Support Care Cancer (2013) 21:1871–1878
DOI 10.1007/s00520-013-1734-6
ORIGINAL ARTICLE
The frequency, cost, and clinical outcomes of hypernatremia
in patients hospitalized to a comprehensive cancer center
Abdulla K. Salahudeen & Simit M. Doshi & Pankaj Shah
Received: 25 August 2012 / Accepted: 28 January 2013 / Published online: 13 February 2013
# Springer-Verlag Berlin Heidelberg 2013
Abstract
Purpose To study the frequency of hypernatremia in hospi-
talized cancer patients and its impact on clinical outcomes
and healthcare cost.
Methods Cross-sectional analysis of data obtained from
patients admitted to the University of Texas M. D. Anderson
Cancer Center over a 3-month period in 2006. The clinical
outcomes and hospital costs were compared among hyperna-
tremics, eunatremics, and hyponatremics (serum sodium val-
ues include >147, 135–147, and <135 mEq/L, respectively).
Results Of 3,446 patients with at least one serum sodium
value, 51.4 % were eunatremic, 46.0 % hyponatremic, and
2.6 % hypernatremic with most of the hypernatremia (90 %)
acquired during hospital stay. The multivariate hazard ratio
(HR) for mortality in hypernatremic was 5-fold higher than
eunatremic (HR for 90 days—5.09 (95 % CI, 3.32–7.81);
p<0·01) and over 2-fold higher than hyponatremic (HR for
90 days—2.79 (95 % CI, 1.91–4.11), p<0.01). The length of
hospital stay in hypernatremic was 2-fold higher than in
hyponatremic and 4-fold higher than in eunatremic (e.g.,
27±22 days in hypernatremic vs. 6±5 days in eunatremic;
mean ± SD, p<0.01). The hospital bill was higher for hyper-
natremic compared with the rest of the groups (46 % over
eunatremic and 37 % over hyponatremic, p<0.01 for both).
A. K. Salahudeen : S. M. Doshi
Division of Internal Medicine, UT MD Anderson Cancer Center,
Houston, TX, USA
P. Shah
Division of Endocrinology, Mayo Clinic, Rochester, MN, USA
A. K. Salahudeen (*)
Department of General Internal Medicine, University of Texas MD
Anderson Cancer Center, 1400 Pressler Street, Unit 1465,
FCT13.5050,
Houston, TX 77030, USA
e-mail: aksalahudeen@mdanderson.org
Conclusions Although hypernatremia was far less frequent
than hyponatremia in the hospitalized cancer patients, most
hypernatremia were acquired in the hospital and had sub-
stantially higher mortality, hospital stay, and hospital bills
than eunatremic or even hyponatremic patients. Studies are
warranted to determine whether avoidance of hypernatremia
or its prompt and sustained correction improves clinical
outcomes.
Keywords Hypernatremia . Cancer . Healthcare cost .
Survival . Serum sodium . Length of hospital stay .
Hyponatremia . Electrolyte disorders . Outcomes . Leukemia
Introduction
Water homeostasis, regulated by thirst and the effect of
arginine vasopressin on kidneys’ ability to excrete water,
plays a key role in maintaining serum sodium concentration
[1]. An increase or decrease in total body water leads to
hyponatremia or hypernatremia, respectively [2]. A large
body of data is available on hyponatremia in hospitalized
patients, such as its frequency and impact on clinical out-
comes and hospital cost, but such data are limited in hyper-
natremia, and none to our knowledge exists for patients with
cancer [3–9]. Furthermore, as a subgroup, cancer patients
are very vulnerable to fluid and electrolytes derangements,
which can often be the consequences of several factors, such
as paraneoplastic syndromes, anticancer agents, neoplastic
spread, surgical procedures, renal pathology, or corticoste-
roid use [10]. Moreover, electrolyte imbalances have been
shown to impact and, in fact, even predict survival in cancer
patients [6, 11]. The objective of this study was to evaluate
the frequency of hypernatremia in patients admitted to a
comprehensive cancer center and to examine its effect on
patient outcomes and healthcare cost.
1872
Methods
The Institutional Review Board at University of Texas M.
D. Anderson Cancer Center (MDACC) approved the col-
lection of the data and the current cross-sectional analysis.
The data were collected into an electronic database on all
patients admitted consecutively to MDACC for 3-months
(May–July 2006). An admission was defined as a stay
of >23 h in the hospital that should also include midnight.
Data were checked at least once a week during this period
for accuracy and for inclusion of any missing data when
available. The eligibility for inclusion in this analysis was
any patient admitted to MDACC during this period with at
least one value for serum sodium at admission or during
hospital stay. For each patient, information on demograph-
ics, medical conditions, laboratory data, treatments, clinical
outcomes, and billing data were electronically collected.
Part of the data used here were previously analyzed, and a
paper on hyponatremia and another on acute kidney injury
were recently published [6, 12]. The results of this hyper-
natremia analysis have not been previously published. The
primary outcome variable was hypernatremia defined as a
serum sodium value of >147 mEq/L. The normal range of
serum sodium at MDACC laboratory is 135–147 mEq/L.
Among the 3,886 patients admitted, serum sodium values
were not available in 440 patients (11.3 %) yielding a total
of 3,436 patients for analysis. The data collected on their
first admission (i.e., n=3,436) were analyzed. To render the
large data of serum sodium values recorded during the
admission and hospital stay manageable for analysis, the
serum sodium values were extracted as values at “admis-
sion,” “peak,” “low,” and “discharge.” For sensitivity anal-
yses, multiple imputation method available in SPSS was
used to impute the missing sodium values. All outcomes
were analyzed with and without imputed serum sodium
values, and the results were found to be not statistically
different. The results presented here are based on data ex-
cluding patients with missing serum sodium values. The
accuracy and validity of the dataset used herein were con-
firmed by direct patients’ electronic chart checking that we
undertook to determine the possible causes for hypernatre-
mia and hospital death. Hypernatremia was defined as a
serum sodium value of >147 mEq/L at admission or any
time during hospital stay, and all patients with serum sodium
values were included in the analysis. To assess the effect of
hypernatremia on hospital cost and mortality, we obtained the
final hospital bill for each patient obtained from the hospital
database, and the data on patients’ mortality from the hospital
cancer registry. Each patient with admission hypernatremia or
acquired hypernatremia during hospital stay (i.e., patients with
normal serum sodium on admission but had peak serum
sodium above normal after admission) was also identified
and data between the two groups were compared.
Support Care Cancer (2013) 21:1871–1878
Statistical analysis
The demographics and clinical characteristics were tabulated
and compared with Chi-square test for categorical variables and
t test for continuous variables. The days of hospital stays were
compared by ANOVAwith multiple comparisons adjusted with
Tukey–Kramer method or Games–Howell test, the latter for
groups with unequal variance. Time to discharge was measured
from the date of hospital admission to the date of hospital
discharge. Patients who died before discharge were considered
censored at their dates of deaths. Time-to-90-day mortality was
measured from the date of admission to the date of death or last
follow-up at 90 days. The Kaplan–Meier product limit method
was used to estimate the survival outcomes of all patients by
serum sodium groups, and comparisons among the groups
were achieved with the log-rank statistic. The natremia groups
were compared using Cox proportional hazard analyses for
mortality. The proportional hazards assumption was tested with
examination of Pearson correlation between Schoenfeld resid-
uals and the rank of survival time for cases that progressed to an
event. After adjusting for other patient and clinical character-
istics, the model was fitted to determine the association of
serum sodium levels with time-to-event outcomes. All results
from the Cox model are expressed in hazard ratios (HR) and
95 % confidence intervals (CIs). For multivariate analyses,
clinically relevant variables significant in the univariate analy-
ses were included for adjustment in the final multivariate
models. The relationship between serum sodium values and
survival was nonlinear. To overcome the nonlinearity, we
employed restrictive cubic splines using four knots. The “post-
rcspline package” for Stata version 10.0 was used to generate
the curves and analyze the associations. All other survival
analyses were carried out by using the SPSS (version 16.0;
Chicago, IL). We also carried out multivariate regression anal-
ysis for hospital cost to compare the three groups for total cost
of hospital stay. Age, type of malignancy, antibiotic and che-
motherapy administered, length of hospital stay, mean serum
creatinine, and mean hematocrit values were used as covariates.
Total cost of hospital stay was log transformed prior to analysis.
Results
From 1 May to 31 July of 2006, a total 3,886 patients were
admitted to MDACC, of which 75 % were admitted once,
22 % for two to three times, and 3 % for more than three times.
For this analysis, we used data from the first admission on
3,446 patients in whom serum sodium values were available.
Natremia breakdown and patient characteristics
On admission, 69.9, 29.9, and 0.2 % were in the eunatremia,
hyponatremia, and hypernatremia categories, respectively.
Support Care Cancer (2013) 21:1871–1878
The respective values for the natremia breakdown for hospital
stay (including admission) were 51.8, 45.6, and 2.6 % (Fig. 1)
indicating the development of new hyponatremia and hyper-
natremia during patients’ stay in the hospital. On admission,
only seven patients were hypernatremic which increased over
10-fold to 90 patients during hospital stay. The clinical char-
acteristics and demographics features of all patients and of
groups based on natremia breakdowns are given in Table 1.
The prevalence of hematological malignancies, especially
leukemia, was significantly higher in patients who had hyper-
natremia as compared with the rest (Table 1). The distribution
of the primary malignancies was: hematological (20.2 %),
genitourinary (13.5 %), gastrointestinal (14.5 %), head, neck,
and lung (16.2 %), and other (melanoma, breast, thyroid,
unknown, or unidentified; 36.0 %).
Length of hospital stay
The overall length of stay of all patients regardless of serum
sodium values was 8.2±9.2 days (mean ± SD). The length
of stay of patients with hypernatremia was significantly
higher at 26.7±22.3 days compared with 10.3±10.0 days
in patients with hyponatremia (p<0·01) and 5.6±4.9 days in
patients with eunatremia (p<0·01). That hypernatremics had
longer hospital stay even when compared with hyponatre-
mics is graphically depicted in Fig. 4a. During the hospital
stay, the percent of patients stayed in the critical care unit
was 10.9, 15.1, and 30.0 % for eunatremics, hyponatremics,
and hypernatremics, respectively (p<0.01 for all compari-
sons). That hypernatremics had the highest rate of ICU
admissions even when compared with hyponatremics is
graphically depicted in Fig. 4a.
Hospital cost analysis
In the multivariate regression model adjusted for age, type
of malignancy, antibiotic and chemotherapy administered,
hospital stay, mean serum creatinine, and mean hematocrit,
Hypernatremia
(>147 mEq/L)
2.6%
Eunatremia
(135-147 mEq/L)
51.8% Hyponatremia
(<135 mEq/L)
45.6%
Fig. 1 The natremia breakdowns in hospitalized cancer patients
1873
the total hospital cost or bill was 45.7 % (95 % CI, 34.2–
57.9 %) higher in the hypernatremic than in eunatremic
patients (p<0.01) and was 36.9 % (95 % CI, 25.8–45.9 %)
higher than in hyponatremic patients (p<0.01) (Table 3).
The use of chemotherapy and antibiotic were also associated
with significantly higher costs and so was as the longer
hospital stay as would be expected (Table 3).
That hypernatremics had higher hospital bills even when
compared with hyponatremics is graphically depicted in
Fig. 4a.
Mortality
The overall crude in-hospital mortality rate regardless of
serum sodium was 5.4 %. When grouped by natremia
groups, this was 2.4, 7.3, and 39.3 % in the eunatremia,
hyponatremia, and hypernatremia groups (p<0.01 for all
comparisons). The causes of death in 35 hypernatremic
patients who died in the hospital was multifactorial with
infection as the most common denominator. Pneumonia
followed by respiratory failure was noted in 37.0 % of the
patients; whereas, 17.4 % of the patients had sepsis due to
infections other than pneumonia.
The 90-day mortality and serum sodium values were plot-
ted as restrictive cubic splines, which showed a “U”-shaped
relationship that was steeper for hypernatremia than for hypo-
natremia (Fig. 2). The overall 90-day mortality rate was
11.7 % and based on natremia groups were 7.5, 14.4, and
47.2 % in the eunatremia, hyponatremia, and hypernatremia
groups (p<0.01). Multivariate Cox regression analysis was
carried out using the eunatremia group as reference category
(Table 2). Adjusting for age, type of malignancy, chemother-
apy and antibiotic therapy, length of hospital stay, and serum
creatinine and hematocrit values, the HR for 90-day mortality
in hypernatremics were significantly higher compared with
eunatremics (5.09 (95 % CI, 3.32–7.81); p<0·01). Repeating
the analysis with hyponatremia as the reference category,
analysis still showed a higher 90-day mortality in the hyper-
natremia group (HR, 2.79 (95 % CI, 1.91–4.11); p<0.01). The
Kaplan–Meier survival analysis also demonstrated signifi-
cantly higher rates of mortality in hypernatremic than euna-
tremic or hyponatremic patients (both p<0·01) (Fig. 3).
That hypernatremics had the highest rate of mortality
even when compared with hyponatremics is graphically
depicted in Fig. 4b.
Admission hypernatremia vs. hospital-acquired hypernatremia:
patient characteristics and outcome analyses
Table 4 displays the comparison between the two groups.
Leukemic and stem cell transplant patients were more fre-
quent in the acquired groups, and most of them during admis-
sion were receiving chemotherapy and were admitted to
1874 Support Care Cancer (2013) 21:1871–1878

Table 1 The patient characteristics: all patients, and groups based on natremia breakdowns

Variables Total (N=3,446, 100 %) Natremia groups

Hypernatremia Hyponatremia Eunatremia p value*

(>147 mEq/L; (<135 mEq/L; (135–147 mEq/L;
N=90, 2.6 %) (N=1,571, 45.6 %) N=1,785, 51.8 %)

Age (years, mean ± SD) 55.8±16.7 59.2±16.9 57.3±15.9 54.5±17.3 <0.01

Gender (male %) 51.8 46.1 53.7 50.4 <0.05
Race (%) 0.09
White 72.8 69.7 72.7 73.0 0.35
Black 9.6 14.6 8.9 10.1
Hispanic 12.9 13.5 13.2 12.6
Othersa 4.7 2.2 5.3 4.3
Primary cancer (%)
Hematological malignanciesb 686 (20.2) 23 (27.4) 292 (18.9) 371 (21.1) <0.01
Genitourinary 464 (13.5) 9 (10.1) 206 (13.2) 249 (14.0) <0.01
Gastrointestinal 499 (14.5) 11 (12.2) 271 (17.3) 217 (12.4) 0.50
Head, neck, and lung 549 (16.2) 14 (16.7) 261 (16.9) 274 (15.6) <0.01
Othersc 1,237 (36.0) 32 (35.9) 536 (34.2) 669 (37.5) 0.60

*p values (overall)
a
Include Asian, Filipino, Pacific Islander, and American Indian
b
Include leukemia, lymphoma, and myeloma
c
Include melanoma, breast, and thyroid malignancies

critical care unit. These suggest that the acquired hypernatre-
mic as a group were severely ill. Consistently, the length of
stay, hospital bills and the crude mortality rate were higher in
the acquired-hypernatremic group than the admission-
hypernatremic group, although mortality did not reach statis-
tical significance (Table 4). The sample size of patients in the
admission-hypernatremia group was quite small and may not
fully represent the true population.
Discussion
Our cross-sectional analysis of data collected from a large
number of patients admitted consecutively to the UT
Fig. 2 The restrictive cubic
MDACC over a 3-month period demonstrated that hyperna-
tremia in hospitalized cancer patients was seen in 2.6 %, far
less than the very high rates of 46 % of hyponatremia seen in
concurrently admitted cancer patients. Most of the hyperna-
tremia—nearly 90 %—were acquired in the hospital and
hypernatremia although was infrequent, however, associated
with unexpectedly worse clinical outcomes than hyponatre-
mia, i.e., higher mortality, hospital stay, and hospital bills.
In a prospective study of patients admitted to general
medical wards of a large urban university hospital for
3 months, hypernatremia as defined as >150 mEq/L of serum
sodium was noted in 1 % of patients [8]. Although no com-
parison to eunatremic or hyponatremic patients were provid-
ed, the mortality rate of 41 % reported for patients with
Predicted probability of 90-day mortality

spline showing the relationship

Predicted probability of in-hospital
.8

.8

between serum sodium values

during hospital stay and in-
hospital and 90-day mortality.
.6

.6

Mortality risk is higher with

hypernatremia than
hyponatremia
.4

.4
.2

.2
0
0

110 120 130 140 150 160 110 120 130 140 150 160
Mean Serum sodium (mEq/L) Mean Serum sodium (mEq/L)
Support Care Cancer (2013) 21:1871–1878 1875

Table 2 Multivariate hazard ratios (HR) for in-hospital and 90-day mortality showing significant association with hypernatremia

HR for in-hospital mortality (N=3,446) HR for 90-day mortality (N=3,446)

HR (95 % CI) p value HR (95 % CI) p value

Multivariate
Age (year) 0.99 ( 0.98–1.01) 0.49 1.01 (1.001–1.02) <0.01
Malignancies (hematological vs. non-hematological) 1.16 (0.78–1.74) 0.47 1.61 (1.22–2.12) <0.01
Chemotherapy (yes vs. no) 0.49 (0.34–0.72) <0.01 1.10 (0.87–1.42) 0.42
Antibiotic use (yes vs. no) 0.61 ( 0.39–0.94) 0.02 1.17 (0.90–1.52) 0.24
Hospital stay in days – – 1.002 (0.99–1.01) 0.36
Serum creatininea 1.25 (1.12–1.40) <0.01 1.29 (1.18–1.41) <0.01
Hematocrita 0.91 (0.87–0.95) <0.01 0.96 (0.93–0.98) 0.01
Serum sodium (categories)
Eunatremia (135–147 mEq/L) 1.00 (ref) 1.00 (ref)
Hyponatremia (<135 mEq/L) 1.21 ( 0.83–1.74) 0.34 1.81 (1.44-2.29) <0.01
Hypernatremia (>147 mEq/L) 2.17 ( 1.30–3.64) <0.01 5.09 (3.32–7.81) <0.01
a
Mean of all values recorded during hospital stay

hypernatremic in this study suggest that mortality in this
subgroup is likely much higher than non-hypernatremic
patients. In another study of 8,142 consecutive adults admitted
to a medical–surgical intensive care unit (ICU) between 2000
and 2006, the frequency of ICU-acquired hypernatremia in-
terestingly was higher than hyponatremia (26 vs. 11 %), but
important to this discussion, hypernatremia was associated
with higher hospital mortality compared with hyponatremia
(34 vs. 28 %) [7]. A recent analysis based on retrospective
case series of 43,911 patients with serum sodium seen at a
university hospital’s Emergency Room (ER) in Switzerland,
hypernatremia was noted in 2 % and hyponatremia in 10 %.
Although the data on outcomes provided were limited, mor-
tality rates among the hypernatremic patients were over 2-fold
higher than the hyponatremic patients [9]. Thus in four set-
tings, ICU, general medical hospital, ER, and our comprehen-
sive cancer center, hypernatremia was found to be consistently
associated with higher mortality even when compared against
patients with hyponatremia.
Hyponatremia is known to be associated with higher cost,
but there are no published studies to our knowledge on the
impact of hypernatremia on hospital cost. Our study dem-
onstrated that even after adjusting for several factors likely
to affect the cost of patients care in the hospital (Table 3),

Fig. 3 The Kaplan–Meier

survival for 90 days: patients
were stratified according to
natremia groups
Cumulative Survival

Serum sodium levels

Eunatremia (135-147 mEq/L)

Hyponatremia (<135 mEq/L)
Hypernatremia (>147 mEq/L)

Log rank P<0·001

Days

Number at risk
Eunatremia 1785 1756 1733 1719 1712 1703 1647
Hyponatremia 1571 1489 1445 1426 1408 1396 1340
Hypernatremia 90 81 69 65 61 58 47
1876 Support Care Cancer (2013) 21:1871–1878

500 Cost of stay stem cell transplant service. We speculate that hospital-
Three outcomes expressed as percent Critical care stay acquired hypernatremia may likely be related to the use of
Length of stay
diuretics and chemotherapy and the practice of keeping
400 patients “dry” to avoid pulmonary edema, such as in patients
of same fo eunatremia (%)

soon after stem cell transplant. It is also to be noted that

hypernatremic patients are extremely sick, with most in the
300
critical care unit. Being sick likely predisposes the patient to
be hypernatremic rather than hypernatremia predisposes
200
them to be sick. However, this issue needs future prospec-
tive studies to be resolved.
The precise timing when hospital-acquired hypernatre-
100 mia occurred in our patients could not be known from our
dataset as we have not collected serial serum sodium values
but instead extracted admission, peak, low, and discharge
0 serum sodium values. However, we attempted to assess the
Hyponatremia Hypernatremia
volume status of hypernatremia in our patients using hospi-
tal chart review for individual patients. Our chart-based data
on 24-h fluid intake and output were not helpful as accurate
Crude mortality expressed as it relates

1800 measurement of routine fluid intake and output remains a

In-hospital
to the same for eunatremics (%)

1600 mortality challenge. Many of our patients receive potentially nephro-

90-day
1400
mortality
1200
1000
800
600
400
200
0
Hyponatremia Hypernatremia
Fig. 4 a Three outcomes as percentages are shown in bar graphs as
they relate to eunatremics, e.g., length of stay in hypernatremics was
450 % of eunatremics. b Crude mortality in hyponatremics and hyper-
natremics expressed in relation to crude mortality in eunatremics in
percentages, e.g., over 1,500 % increase in in-hospital mortality was
noted in hypernatremics in relation to eunatremics and hyponatremics
patients with hypernatremia still had higher hospital cost,
and this persisted even when compared with patients with
hyponatremia. Our hypernatremic patients had more fre-
quent critical care unit stay than the rest suggesting a sicker
patient population and partly explaining for the higher hos-
pital cost. This is confirmed in our analysis between admis-
sion hypernatremia vs. hospital-acquired hypernatremia
(Table 4). Our data indicate for the first time to our knowl-
edge that majority of the hypernatremia in hospitalized
cancer patients were actually acquired in the hospital.
Our objective in this cross-sectional analysis was to de-
termine the frequency of hypernatremia and its effects on
outcomes and, therefore, our ability to describe the cause or
mechanism for hypernatremia is limited. However, it is
evident in our analysis that hypernatremia was seen more
frequently in leukemia patients and patients admitted to
toxic drugs, such as platinum compounds, ifosfamide, meth-
otrexate, amphoteracin, cidafovir, foscarnet, and the like that
can cause direct tubular toxicity vis-a-vis renal concentrat-
ing defect contributing to unregulated water loss. This com-
bined with other factors such as vomiting and diarrhea along
with insensible loss of water from neutropenic fevers when
present can dehydrate patients. In our study, preliminary
analysis suggested that hypernatremics received more diu-
retics and steroid therapy than the eunatremic patients and
were more likely to be transferred to the critical care unit.
Whether the specific chemotherapy or specific patient care
pattern, e.g., liberal use of diuretics coupled with fluid
restrictions contributes to the higher incidence of hyperna-
tremia in leukemia patients is not clear. Among our patients
with hypernatremia, nearly a quarter of patients had one or
more surgical procedures during hospitalization and 15 %
had vomiting diarrhea or a combination accounting for
possible preferential water loss with reduced fluid intake.
In our series, only one patient was diagnosed with diabetes
insipidus (DI), although we could have likely missed DI
patients who were receiving hormonal replacement.
The strength of our study is that the data are prospective-
ly collected with one of us (PS) actively verifying and
validating the data as they were accruing. The validity of
the dataset was also confirmed during chart analysis. The
dataset represents large, unselected, and consecutively ad-
mitted patients to a large comprehensive center and, more-
over, none were excluded from analysis except for patients
with missing serum sodium values. Our study is probably
the first to study hypernatremia in a large population of
cancer patients. The clinical outcomes are based on hard
end points. For statistical analysis, we had undertaken alter-
nate approaches, such as imputing missing data followed by
Support Care Cancer (2013) 21:1871–1878 1877

Table 3 Multivariate regression

model for total hospital cost Total hospital cost (log transformed; N=3,446)
showing strong correlation with
hypernatremia Coefficient (95 % CI) p value

Multivariate regression
Age (years) 0.001 (−0.001–0.002) 0.24
Malignancies (hematological vs. non-hematological) −0.060 (−0.116–-0.005) 0.03
Chemotherapy (yes vs. no) 0.242 (0.194–0.289) <0.01
Antibiotic use (yes vs. no) 0.327 (0.283–0.370) <0.01
Hospital stay in days 0.055 (0.052–0.057) <0.01
Serum creatininea 0.005 (−0.026–0.035) 0.76
Hematocrita −0.002 (−0.006–0.002) 0.38
a
Mean of all values during hos- Serum sodium (categories)
pital stay. In this multivariate
model, hypernatremia had strong Eunatremia (135–147 mEq/L) 0.00 (ref)
correlation to cost; the cost in Hyponatremia (<135 mEq/L) 0.094 (0.054–0.134) <0.01
hypernatremics was nearly 46 % Hypernatremia (>147 mEq/L) 0.457 (0.342–0.571) <0.01
higher than eunatremics

sensitivity analyses, employing restrictive cubic splines for
nonlinear data, and multivariate regression analysis for cost
analysis. A main limitation of our study is lack of details on
the causes of hypernatremia in this population as alluded to.
Prospective studies are required in cancer patients to deter-
mine the nature and cause of evolution of hypernatremia and
to determine whether treatment to mitigate hypernatremia
will change the outcomes. Our inability to incorporate all
covariates—known or unknown—as well as comorbidities
and their severity into the regression models could have led
to some unmeasured confounding in the multivariate analy-
ses. However, we had included common variable relevant to
the survival of cancer population.
In summary, we provide what we believe is the first
report on a detailed outcome analysis of hypernatremia in
hospitalized cancer patients. Hypernatremia was seen in

Table 4 Admission hypernatre-

mia vs. hospital-acquired hyper- Variable Admission hypernatremia Acquired hypernatremia p value
natremia: patient characteristics (n=7) (n=83)
and outcome analyses
Age (years; median and 25th 54, 48–59 62, 52–70 0.42
and 75th percentiles)
Gender (male %) 33 51 0.70
Race (%)
White 57 67 0.55
Black 28 16
Hispanic 14 12
Asian 0 4
Admission service (%)
Medical oncology 57 15 0.04
Surgical oncology 15 31
Stem cell transplant 14 48
Pain management 0 3
General internal medicine 14 3
Leukemia (%) 0 26 0.15
Chemotherapy (%) 0 58 0.04
AKIa (%) 28 56 0.27
ICU admission (%) 0 69 0.13
Length of stay (days; median and 5, 3–7 27, 16–43 <0.01
Apply caution when interpreting
25th and 75th percentiles)
the results as sample size of
Hospital bill (US $; median and 24,966 (15,497–34,707) 263,947 (136,319–479,726) <0.01
patients in admission-hyperna-
25th and 75th percentile)
tremia was small
a
Crude in-hospital mortality 14 40 0.22
Based on RIFLE criteria
1878
2.6 % of patients compared with 46 % of patients with
hyponatremia. Most hypernatremia were acquired in the
hospital. Hypernatremia was seen much more commonly
in critically ill patients especially in patients with leukemia
and stem cell transplants. Patients with hypernatremia had
substantially worse clinical outcomes and higher health care
cost than eunatremic or even hyponatremic patients (Fig. 4).
If our findings are to be confirmed in prospective studies in
cancer or noncancer patients, future studies are warranted to
examine whether avoidance or prompt and sustained cor-
rection of hypernatremia could lead to a better outcome.
Conflict of interest statement We are submitting herewith the man-
uscript titled “The frequency, cost, and clinical outcomes of hypernatre-
mia in patients hospitalized to a comprehensive cancer center”—to be
considered for publication as an original article in Supportive Care in
Cancer. There are no financial relationships to disclose for any of the
authors. The manuscript has not been published elsewhere nor is it being
considered for publication. We have full control of all primary data, and
the journal is free to review these data if required.
References
1. Adrogue HJ, Madias NE (2000) Hypernatremia. N Engl J Med 342
(20):1493–1499
Support Care Cancer (2013) 21:1871–1878
2. Schrier RW (2008) Vasopressin and aquaporin 2 in clinical disor-
ders of water homeostasis. Semin Nephrol 28(3):289–296
3. Anderson RJ (1986) Hospital-associated hyponatremia. Kidney Int
29(6):1237–1247
4. Upadhyay A, Jaber BL, Madias NE (2009) Epidemiology of
hyponatremia. Semin Nephrol 29(3):227–238
5. Waikar SS, Mount DB, Curhan GC (2009) Mortality after hospi-
talization with mild, moderate, and severe hyponatremia. Am J
Med 122(9):857–865
6. Doshi SM, Shah P, Lei X, Lahoti A, Salahudeen AK (2012)
Hyponatremia in hospitalized cancer patients and its impact on
clinical outcomes. Am J Kidney Dis: Offic J National Kidney
Found 59(2):222–228
7. Stelfox HT, Ahmed SB, Khandwala F, Zygun D, Shahpori R,
Laupland K (2008) The epidemiology of intensive care unit-
acquired hyponatraemia and hypernatraemia in medical–surgical
intensive care units. Crit Care 12(6):R162
8. Palevsky PM, Bhagrath R, Greenberg A (1996) Hypernatremia in
hospitalized patients. Ann Intern Med 124(2):197–203
9. Arampatzis S, Frauchiger B, Fiedler GM, Leichtle AB, Buhl D,
Schwarz C et al (2012) Characteristics, symptoms, and outcome of
severe dysnatremias present on hospital admission. The American
journal of medicine 125(11):1125, e1–e7
10. Salahudeen AK, Bonventre JV (2013) Onconephrology: the latest
frontier in the war against kidney disease. J Am Soc Nephrol:
JASN 24(1):26–30
11. Elsayem A, Mori M, Parsons HA, Munsell MF, Hui D, Delgado-
Guay MO et al (2010) Predictors of inpatient mortality in an acute
palliative care unit at a comprehensive cancer center. Support Care
Cancer 18(1):67–76
12. Salahudeen AK, Doshi SM, Pawar T, Nowshad G, Lahoti A, Shah
P (2013) Incidence rate, clinical correlates, and outcomes of AKI
in patients admitted to a comprehensive cancer center. Clin J Am
Soc Nephrol. doi:10.2215/CJN.03530412