Introduction
Cancer is the second most prevalent cause of mortality
across all leading causes of death.[] There is an
increasing incidence of malignant tumors worldwide, and a progressively higher frequency of emergency
department (ED) visits by cancer patients due to various
reasons.[] Oncology patients constitute a risk
group across ED patients.[] The incidence of ED visits
ranges from 7% to 12% in cancer patients, and some risk factors for ED visits, such as the black race, male
gender, advanced age, medication non-adherence, and
living alone, have been previously reported.[] Most
cancer patients need hospitalization in the last months
of their lives and their frequency of ED visits increases
due to inadequate provision of palliative care.[] However,
ED crowding is increasingly a problem, and cancer
patients often experience longer ED waiting times
due to their possible multisystem pathologies.[] In
addition to the systemic effects of malignancy, the potential
for metastatic organ involvement may result in
excessive use of hospital resources, such as laboratory
procedures and various imaging tools, during the ED
stay of cancer patients.[]
Most ED admissions or visits of cancer patients
are not due to cancer-related problems; instead, are
attributable to pain, nausea, vomiting, infection, local
mass effect, pleural effusion, ascites, chemotherapy/radiotherapy
side effects, and hematologic or metabolic
disorders.[-] Because patients admitted to the ED
need a multidisciplinary approach and as the oncology,
palliative care, and intensive care units are similarly
overburdened as the ED, it is crucial for the emergency
physician to be aware of the characteristics and risk
factors of cancer patients for effective resource management
and swiftly making a precise diagnosis.
This study was conducted to determine the risk
factors for hospitalization and in-hospital mortality
(IHM) in adult ED patients with cancer.
Methods
This retrospective observational study was approved
by the Non-Interventional Research Ethics Committee,
Gülhane School of Medicine, University of Health
Sciences (approval no. 19/253, June 11, 2019). We reviewed
the patient charts and records in the Hospital
Data Management System to identify patients who presented
to the general ED of the hospital in the period
from October 1, 2016, to April 1, 2019. Patients with a
diagnosis of any type of cancer were eligible for study
inclusion. We excluded patients younger than 18 years,
with missing data, of patients referred to an external
treatment center (Fig. 1).
Fig 1: Flowchart of the study.
For the eligible cancer patients included in this
study, we identified and recorded the following information:
age, gender, triage category, and vital signs at
admission; test results of complete blood count, routine
biochemistry, coagulation panels, cardiac panels,
and blood gas parameters obtained at ED visit; and
whether the patients arrived at the ED by ambulance or underwent cardiopulmonary resuscitation (CPR) in
ED. Triage categories were classified according to the
Turkish Ministry of Health triage system, which comprises
different codes to specify priority to ED patients:
a black code indicates patients with the cardiopulmonary
arrest at the admission; a red code indicates
patients who need emergent resuscitative efforts for
survival; a yellow code specifies patients who require
close observation, multiple resources (e.g., laboratory,
radiological, intravenous intervention), having a high-
-risk medical history (e.g., diabetes, pregnancy, prior
heart attack), or admitted with a critical complaint
(e.g., chest pain, severe abdominal pain), as they may
be stable but at the risk of immediate danger; and the
green code is for patients with stable vital signs (such
as pulse rate, respiratory rate, body temperature, oxygen
saturation, arterial blood pressure), without any
critical complaints, and are not expected to develop a
dangerous medical condition during their ED stay.[]
We stratified our study participants into hospitalized
and non-hospitalized patient subgroups, as well
as a survivor (no-IHM) and non-survivor hospitalized
patient (IHM) groups, based on mortality from the
time of hospitalization until discharge.
Statistical Analysis: Demographic data were presented
as frequencies and percentages. Normally distributed
continuous variables are presented as mean
and standard deviation, whereas non-normally distributed
variables were presented as the median and
interquartile range (IQR). The chi-square test was used
for the comparison of two groups concerning categorical
variables. The Mann-Whitney U test and Student's
t-test were used for pairwise comparisons of continuous
data that did not or did conform to a normal distribution,
respectively. All statistical analyses were conducted
in SPSS 18.0 software. A p-value of <0.05 was
considered statistically significant.
Power Analysis: For an alpha value of 0.05, 80%
power, enrollment ratio of 1, and an expected hospitalization
rate of 50%, we calculated that this study should
include 338 patients.
Results
We identified 781 patients, excluded 288 patients, and
included 493 patients (males 283 [57.4%], females 210
[42.6%]; mean age±SD [range] 59.58±17.07 [18?96]
years) in this study, a study flowchart presents patient
disposition (Fig. 1). There were no significant betweengroup
differences in age and sex between the hospitalized
and non-hospitalized patient groups and the no-
IHM and IHM groups (Table 1). The mean length of
ED stay was 189.50 (IQR: 91?321.25) minutes.
Table 1: Comparison of the demographic data and vital signs at presentation by the hospitalization and in-hospital mortality status
The commonest presenting complaints were pain
(n=187, 37.9%), fever (n=71, 14.4%), nausea/vomiting
(n=56, 11.4%), and dyspnea (n=55, 11.2%). The presenting
complaints were not significantly different in
the hospitalization and IHM groups (Table 2). Moreover,
327 (66.3%) arrived at the ED by ambulance and
166 (33.7%) by their vehicles. Stratification of triage
codes indicated 138 (28.0%) red triage, 350 (71.0%)
yellow triage, and 5 (1.0%) black triage code patients;
no cancer patients were assigned a green triage code.
We detected a statistically significant relationship
between the triage codes and hospitalization rates
(p=0.001, chi-square test), but this was not related to
IHM (p=0.431, chi-square test).
Table 2: Distribution of the patients' complaints
Among the patients who visited the ER, 130 (26.4%)
were hospitalized and 363 (73.6%) were discharged;
moreover, 16 (3.2%) patients were transferred from the
ED to another hospital. The hospitalized patients were
further stratified by hospitalization in the medical oncology
clinic (n=182, 36.9%), intensive care unit (n=18,
3.7%), palliative care clinic (n=2, 0.4%), and other
clinical departments (n=202, 59.0%). Sixteen (3.2%)
patients underwent cardiopulmonary resuscitation recomduring
the ED stay, and IHM occurred in 90 (18.3%)
patients. The identified infections were pneumonia
(n=22, 4.5%), urinary tract infections (n=16, 3.2%),
upper respiratory tract infections (n=6, 1.2%), cellulitis
(n=6, 1.2%), biliary tract infections (cholecystitis; n=1,
0.2%), and pancreatitis (n=1, 0.2%).
We detected significant between-group differences
in the levels of hemoglobin (Hb), aspartate aminotransferase
(AST), alanine aminotransferase (ALT), lactate
dehydrogenase (LDH), C-reactive protein (CRP), lactate,
and prothrombin time (PTT) (Tables 1-4) in the
hospitalized and non-hospitalized groups. Moreover,
we found significant between-group differences in the
pulse rate, diastolic blood pressure, oxygen saturation
(SaO2), white blood cell count (WBC), PTT, and international
normalized ratio (INR), and the levels of glucose,
urea, AST, ALT, LDH, CRP, PTT, troponin, and
lactate (Tables 1-4) in the IHM and no-IHM groups.
Table 3: Comparison of the laboratory parameters by the hospitalization and in-hospital mortality status
Table 4: Comparison of the parameters of coagulation, cardiac functions, and blood gases by the hospitalization and in-hospital mortality status in cancer patients
Discussion
The results of the present study show that emergency
physicians can predict the IHM of cancer patients via
a detailed examination of the vital signs, liver function
tests, and coagulation parameters at the time of ED admission.
The age and sex distribution of cancer patients admitted
to the ED were consistent with the reports in the literature.[-] Given that cancer is commonest at the
advanced age (fifth and sixth decades), our study had a
distribution of cancer patients who visited the ED that
was in agreement with this information. As cancer patients
constitute a risk group among other patients with
ED, they are hospitalized more commonly compared
to the non-cancer patients.[,-] Moreover, the IHM
rate among cancer patients in our study was consistent
with reports in the available literature [,,,],
which supports the more frequent observation of high-
-risk triage categories and more frequent hospitalizations
in our study.
Conventional cancer treatment includes chemotherapy
and radiotherapy. Cancer staging determines
the treatment modality and mortality in these patients
but has limited value for the emergency physician.[]
This is because acute and life-threatening conditions,
such as hematological disorders (e.g., neutropenia,
bleeding diathesis), symptoms due to compression by
primary or metastatic mass, acute pain, predisposition
to thromboembolic events, as well as radiotherapy-
and chemotherapy-induced immunosuppression,
are primarily evaluated during ED procedures.[] In
the present study, these conditions are indicated by an
increase in infective parameters, such as WBC, CRP,
bleeding diathesis (caused by INR elevation), and increased
levels of liver enzymes. Therefore, we recommend that emergency physicians and other clinicians
should evaluate laboratory findings together with the
patients' vital signs (hypotension, tachycardia and hypoxemia).
Cancer patients visited the ED mostly between 12
p.m. and 3 a.m., similar to non-cancer patients, and
experienced significantly longer ED wait times than
non-cancer patients.[,,,] Yates et al. reported no
association between the cause of hospitalization and
length of stay.[] In the present study, we found that the
mean length of ED stay of cancer patients was shorter
than the time reported in the literature.[,] Notably,
ED practices in Turkey are regulated by local legislation,
which requires strict compliance. A decree of
the Emergency Department Legislation, issued by the
Republic of Turkey's Ministry of Health, stipulates patients
with a defined diagnosis should be hospitalized
or discharged from the ED within four hours, whereas
patients without any defined diagnoses must be hospitalized
or discharged from the ED within eight hours.
Therefore, the short ED waiting period identified in the
present study is attributable to the implementation of
regulatory requirements, undertaken to reduce ER wait
time in the respective hospital to prevent non-compliance
to legislation and future potential sanctions.
Inflammatory processes play a role in the development
and progression of cancer and infections and
are reported as the commonest cause (19.4%) of hospitalization
across all identified etiological factors.[,] Wulaningsih et al. reported that high levels of CRP
were associated with mortality in cancer patients, and
are potentially related to chronic inflammation.[]
We found increased levels of inflammatory markers
(WBC counts, CRP levels) in hospitalized patients
and in the IHM group. Thus, infections possibly prolong
the length of hospital stay and increase the risk of
mortality, and infectious processes are simultaneously
involved in tumor progression.
Cancer affects the management of diabetes adversely
in association with increased mortality in cancer
patients.[] In contrast, Boursi et al. reported that
high blood glucose levels are associated with better
prognosis in cancer patients.[] We found low glucose
levels in the IHM group, which is consistent with
the findings of Boursi et al. However, the information
in the literature on the value of glucose levels in predicting
mortality in cancer patients is controversial,
and the exact mechanism underlying these findings
has not been clarified yet.
Yang et al. reported that high levels of BUN and
creatinine were risk factors for mortality in septic cancer patients.[] We found higher urea levels of IHM
patients, indicating that the association between increased
urea levels and death might be secondary to
the evolution of multiorgan failure. Oh et al. reported
that increased ALT and GGT levels were associated
with mortality in patients younger than 60 years.[]
Wulaningsih et al. reported that higher LDH levels
might be associated with mortality in cancer patients
due to chronic hypoxia with excessive energy use in
rapidly proliferating cancer cells.[] In concordance
with these reports, we found higher levels of AST, ALT,
and LDH in both hospitalized and IHM groups than
non-hospitalized and non-IHM subgroups. However,
there are no proposed mechanisms in the literature to
precisely explain this observation. Castle et al. reported
that prolonged PTT or INR values in pancreatectomy
patients were associated with increased 30-day mortality
rates.[] Fischer et al. reported a close relationship
between inflammation and several coagulation parameters,
including INR and PTT levels, and platelet
counts in predicting mortality in patients with ED with
suspected infections.[] These findings may have arise
from dysregulation of anticoagulant and procoagulant
protein release from hepatocytes, endothelium, and
circulating cells.[] Furthermore, the homeostasis of
these systems purportedly shifts to a procoagulant state
during inflammatory processes.[] Our study results
are in alignment with the literature, as prolonged PTT
and INR values potentially are risk factors for hospitalization
and IHM.
Yang et al. reported that lactate levels over >4
mmol/L were predictive of mortality in septic cancer
patients.[] Similarly, we found a high lactate levels
may indicate high risk and can be a predictor for hospitalization
and IHM. Yang et al. reported that cardiac
troponin I and CK-MB are independent risk factors for
mortality in cancer patients, and CK and CK-MB levels,
but not brain natriuretic peptide, were associated
with mortality.[] Therefore, we recommend the use
of troponin levels as a marker for IHM in cancer patients,
provided emergency physicians carefully evaluate
the differential diagnoses of the acute coronary
syndrome in patients with high troponin levels.
Limitations of the Study
This study is retrospective, which is subject to the inherent
risks of bias associated with this research design,
and we could neither evaluate the time between
the end of treatment and ED admission nor the patient
prognosis post-discharge.
Conclusion
Cancer patients undertake a higher number of ED visits.
Despite the chronic disease process and relative rarity
of oncological emergencies, such as superior vena
cava syndrome, emergency physicians should remain
updated on the management of cancer patients. The
pain was one of the commonest reasons for ED visits;
emergency physicians should conduct a thorough differential
diagnosis to identify the pain source and provide
appropriate pain-control strategies for the patient.
Hospitalization and IHM rates are higher in cancer
patients compared to non-cancer patients. Therefore,
complete blood counts, liver and kidney function tests,
coagulation parameters, and lactate levels should be
evaluated in these patients concomitantly.
Peer-review: Externally peer-reviewed.
Conflict of Interest: The authors declare that they have no
conflict of interest.
Ethics Committee Approval: This retrospective observational
study was approved by the Non-Interventional
Research Ethics Committee, Gülhane School of Medicine,
University of Health Sciences (Approval no. 19/253, June 11,
2019).
Financial Support: The authors declared that this study received
no financial support.
Authorship contributions: Concept - S.B., Y.A.A.; Design
- S.B., Y.A.A.; Supervision - S.B., Y.A.A.; Funding - None;
Materials ? None; Data collection and/or processing - S.B.,
Y.A.A.; Data analysis and/or interpretation - S.B., Y.A.A.;
Literature search - S.B., Y.A.A.; Writing - S.B., Y.A.A.; Critical
review - S.B., Y.A.A.
References
Miranda B, Vidal SA, Mello MJ, Lima JT, Rêgo JC,
Pantaleão MC, et al. Cancer patients, emergencies service
and provision of palliative care. Rev Assoc Med
Bras (1992) 2016;62:207-11.
Rivera DR, Gallicchio L, Brown J, Liu B, Kyriacou DN,
Shelburne N. Trends in Adult Cancer-Related Emergency
Department Utilization: An Analysis of Data
From the Nationwide Emergency Department Sample.
JAMA Oncol 2017;3:e172450.
Hsu J, Donnelly JP, Moore JX, Meneses K, Williams
G, Wang HE. National characteristics of Emergency
Department visits by patients with cancer in the
United States. Am J Emerg Med 2018;36:2038-43.
Lash RS, Bell JF, Reed SC, Poghosyan H, Rodgers J,
Kim KK, et al. A Systematic Review of Emergency
Department Use Among Cancer Patients. Cancer
Nurs 2017;40:135-44.
Yang Z, Yang R, Kwak MJ, Qdaisat A, Lin J, Begley
CE, et al. Oncologic emergencies in a cancer center
emergency department and in general emergency
departments countywide and nationwide. PLoS One
2018;13:e0191658.
Ahn S, Lee YS, Lim KS, Lee JL. Emergency department
cancer unit and management of oncologic emergencies:
experience in Asan Medical Center. Support Care
Cancer 2012;20:2205-10.
Sadik M, Ozlem K, Huseyin M, AliAyberk B, Ahmet
S, Ozgur O. Attributes of cancer patients admitted to
the emergency department in one year. World J Emerg
Med 2014;5:85-90.
Yates M, Barrett A. Oncological emergency admissions
to the Norfolk and Norwich University
Hospital: an audit of current arrangements and
patient satisfaction. Clin Oncol (R Coll Radiol)
2009;21:226-33.
Erimşah ME, Yaka E, Yilmaz S, Kama A, Pekdemir
M. Inter-rater reliability and validity of the Ministry
of Health of Turkey's mandatory emergency triage instrument.
Emerg Med Australas 2015;27:210-5.
Amado JP, Vasquez R, Huari RW, Sucari AS, Oscanoa
TJ. Patients with End-stage Oncologic and Nononcologic
Disease in Emergency Service of an Urban Tertiary
Hospital. Indian J Palliat Care 2018;24:25-7.
Nafsi T, Russell R, Reid CM, Rizvi SM. Audit of deaths
less than a week after admission through an emergency
department: how accurate was the ED diagnosis
and were any deaths preventable?. Emerg Med J
2007;24:691-5.
Brown J, Grudzen C, Kyriacou DN, Obermeyer Z,
Quest T, Rivera D, et al. The Emergency Care of Patients
With Cancer: Setting the Research Agenda. Ann
Emerg Med 2016;68:706-11.
Wulaningsih W, Holmberg L, Ng T, Rohrmann S, Van
Hemelrijck M. Serum leptin, C-reactive protein, and
cancer mortality in the NHANES III. Cancer Med
2016;5:120-8.
Hope C, Robertshaw A, Cheung KL, Idris I, English
E. Relationship between HbA1c and cancer in people
with or without diabetes: a systematic review. Diabet
Med 2016;33:1013-25.
Boursi B, Giantonio BJ, Lewis JD, Haynes K, Mamtani
R, Yang YX. Serum glucose and hemoglobin A1C
levels at cancer diagnosis and disease outcome. Eur J
Cancer 2016;59:90-8.
Yang Z, Qdaisat A, Hu Z, Wagar EA, Reyes-Gibby C,
Meng QH, et al. Cardiac Troponin Is a Predictor of
Septic Shock Mortality in Cancer Patients in an Emergency
Department: A Retrospective Cohort Study.
PLoS One 2016;11:e0153492.
Oh CM, Won YJ, Cho H, Lee JK, Park BY, Jun JK, et al.
Alanine aminotransferase and gamma-glutamyl transferase
have different dose-response relationships with
risk of mortality by age. Liver Int 2016;36:126-35.
Wulaningsih W, Holmberg L, Garmo H, Malmstrom
H, Lambe M, Hammar N, et al. Serum lactate dehydrogenase
and survival following cancer diagnosis. Br
J Cancer 2015;113:1389-96.
Castle J, Mazmudar A, Bentrem D. Preoperative
coagulation abnormalities as a risk factor for adverse
events after pancreas surgery. J Surg Oncol
2018;117:1305-11.
Fischer CM, Yano K, Aird WC, Shapiro NI. Abnormal
coagulation tests obtained in the emergency department
are associated with mortality in patients with
suspected infection. J Emerg Med 2012;42:127-32.
Aird WC. The hematologic system as a marker of organ
dysfunction in sepsis. Mayo Clin Proc 2003;78:869-81.
Esmon CT. The interactions between inflammation
and coagulation. Br J Haematol 2005;131:417-30.
