.1)
ALAT in U/L	106	18; 10 to 34	18; 8 to 47	-	27 (25.4)	0 to 41	0 to 31
ASAT in U/L	104	39; 23 to 83	36; 28 to 65	-	62 (59.6)	0 to 37	0 to 31
gGT in U/L	108	61; 44 to 145	95; 45 to 343	-	81 (75.0)	0 to 50	0 to 35
ALP in U/L	108	139; 89 to 301		-	65 (60.2)	0 to 120
Bili in umol/L (in mg/dL)	100	11; 7 to 20 (0.6; 0.4 to 1.2)		-	27 (27.0)	0 to 17 (0 to 1.0)
Albumin in g/L	115	23; 20 to 27		112 (97.4)	-	35 to 50
Na in mmol/l	114	137; 133 to 140		41 (36.0)	10 (8.8)	135 to 145
K in mmol/L	113	4.2; 3.7 to 4.7		13 (11.5)	18 (15.9)	3.5 to 5.0
Ca_corr in mmol/L (in mg/dL)	94	2.76; 2.63 to 2.95 (11.0; 10.5 to 11.8)		-	64 (68.1)	2.20 to 2.65 (8.8 to 10.6)
eGFR	115	63; 33 to 94		Displayed in Table 4

For Hb, Bili, and Ca also, the values in U.S. units are given between parentheses.

ALAT, alanine aminotransferase; ALP, alkaline phosphatase; ASAT, aspartate aminotransferase; Bili, total bilirubin; Ca, calcium; Ca_corr, calcium level corrected for albumin level; eGFR, estimated glomerular filtration rate; gGT, gamma-glutamyl transferase; Hb, hemoglobin; IQR, interquartile range; K, potassium; Na, sodium.

The Ca levels were corrected for albumin levels according to the following equation: Ca +0.025 (40-albumin).³² The corrected Ca levels (Ca_corr) are presented in Table 2.

The eGFR was calculated using the Modification of Diet in Renal Disease (MDRD) formula.³³ This MDRD formula includes four variables: serum creatinine, age, ethnicity, and gender and is accurate for estimating eGFR across a wide range of subgroups for eGFR <60 mL/min/1.72 m². An eGFR <30 mL/min is considered to reflect poor kidney function.

Other variables

Age, gender, diagnoses, and duration of admission were extracted from the electronic medical records; the primary diagnoses and the number of comorbidities were evaluated. The primary diagnoses refer to the World Health Organization's (WHO's) International Classification of Diseases, 10th revision (ICD-10 classification) coding for the patient's terminal illness. For each patient pre-existing liver diseases (including primary or secondary liver malignancies, cirrhosis, and hepatitis) and pre-existing kidney diseases (including primary or secondary kidney malignancies) were recorded.

The prescriptions of regular medication were extracted from the pharmacy database, including drug name, dose, frequency, route of administration, and dates of start and discontinuation of the prescription. Drugs were prescribed according to the symptom-specific Dutch national palliative guidelines.²⁶ For this study, data of morphine (administered orally or subcutaneously) and non-steroidal anti-inflammatory drugs (NSAIDs, administered orally or rectally) were analyzed.

Daily morphine doses were calculated taking into account the route of administration, so a subcutaneous equivalent was used for analysis.^26,34 A daily subcutaneous morphine dose <100 mg/24 hours was considered a low-to-moderate dose.^35–37

Statistical analysis

Data were analyzed using descriptive statistics. Data are presented as mean (standard deviation, SD) in case of normally distributed variables and as median (interquartile range = IQR, or minimum-maximum range = range) in case of non-normally distributed variables. Data analyses were performed using IBM SPSS Statistics 20 (IBM Corp., Armonk, NY). A p value of <0.05 (two-sided) was deemed statistically significant.

Ethics

Given the retrospective nature of the study ethical approval was waived according to Dutch law. All patient data were handled and processed in accordance with the recommendations of Good Clinical Practice. For patients participating in a drug study (48/125) permission from the Medical Ethics Review Board of the Erasmus University Medical Center was granted and written informed consent had been obtained from the patients for study purposes, including laboratory testing.

Results

Participants

Laboratory results from 285 patients were available over the given time period. Final week results were available for 125 patients, of which the last (or only) ones had been assessed at a median of 3 days (IQR 1 to 4) before death. These 125 patients' median age was 75 years (IQR 64 to 81 years), 52% were female, and the median length of stay was 13 days (IQR 7 to 34 days). An advanced malignancy, mainly of the respiratory or digestive organs, was the reason for admission in 116 patients (93.6%). A pre-existing liver disease was recorded in 38 patients (30.4%), and a pre-existing kidney disease in 18 patients (14.4%). See further Table 3.

Table 3.

Patient Characteristics

Characteristics	N = 125
Gender in %
Male/female	48/52
Age in years
Median (IQR)	75 (64 to 81)
Duration of admission in days
Median (IQR)	13 (7 to 34)
Assessment days before death
Median (IQR)	3 (1 to 4)
Primary diagnose in N (%)
Neoplasms	117 (93.6)
Respiratory and intra-thoracic organs	32 (27.4)
Digestive organs	29 (24.8)
Lymphoid, hematopoietic, and related tissue	13 (11.1)
Breast	9 (7.7)
Female genital organs	7 (6.0)
Male genital organs	6 (5.1)
Urinary tract	5 (4.3)
Eye, brain, and other parts of central nervous system	4 (3.4)
Ill–defined, secondary, and unspecified sites	4 (3.4)
Other	8 (6.8)
Disease of circulatory system	4 (3.2)
Disease of digestive system	2 (1.6)
Other	2 (1.6)
Pre-existing liver disease in N (%)
Yes	38 (30.4)
Secondary malignancy in liver	33 (86.8)
Primary malignancy in liver	3 (7.9)
Other	2 (5.3)
No or not documented clearly	87 (69.6)
Pre-existing kidney disease in N (%)
Yes	18 (14.4)
Not otherwise specified	8 (44.5)
Primary malignancy in kidney	4 (22.2)
Secondary malignancy in kidney	2 (11.1)
Drug induced kidney disease	2 (11.1)
Other	2 (11.1)
No or not documented clearly	107 (85.6)
Co-morbidities
Median (IQR) number of diagnoses	3 (1 to 4)

IQR, interquartile range.

Laboratory results

In 80.2% of 86 patients the Hb levels were below the reference values. Median Hb levels were 6.3 mmol/L (10.1 g/dL) for men and 6.4 mmol/L (10.3 g/dL) for women. Albumin levels were markedly below the reference values in 97.4% of 115 patients, with a median level of 23 g/L (IQR 20 to 27). The median levels of several liver enzymes were too high for a major proportion of patients: gGT in 75.0% of 108 patients, ALP in 60.2% of 108 patients, and ASAT in 59.6% of 104 patients. The median Ca_corr levels were above the normal range for 68.1% of 94 patients. ALAT, Bili, Na, and K were abnormal in smaller proportions (8.8% to 36.0%) of patients. The median eGFR was 63 mL/min (IQR 33 to 94). See further Table 2.

In 25/115 patients (21.7%) the eGFR was <30 mL/min, indicating poor kidney function. Only 10 of the patients (40.0%) were known to have a pre-existing kidney disease (Table 4). In 12 of the patients (48.0%) an eGFR <30 mL/min was accompanied by an increased Ca_corr (>2.65 mmol/L).

Table 4.

Renal Function in Patients with and without a Pre-existing Kidney Disease

eGFR categories (in mL/min/1.73m²)	Total N = 125 Number (%) of patients	Pre-existing kidney disease N = 18 Number (%) of patients	No knownpre-existing kidney disease N = 107 Number (%) of patients
0–14 (kidney failure)	6 (4.8)	3 (16.7)	3 (2.8)
15–29 (severe kidney disease)	19 (15.2)	7 (38.9)	12 (11.2)
30–59 (moderate kidney disease)	31 (24.8)	6 (33.3)	25 (23.4)
60–89 (mild kidney disease)	27 (21.6)	1 (5.6)	26 (24.3)
≥90 (very mild or no kidney disease)	32 (25.6)	1 (5.6)	31 (29.0)
eGFR was not assessed in the final week	10 (8.0)	–	10 (9.3)

eGFR, estimated glomerular filtration rate.

Levels of liver enzymes (Table 5) were within the normal range in fewer patients with a documented pre-existing liver disease (10.5% to 47.4% of 38 patients) than in patients without that disease (26.4% to 71.3% in 87 patients). Moreover, the median levels of ASAT, ALP, and gGT were distinctly higher in the patients with a pre-existing liver disease.

Table 5.

Results of Liver Test for Patients with and without Pre-existing Liver Disease

		Pre-existing liver disease N = 38		No known pre-existing liver disease N = 87
Liver tests	Total N = 125 Number (%) of patients within normal range	Number (%) of patients within normal range	Laboratory result Median (IQR)	Number (%) of patients within normal range	Laboratory result Median (IQR)
Bili	73/100 (73.0)	18 (47.4)	14 (7 to 53)	55 (63.2)	10 (7 to 15)
ALAT	79/106 (74.5)	17 (44.7)	34 (17 to 79)	62 (71.3)	14 (9 to 27)
ASAT	42/106 (39.6)	6 (15.8)	85 (36 to 273)	36 (41.4)	34 (23 to 45)
ALP	43/108 (39.8)	6 (15.8)	312 (158 to 766)	37 (42.5)	117 (81 to 153)
gGT	27/108 (25.0)	4 (10.5)	315 (95 to 897)	23 (26.4)	56 (32 to 103)

ALAT, alanine aminotransferase; ALP, alkaline phosphatase; ASAT, aspartate aminotransferase; Bili, total bilirubin; gGT, gamma-glutamyl transferase; IQR, interquartile range.

Prescriptions of NSAIDs and morphine

Medication data during the last week of life in combination with an eGFR value were available for 102 of 125 patients (81.6%), for 22 of 25 (88.0%) patients with poor kidney function, and for 80 of 90 (88.9%) patients with an eGFR ≥30 mL/min.

Twelve patients were on regular treatment with oral or rectal NSAIDs in the last week of life. One of six of these patients whose kidney function was checked in the last week had severe renal failure (eGFR 20 mL/min).

Morphine was prescribed on a regular basis via the oral or subcutaneous route for 94 of 102 patients (92.2%). Kidney function was checked in 60 of them (63.8%) and 13 of those (21.7%) had severe renal failure: median eGFR of 21 mL/min (range 5 to 28). A low-to-moderate subcutaneous equivalent morphine dose (<100 mg/day) was administered to 12 of 13 patients (92.3%) with poor kidney function versus 39 of 47 (83.0%) patients with an eGFR ≥30 mL/min. The median morphine daily dose in patients with severe kidney failure was 30 mg (IQR 15 to 60) and not different from the 30 mg (IQR 15 to 75) morphine dose in patients with eGFR ≥30 mL/min.

Discussion

This study found that most patients had anemia (80%) or hypoalbuminemia (97%) in their last week of life and that many (60% to 75%) had elevated levels of gGT, ALP, ASAT, or Ca_corr levels. Notably, levels of ALAT, bilirubin, Na, and K were within the normal range for most patients (abnormal in 8.8% to 36.0%). In 60% of patients whose kidney function proved to be poor (eGFR <30 mL/min) this was not documented in the medical history. Some patients with severe kidney failure were on treatment with an NSAID and morphine (1 and 13 patients, respectively), which is not recommended in palliative treatment guidelines.^25,26

Both lowered albumin and hemoglobin levels are related to chronic disease and are therefore not surprising in this population of mainly oncology patients. Severe anemia (Hb <5.0 mmol/L, 8.0 g/dL)³⁸ was found in 20% of patients. The prevalence of anemia in the present study (80%) is comparable to that in other studies in palliative care patients (72% to 82%).^27,28,39 Differences in case mix or patient population and in timing of laboratory testing (Table 6) will influence the prevalence of anemia. To illustrate this, one study excluded the very ill patients and found anemia in 77% of men and 68% of women.³⁹

Table 6.

Overview of Literature about Laboratory Parameters in Palliative Patients, Displaying Differences in Case Mix

			Patients				Prognosis
Reference	Design	Primary aim	N	Setting	Type	Deceased	Duration of admission	Survival after assessment
Present study	Cohort study	To describe laboratory parameters	125	Palliative care center	94% advanced cancer	125/125	Median 13 days (IQR 7 to 34)	Median 3 days (IQR 1 to 4)
Jiménez-Gordo 2009²⁷	Cohort study	To describe clinical and laboratory parameters	406	Hospital	All advanced cancer	368/406	Median 27 days (range 0 to 547)	–
Morita 2006²⁹	Observational study	To relate hydration volume and laboratory parameters	125	Various (oncology units, palliative care units, home based programs)	All advanced lung or abdominal cancer Note: patients with liver cirrhosis or liver cancer, renal failure, or hypercalcemia were excluded	125/125	–	All assessments were done in the last week of life
Sarhill 2003²⁸	Cohort study	To evaluate nutritional status assessments	352	Various (inpatient units and outpatient clinics)	All advanced cancer Note: actively dying patients were excluded	– Note: Survival was recorded for 53% of patients	–	Median about 1 month (range 1 day to 7 months)
Dunn 2003³⁹	Cohort study	To determine the prevalence and causes of anemia	105	Hospice inpatient unit	91% advanced cancer Note: patients recognized to be near death were excluded	–	13% of patients deceased within 10 days of admission	–
Ellershaw 1995³⁰	Cohort study	To relate symptoms and dehydration	82	Hospice	All advanced cancer	82/82	–	Median 2 days (range 1 to 5 days)
Oliver 1984³¹	Secondary analysis	To study hypercalcemia upon admission	22	Hospice	All advanced cancer	22/22	–	12 patients within 1 day, 10 patients between 1 and 2 days

IQR, interquartile range.

The prevalence of hypoalbuminemia (albumin <35 g/L) differs between studies. In the present study almost all patients had hypoalbuminemia (97%) compared with only two thirds in the referred studies.^27,28 This discrepancy might be explained by a difference in the median time till death (Table 6); 3 days in the present study versus about 1 month in the other studies.^27,28 Moreover, hypoalbuminemia is expected in end-of-life patients because these patients often suffer from edema, ascites, and carcinomatous pleural effusion and are in a cachectic state.²⁸ Some studies suggest that albumin level may help predict the time until death in oncology or palliative patients; the lower the level, the nearer to death.^40–42 However, this prognostic value of albumin levels was not confirmed in all studies; especially in studies testing multiple factors different outcomes are described.^43–45

Albumin is an important protein for the transport of various drugs in plasma.^46,47 Therefore, hypoalbuminemia may lead to a significant decrease of protein binding capacity of drugs and an increase in the unbound drug fraction, which could strengthen the pharmacological effect. The extent of this phenomenon is debated; the theory and clinical relevance may be contradictory. Benet and Hoener concluded that, clinically, for changes in plasma protein binding adjustment of a drug dose is not necessary, except for drugs with a high extraction ratio and a narrow therapeutic index that are given parenterally.⁴⁸ Midazolam and haloperidol are often prescribed in palliative care²⁴ and both are listed in the referred article as drugs in which protein binding may influence clinical drug exposure.⁴⁸ However, the actual consequences in clinical practice remain unknown.

The abnormal liver enzyme test results in the present study do not all tally with those in previous studies.²⁷ Laboratory tests of liver enzymes are notoriously bad indicators for liver function. Unfortunately, the levels of liver enzymes, even if severely elevated, do not predict a change in hepatic function, including drug metabolism.^49,50 Moreover, liver enzymes may be elevated for other reasons than liver disease only, such as polypharmacy, alcohol consumption, metastatic disease, or other organ diseases.⁵¹ Bilirubin levels may reflect bile duct related problems (obstruction), while metabolic liver capacity is normal.

Hypercalcemia was found in 68% of patients in the present study versus 7% in another study.²⁷ This discrepancy might be explained by the median time till death (Table 6), that is, 3 days versus 26.5 days. In 48.0% of patients hypercalcemia was accompanied by kidney failure, possibly due to severe dehydration.

The median eGFR in our study population was 63 mL/min. Because the median age of this population was 75 years, this value reflects the physiological changes in renal function associated with advanced age.⁵² Severe kidney failure, however, was found in 19% of patients. Kidney function impairment may be the result of a variety of underlying causes in cancer patients, including dehydration and extra-renal obstruction.¹¹ Although altered kidney function may be caused by drug treatment, poor renal function in its turn can have pharmacological consequences. This is illustrated by the finding of higher metabolite concentrations in patients with a worse kidney function.^7,53 Those patients with kidney failure might be at risk for opioid-induced side effects, because metabolites will accumulate. In our study most patients with poor kidney function were on treatment with a low-to-moderate morphine daily dose (92%), perhaps because the accumulation had resulted in stronger effects, and the dose needed to be reduced.

Remarkably, most of our patients had normal levels of Na and K. This is also described in other studies during the last month²⁷ and the last week of life,^29–31 indicating that these electrolytes even shortly before death are balanced.

To the best of our knowledge, this is the first overview of laboratory results in the week before death in hospice patients. We were also able to include data on actual morphine and NSAID use. An additional strength of our study is that we related the laboratory results to probable pharmacokinetic effects. Generally, in such a vulnerable patient population, the possibly negative impact of blood taking should be weighed against the benefits of the laboratory results. However, even in end-of-life patients laboratory testing may be used to monitor or guide pharmacological treatment, to assess the need for additional treatment such as blood transfusion or to explain symptoms such as fatigue, confusion, or vomiting. This is illustrated by one study in a palliative care unit that concluded that 96% of the cases of laboratory testing were clinically relevant to confirm or exclude a diagnosis and that 30% resulted in a change of drug therapy.⁵⁴

Several limitations of this study have to be addressed, however. This was a single-center study in patients with mainly oncologic disease, so that findings should be extrapolated with caution. Further, clinical details on diagnoses, mainly liver and kidney disease, could be incomplete. In a palliative care setting the observation of impaired kidney function will not lead to an exhaustive diagnostic work-up to identify the cause. This might have consequences for the classification of patients having a pre-existing kidney or liver disease. In addition, the exact severity of the kidney or liver disease was not detailed and all diseases states were therefore grouped together. Unfortunately, symptoms had not been routinely recorded with validated objective scoring instruments. And then, medication data were not available for all included patients. It would be of added value to relate laboratory results to validated symptom scores and administered medication in future research, because this may give additional insights into the clinical relevance of the abnormal laboratory results.

Conclusion

In conclusion, although clinically relevant abnormal values were found for several laboratory parameters, 5 out of 11 parameters studied were within the normal range for most patient in their last week of life. Anemia and hypoalbuminemia were not unexpected, due to the physiological changes that occur during the last phase of life. Remarkably, electrolytes were balanced even shortly before death. The most clinically relevant laboratory parameter seems to be eGFR, reflecting kidney function, as for some drugs (including morphine and NSAIDs) adverse effects may negatively impact on quality of life and knowledge of eGFR may assist in deciding on drugs and their dosing.

Footnotes

Author Disclosure Statement

This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors. No competing financial interests exist.

References

Potter

, Hami

, Bryan

, et al.: Symptoms in 400 patients referred to palliative care services: prevalence and patterns. Palliat Med, 2003; 17:310–314.

NHPCO's Facts and Figures: Hospice Care in America. 2015. Alexandria, VA: National Hospice and Palliative Care Organization.

Lunney

, Lynn

, Foley

, et al.: Patterns of functional decline at the end of life. JAMA, 2003; 289:2387–2392.

Murray

, Kendall

, Boyd

, et al.: Illness trajectories and palliative care. BMJ, 2005; 330:1007–1011.

Raijmakers

, Van Zuylen

, Costantini

, 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; 22:1478–1486.

Ellershaw

, Ward

: Care of the dying patient: the last hours or days of life. BMJ, 2003; 326:30–34.

Strouse

: Pharmacokinetic drug interactions in palliative care: focus on opioids. J Palliat Med, 2009; 12:1043–1050.

Domeisen Benedetti

, Ostgathe

, Clark

, et al.: International palliative care experts' view on phenomena indicating the last hours and days of life. Support Care Cancer, 2013; 21:1509–1517.

Mihelic

: Pharmacology of palliative medicine. Semin Oncol Nurs, 2005; 21:29–35.

10.

Cheng

, Roemer-Becuwe

, Pereira

: When midazolam fails. J Pain Symptom Manage, 2002; 23:256–265.

11.

Humphreys

, Soiffer

, Magee

: Renal failure associated with cancer and its treatment: an update. J Am Soc Nephrol, 2005; 16:151–161.

12.

Haddad

, Davis

, Lagman

: The pharmacological importance of cytochrome CYP3A4 in the palliation of symptoms: review and recommendations for avoiding adverse drug interactions. Support Care Cancer, 2007; 15:251–257.

13.

Vestal

: Aging and pharmacology. Cancer, 1997; 80:1302–1310.

14.

Crome

: What's different about older people. Toxicology, 2003; 192:49–54.

15.

Mangoni

, Jackson

: Age-related changes in pharmacokinetics and pharmacodynamics: basic principles and practical applications. Br J Clin Pharmacol, 2004; 57:6–14.

16.

Shah

: Drug development and use in the elderly: search for the right dose and dosing regimen (Parts I and II). Br J Clin Pharmacol, 2004; 58:452–469.

17.

Schwartz

: The current state of knowledge on age, sex, and their interactions on clinical pharmacology. Clin Pharmacol Ther, 2007; 82:87–96.

18.

Eldesoky

: Pharmacokinetic-pharmacodynamic crisis in the elderly. Am J Ther, 2007; 14:488–498.

19.

Hilmer

, Mclachlan

, Le Couteur

: Clinical pharmacology in the geriatric patient. Fundam Clin Pharmacol, 2007; 21:217–230.

20.

Power

, Forbes

, Van Heerden

, et al.: Pharmacokinetics of drugs used in critically ill adults. Clin Pharmacokinet, 1998; 34:25–56.

21.

Yogaratnam

, Miller

, Smith

: The effects of liver and renal dysfunction on the pharmacokinetics of sedatives and analgesics in the critically ill patient. Crit Care Nurs Clin North Am, 2005; 17:245–250.

22.

Boucher

, Wood

, Swanson

: Pharmacokinetic changes in critical illness. Crit Care Clin, 2006; 22:255–271, vi.

23.

Roberts

, Hall

: Drug absorption, distribution, metabolism and excretion considerations in critically ill adults. Expert Opin Drug Metab Toxicol, 2013; 9:1067–1084.

24.

Masman

, Van Dijk

, Tibboel

, et al.: Medication use during end-of-life care in a palliative care centre. Int J Clin Pharm, 2015; 37:767–775.

25.

Caraceni

, Hanks

, Kaasa

, et al.: Use of opioid analgesics in the treatment of cancer pain: evidence-based recommendations from the EAPC. Lancet Oncol, 2012; 13:e58–e68.

26.

IKNL. Guidelines for diagnosis and treatment in palliative care. 2010. www.pallialine.nl (Last accessed November 2014 ).

27.

Jimenez-Gordo

, Feliu

, Martinez

, et al.: Descriptive analysis of clinical factors affecting terminally ill cancer patients. Support Care Cancer, 2009; 17:261–269.

28.

Sarhill

, Mahmoud

, Walsh

, et al.: Evaluation of nutritional status in advanced metastatic cancer. Support Care Cancer, 2003; 11:652–659.

29.

Morita

, Hyodo

, Yoshimi

, et al.: Artificial hydration therapy, laboratory findings, and fluid balance in terminally ill patients with abdominal malignancies. J Pain Symptom Manage, 2006; 31:130–139.

30.

Ellershaw

, Sutcliffe

, Saunders

: Dehydration and the dying patient. J Pain Symptom Manage, 1995; 10:192–197.

31.

Oliver

: Terminal dehydration. Lancet, 1984; 2:631.

32.

Payne

, Little

, Williams

, et al.: Interpretation of serum calcium in patients with abnormal serum proteins. Br Med J, 1973; 4:643–646.

33.

Levey

, Coresh

, Greene

, et al.: Expressing the Modification of Diet in Renal Disease Study equation for estimating glomerular filtration rate with standardized serum creatinine values. Clin Chem, 2007; 53:766–772.

34.

Hanks

, Conno

, Cherny

, et al.: Morphine and alternative opioids in cancer pain: the EAPC recommendations. Br J Cancer, 2001; 84:587–593.

35.

Bercovitch

, Adunsky

: Patterns of high-dose morphine use in a home-care hospice service: should we be afraid of it?. Cancer, 2004; 101:1473–1477.

36.

Ensor

, Middlemiss

: Benchmarking opioids in the last 24 hours of life. Intern Med J, 2011; 41:179–185.

37.

Bruera

, Schoeller

, Wenk

, et al.: A prospective multicenter assessment of the Edmonton staging system for cancer pain. J Pain Symptom Manage, 1995; 10:348–355.

38.

WHO: Haemoglobin concentrations for the diagnosis of anaemia and assessment of severity. Vitamin and Mineral Nutrition Information System. Geneva: World Health Organization, 2011.

39.

Dunn

, Carter

: Anemia at the end of life: prevalence, significance, and causes in patients receiving palliative care. J Pain Symptom Manage, 2003; 26:1132–1139.

40.

Vigano

, Bruera

, Jhangri

, et al.: Clinical survival predictors in patients with advanced cancer. Arch Intern Med, 2000; 160:861–868.

41.

Lis

, Grutsch

, Vashi

, et al.: Is serum albumin an independent predictor of survival in patients with breast cancer?. JPEN J Parenter Enteral Nutr, 2003; 27:10–15.

42.

Vigano

, Dorgan

, Buckingham

, et al.: Survival prediction in terminal cancer patients: a systematic review of the medical literature. Palliat Med, 2000; 14:363–374.

43.

Kao

, Chen

, Chiang

, et al.: Predicting factors in the last week of survival in elderly patients with terminal cancer: a prospective study in southern Taiwan. J Formos Med Assoc, 2009; 108:231–239.

44.

Maltoni

, Pirovano

, Nanni

, et al.: Biological indices predictive of survival in 519 Italian terminally ill cancer patients. Italian Multicenter Study Group on Palliative Care. J Pain Symptom Manage, 1997; 13:1–9.

45.

Hyodo

, Morita

, Adachi

, et al.: Development of a predicting tool for survival of terminally ill cancer patients. Jpn J Clin Oncol, 2010; 40:442–448.

46.

Principles of Clinical Pharmacology, 2e. Elsevier Academic Press, San Diego, CA, 2007.

47.

Clinical Pharmacokinetics and Pharmacodymanics: Concepts and Applications, 4e. Lippincott Williams and Wilkins, Philadelphia, PA, 2011.

48.

Benet

, Hoener

: Changes in plasma protein binding have little clinical relevance. Clin Pharmacol Ther, 2002; 71:115–121.

49.

Rodighiero

: Effects of liver disease on pharmacokinetics. An update. Clin Pharmacokinet, 1999; 37:399–431.

50.

Morgan

, Mclean

: Clinical pharmacokinetic and pharmacodynamic considerations in patients with liver disease. An update. Clin Pharmacokinet, 1995; 29:370–391.

51.

Giannini

, Testa

, Savarino

: Liver enzyme alteration: a guide for clinicians. CMAJ, 2005; 172:367–379.

52.

Wetzels

, Kiemeney

, Swinkels

, et al.: Age- and gender-specific reference values of estimated GFR in Caucasians: the Nijmegen Biomedical Study. Kidney Int, 2007; 72:632–637.

53.

Ashby

, Fleming

, Wood

, et al.: Plasma morphine and glucuronide (M3G and M6G) concentrations in hospice inpatients. J Pain Symptom Manage, 1997; 14:157–167.

54.

Mccormack

, Sui

, Conroy

, et al.: The usefulness of phlebotomy in the palliative care setting. J Palliat Med, 2011; 14:297–299.