Neurosurg Focus 30 (2):E6, 2011

Syndrome of alternating hypernatremia and hyponatremia

after hypothalamic hamartoma surgery
Adib A. Abla, M.D.,1 Scott D. Wait, M.D.,1 Jonathan A. Forbes, M.D., 4
Sandipan Pati, M.D., 3 Roger E. Johnsonbaugh, M.D., Ph.D., 2 John F. Kerrigan, M.D., 3
and Yu-Tze Ng, M.D. 3

Divisions of 1Neurological Surgery and 3Neurology, Barrow Neurological Institute, St. Joseph’s Hospital and
Medical Center; 2Arizona Pediatric Endocrinology, Phoenix, Arizona; and 4Division of Neurological Surgery,
Vanderbilt University Medical Center, Nashville, Tennessee

Object. In this paper, the authors’ goal was to describe the occurrence of alternating hypernatremia and hypona-
tremia in pediatric patients who underwent resection of hypothalamic hamartomas (HHs) for epilepsy. Hypernatremia
in patients after pituitary or hypothalamic surgery can be caused by diabetes insipidus (DI), whereas hyponatremia
can occur due to a syndrome of inappropriate antidiuretic hormone, cerebral salt wasting, or excessive administration
of desmopressin (DDAVP). The triphasic response after surgery in the pituitary region can also explain variations in
sodium parameters in such cases.
Methods. One hundred fifty-three patients with HH who underwent surgery were enrolled in a prospective study
to monitor outcomes. Of these, 4 patients (2.6%) were noted to experience dramatic alterations in serum sodium
values. The medical records of these patients were identified and evaluated.
Results. Patients’ ages at surgery ranged from 1.2 to 6.0 years. All patients were girls. Two patients had Delal-
ande Type IV lesions (of 16 total Type IV lesions surgically treated) and 2 had Type III lesions (of 39 total Type III
lesions). All patients had a history of gelastic seizures refractory to medication. Seizure frequency ranged from 3 to
300 per day. After surgery, all patients experienced hypernatremia and hyponatremia. The largest fluctuation in serum
sodium concentration during hospitalization in a single patient was 53 mEq/L (range 123­–176 mEq/L). The mean
absolute difference in maximum and minimum sodium values was 38.2 mEq/L.
All patients exhibited an initial period of immediate DI (independent of treatment) after surgery followed by a
period of hyponatremia (independent of treatment), with a minimum value occurring between postoperative Days
5 and 8. All patients then returned to a hypernatremic state of DI, and 3 patients still require DDAVP for DI man-
agement. A second occurrence of hyponatremia lasting several days without DDAVP administration occurred in 2
patients during their hospitalization between periods of hypernatremia. One patient stabilized in the normal range of
sodium values prior to discharge from rehabilitation without the need for further intervention. At last follow-up, 3
patients are seizure-free.
Conclusions. Severe instability of sodium homeostasis with hypernatremia and hyponatremia is seen in up to
2.6% of children undergoing open resection of HH. This risk appears to be related to HH type, with a higher risk for
Types III (2 [5.1%] of 39) and IV (2 [12.5%] of 16) lesions. Here, the authors describe alternating episodes of hy-
pernatremia and hyponatremia in the postoperative period following HH surgery. Management of this entity requires
careful serial assessment of volume status and urine concentration and will often require alternating salt replacement
therapy with DDAVP administration. (DOI: 10.3171/2010.12.FOCUS10235)

Key Words • hypothalamic hamartoma • diabetes insipidus •

syndrome of inappropriate antidiuretic hormone secretion • serum sodium concentration

T
he relationship between postoperative hyponatre- pituitary surgery, DI was responsible for hypernatremia,
mia and hypernatremia has been well described and suggested that CSW rather than SIADH was respon-
in a prospective study investigating the causes of sible for hyponatremia, evidenced by hypovolemic states
both entities, either alone or in combination, following for both of these abnormalities (DI and CSW) that are at
pituitary surgery.12 Those authors found that, following opposite ends of the sodium spectrum. Two other stud-
ies suggested that SIADH causes most cases of postop-
Abbreviations used in this paper: ADH = antidiuretic hormone; erative hyponatremia after transsphenoidal pituitary sur-
ANP = atrial natriuretic peptide; CSW = cerebral salt wasting; gery.14,19 Diabetes insipidus has been previously demon-
DDAVP = desmopressin; DI = diabetes insipidus; HH = hypotha- strated after HH surgery.1,3,9,10 In 1 study of patients with
lamic hamartoma; SIADH = syndrome of inappropriate ADH. HH, hypernatremia (> 145 mEq/L) was found in 26 of 29

Neurosurg Focus / Volume 30 / February 2011 1

Unauthenticated | Downloaded 12/20/23 08:50 AM UTC
 A. A. Abla et al.

patients, and 55% of these patients developed hyperna- Results

tremia greater than 150 mmol/L.3 Following surgery in
the pituitary region, the sequential occurrence of CSW Four patients are included in this report. Data related
and SIADH has been previously demonstrated to occur to seizure frequency, seizure type, and seizure onset and
in the same patient.5,12,18 However, fluctuations in serum other symptoms are shown in Table 1. None of the patients
sodium levels resulting in sequential development of hy- had preoperative DI, CSW, or SIADH and none were tak-
pernatremia and hyponatremia have not been previously ing DDAVP before surgery. Data involving the hospital
described following HH surgery; this entity is in line with course and outcomes as they relate to seizure control, sodi-
other water electrolyte disturbances experienced after um levels, and other endocrinopathies are given in Table 2.
surgery in this region. Patients’ sodium values, urine specific gravities, base-
Here, we demonstrate alternating hypernatremia and line (dry) weight and nearly daily weights, and urine input
hyponatremia following resection of some of the largest and output values were monitored during their hospital-
HHs at our center. Four patients presented here all exhib- ization, most of which took place in an ICU setting. The
ited sodium values both above and below the normal val- patients’ laboratory values of the above-stated parameters
ue ranges. All HHs were treated surgically and all were are shown graphically in Figs. 1–4, which correspond to
Delalande Type III or IV. the patients in Cases 1–4. Desmopressin administration
is also shown.
Methods All patients immediately went into DI after surgery
and required at least 1 dose of DDAVP intranasally. All
Since 2003, 153 patients have been treated surgically at patients then went into a hyponatremic state that lasted
our center for epilepsy due to HHs with 1 or more of the fol- at least 4 days and persisted without DDAVP administra-
lowing approaches: transcallosal anterior interhemispheric tion; that is, this occurrence was not iatrogenic. In Case
interforniceal, orbitozygomatic, or endoscopic resection of 4, the hyponatremia improved and overshot the normal
HH. Nineteen HH lesions have been treated using Gamma sodium range by Day 4, but the level returned to a hypo-
Knife surgery. Patients are evaluated preoperatively by a natremic state and hit the nadir value around Day 7 after
multidisciplinary team including epileptologists, endocri- surgery (Fig. 4). In Case 4, hyponatremia occurred sev-
nologists, and neurosurgeons for appropriate selection of eral times for shorter periods later in the hospitalization,
treatment modality based on patients’ symptoms, as well as but the effect was likely iatrogenic and related to DDAVP
MR imaging and video electroencephalography findings. (Fig. 4). Another patient experienced a prolonged second
Patients are prospectively enrolled and observed after sur- decrease, with a second hyponatremic state occurring be-
gery with data collection in a secure, proprietary database tween Days 14 and 18 after surgery (Fig. 2, Case 2). In
following informed consent under protocols approved by this patient, as well as the patient in Case 4 and in all
the institutional review board of the Barrow Neurological patients between Days 4 and 8, hyponatremia occurred
Institute at St. Joseph’s Hospital and Medical Center. Fol- independent of ongoing DDAVP administration.
low-up is obtained with surveys, office visits, and telephone Urine specific gravity, urine output, and daily weights
calls at regular intervals. were used in the management of sodium hemostasis (Figs.
Patients are evaluated preoperatively and postopera- 1–4). However, weight and daily fluid output were found
tively for endocrinopathies including weight gain and DI to be unhelpful in differentiating hyponatremia due to
by a pediatric endocrinologist (R.E.J.), seizure outcome, CSW from low sodium due to SIADH. Weights were not
behavioral changes, and cognitive/developmental as well particularly helpful, and they were not measured consis-
as short-term memory capacities. Four patients to date tently in our patients. All 4 patients did eventually weigh
have been identified who experienced large fluctuations more than they did at baseline, which would indicate that
in their sodium values during hospitalization and for up their volume status at the end of hospitalization was not
to several months after surgery in nearly all cases. depleted. Volume status would be expected to be on the
TABLE 1: Preoperative characteristics: seizures and symptoms*

Case Age (yrs), Age at Seizure No. of No. of

No. Sex Onset (mos) Seizures/Day AEDs at Op Seizure Type Symptoms
1 5.6, F 4 3–12 3 gelastic, grand mal, tonic precocious puberty, developmental
delay, rage
2 1.2, F 2 50–300 2 gelastic precocious puberty, developmental
delay†
3 6.0, F 1 5–10 3 gelastic, secondary developmental delay, rage‡
complex partial
4 3.3, F 1 8–12 2 gelastic, complex partial precocious puberty, mild develop-
mental delay, rage

* AEDs = antiepileptic drugs.

† As assessed postoperatively. The patient was too young preoperatively for symptom assessment.
‡ This patient exhibited mild mental retardation and short-term memory loss.

2 Neurosurg Focus / Volume 30 / February 2011

Unauthenticated | Downloaded 12/20/23 08:50 AM UTC
Syndrome of alternating hypernatremia and hyponatremia

hyperthermia to 39.8°C during stay;

hyperthermia to 38.9°C during hos-

elevated; weight gain/aggressive

hyperthermia during hospital stay,
increased appetite/weight gain
Other Endocrine Outcomes

temperature remains slightly

behavior toward food

significant weight gain

(improving)
pital stay

seizure free; taking 1 Na fluctuated for 3 mos; DDAVP

DDAVP 0.05 mg once nightly

0.1 mg bid (last sodium value
seizure free; taking 3 resolved before discharge from

10–20 gelastic/day; 6 mos of Na swings; Na stable

seizure free; taking Na level hovers ~150 mEq/L;

Sodium Outcome

146 mEq/L)
on DDAVP
rehab
Seizure Outcome,

taking 2 AEDs

AED for mood

Medication

1 AED
AEDs
TABLE 2: Hospitalization and follow-up data related to seizure control and sodium homeostasis*

Fig. 1. Case 1. Sodium concentration (A) in mEq/L, urine specific

(seizure or aLOC) Discharge (days) (mos)

* aLOC = altered level of consciousness; FU = follow-up; LOS = length of stay; rehab = rehabilitation.

gravity (B), daily fluid intake (solid line) and daily fluid output (dashed
Hyponatremia Continued on LOS FU

30

41

41

line) in ml (C), and weight in kg (D). The values on the x axes indicate
the postoperative days. The asterisk in panel A represents the time of
26

27
16

21

administration of DDAVP. Time zero for sodium concentration repre-

sents baseline sodium level. Time zero for weight represents baseline
dry weight.
DDAVP

yes

yes

yes
no

dehydration end of the spectrum had patients experienced

both DI (which they did) and CSW (characterized by a
low volume state) without periods of SIADH. This was
Symptoms From

not the case; SIADH was more likely occurring. In most

of the patients, except the one in Case 4, fluid intake and
yes

yes

yes

† This value was obtained outside the normal analytical range.

no

urine output were closely matched. In Case 4, although

demonstrating fluid output greater than fluid input for the
majority of the stay, the patient managed to gain weight.
This raises the concern or limitation that fluid and weight
123–176+†
Na Value

(mEq/L)

measurements in the ICU are not always completely ac-

Range

III (2.1 cm) endoscopic trans- 125–160

124–150

121–160

curate or reliable in predicting volume status.

Most patients exhibited long-term impairments in
the ability to regulate water and electrolytes, continued to
have central DI, and remained on DDAVP at last follow-
Approach

ventricular
IV (2.9 cm) transcallosal

up. One patient had resolution of the fluctuating sodium

III (2.1 cm) transcallosal

IV (3.6 cm) transcallosal

prior to discharge and did not require DDAVP postop-

eratively (Case 1). There were also several other associ-
ated endocrine disturbances including hyperphagia and
hyperthermia (Table 2). Three of the 4 patients with these
Case Type (max

sodium perturbations experienced permanent injury and

length)

continued to have DI. Three of the patients had undergone

transcallosal surgery, and at surgery, a very aggressive
resection was obtained in all 3, with the percentages of
No.

resection based on volumetric calculations totaling 92%,

1

Neurosurg Focus / Volume 30 / February 2011 3

Unauthenticated | Downloaded 12/20/23 08:50 AM UTC
 A. A. Abla et al.

Fig. 2. Case 2. Sodium concentration (A) in mEq/L, urine specific Fig. 3. Case 3. Sodium concentration (A) in mEq/L, urine specific
gravity (B), daily fluid intake (solid line) and daily fluid output (dashed gravity (B), daily fluid intake (solid line) and daily fluid output (dashed
line) in ml (C), and weight in kg (D). The values on the x axes indicate line) in ml (C), and weight in kg (D). The values on the x axes indicate
the postoperative days. Asterisks in panel A represent the time of ad- the postoperative days. Asterisks in panel A represent time of adminis-
ministration of DDAVP. Time zero for sodium concentration represents tration of DDAVP. Time zero for sodium concentration represents base-
baseline sodium level. Time zero for weight represents baseline dry line sodium. Maximum sodium value was outside the range of record-
weight. able values of the laboratory test at its peak (> 176 mEq/L). Time zero
for weight represents baseline dry weight.
99%, and 100% of lesion volume. The remaining patient
(Case 2) underwent endoscopic surgery aimed at discon- tic neuronavigation and intraoperative MR imaging; no
nection only. additional resection took place after intraoperative MR
imaging. The lesion was nearly totally resected (99%)
Illustrative Case (Fig. 5). Afterward, she was immediately extubated and
taken to intensive care.
Case 4
Postoperative Course. The patient experienced few
Presentation and Examination. This 3-year-old girl postoperative neurological deficits after surgery and had
presented to our institution for management of a giant HH. minimal verbal output for the first 72 hours. She said “no”
Preoperatively, she had experienced 8–12 gelastic seizures on Day 3 after surgery but remained aphasic for most of her
per day since the 1st month of life. She had mild cogni- hospitalization; at times, she would communicate, some-
tive delay with expressive language being the most affect- times verbally, with her parents. She had left hemiparesis/
ed capacity on baseline neuropsychological and cognitive neglect, which improved within the first few weeks after
testing. She also exhibited behavioral symptoms including surgery. Anisocoria was present initially, but both pupils
rage attacks and premature menarche at 16 months of age were reactive and her anisocoria resolved slowly during her
(treated with Lupron). stay. She maintained minimal eye contact with the exam-
iner but would follow simple commands. On postoperative
Operation. The patient underwent surgery via a Day 7, she experienced an approximately 3-minute spell
transcallosal anterior interforniceal approach. Her HH is of lip-smacking and right-sided tonic posturing. She expe-
one of the largest lesions treated to date, with a maxi- rienced several more seizures during the same day, which
mum length of 3.6 cm and a volume of 20.4 cm3 (Fig. 5). were attributed to hyponatremia given that these did not
The lesion extended outside the third ventricle into the resemble her preoperative seizure semiology. She also was
prepontine cistern directly adjacent to the basilar artery. believed to experience an altered level of consciousness
Surgery proceeded uneventfully with the aid of stereotac- during hyponatremia with bouts of lethargy. Later during

4 Neurosurg Focus / Volume 30 / February 2011

Unauthenticated | Downloaded 12/20/23 08:50 AM UTC
Syndrome of alternating hypernatremia and hyponatremia

this dose up to 10 mg intranasally if needed on a 1-time

basis. Once patients are placed on standing DDAVP orders,
we will administer doses between 50 and 100 mg orally
twice daily. The patient was discharged 27 days following
surgery after readmission to the pediatric ICU from a regu-
lar ward. She required DDAVP frequently throughout her
hospitalization but did not receive any between Days 2 and
9, at which point she experienced 2 episodes of hyponatre-
mia with a hypernatremic state in between. After her 2nd
week of hospitalization, she continued to suffer from DI
and required twice daily oral DDAVP, which at 2 points
caused her to become hyponatremic. At the same time, she
experienced central neurogenic hyperthermia during her
hospitalization and required antiinflammatory medication
on several occasions.
After discharge from the hospital, this patient re-
mained in neuro-rehabilitation for 1 month for ongoing
physical therapy and then transferred to a subacute care
center closer to home. She continued to have variations in
her sodium measurements for several months. Her parents
weighed every diaper and also measured all fluid intake to
assist in managing her volume status, recording these in a
spreadsheet. She required a peripherally inserted central
catheter for intravenous medication and fluids upon return-
ing to her home state.
At present, the patient is seizure free and is taking
DDAVP 0.05 mg orally once nightly. She has experienced
mild short-term memory loss since the surgery, and her
language abilities remain slightly delayed, although she is
Fig. 4. Case 4. Sodium concentration (A) in mEq/L, urine specific in the 1st grade now and participates in all activities. She
gravity (B), daily fluid intake (solid line) and daily fluid output (dashed continues to undergo speech and language therapy. Her
line) in ml (C), and weight in kg (D). The values on the x axes indicate parents continue to maintain mild fluid restriction of free
the postoperative days. Asterisks in A represent the time of administra- water to avoid precipitous decreases in her sodium levels.
tion of DDAVP. Time zero for sodium concentration represents baseline She undergoes blood sampling every 3 months now, and
sodium. Time zero for weight represents baseline dry weight. her sodium values hover above 150 mEq/L. She has not
had a sodium value of less than 140 mEq/L in 2 years.
her hospitalization, once she was no longer hyponatremic, Her appetite has been reported by her parents as insatia-
she also experienced staring spells but no overt seizures. ble, which triggers aggressive behavior related to taking
The patient’s sodium values and other parameters are food from others.
documented in Fig. 4. On postoperative Day 9, the sodium
value returned to 152 mEq/L. Desmopressin was given in- Discussion
tranasally and orally for all measured sodium values great-
er than 150 mEq/L. Generally speaking, our regimen starts It has been reported that DI can develop after pi-
with 2.5 mg of DDAVP intranasally, and we will increase tuitary region surgery in up to 75% of cases7,8,12,15,17 and
after transsphenoidal surgery in 10%–44% of cases,4,6,12
whereas hyponatremia can occur in up to 35% of cases
after surgery of pituitary region lesions.12,13
The management of hypernatremia and hyponatre-
mia following pituitary or hypothalamic region surgery
can often be complex and dependent on several variables.
Following these types of surgeries, it is generally accepted
that hypernatremia develops in response to decreased cir-
culating levels of ADH in a process known as central DI.
This can be caused by a lack of ADH secretion (after the
appropriate stimulus) following such surgery and likely oc-
curs from surgical disruption of the supraoptic and para-
ventricular nuclei of the hypothalamus, pituitary stalk, or
posterior pituitary. We believe that one explanation for this
Fig. 5. Case 4. Sagittal T2-weighted MR images. Left: Preopera-
tive image illustrating a giant HH (arrow) with components both inside
effect in pediatric patients (such as those who have under-
and outside the third ventricle. Right: Postoperative image showing gone HH surgery) involves the smaller circulating volume,
gross-total resection of this previously identified giant HH with a small perhaps evidenced by large shifts in serum sodium concen-
rim of residual lesion anteriorly (arrow). trations with minimal shifts in intake or output of fluids or

Neurosurg Focus / Volume 30 / February 2011 5

Unauthenticated | Downloaded 12/20/23 08:50 AM UTC
 A. A. Abla et al.

salts. It is also possible that HH is a secretory tissue, and They used free water resorption as a barometer to dictate
removal results in deficits of hypothalamic-pituitary axis DDAVP administration and used fractional sodium excre-
endocrine function responsible for normal sodium homeo- tion as a way to monitor sodium replacement.
stasis, resulting in DI. At our institution, patients are routinely monitored for
In contrast to hypernatremia, there are numerous eti- hypernatremia or hyponatremia with serum sodium mea-
ologies for postoperative hyponatremia. Hyponatremia in surements every 6 hours, or more frequently when sodium
this setting can be due to overzealous ADH administra- values are more volatile. Input and output fluid parameters
tion (via DDAVP medication), a coinciding SIADH secre- are measured hourly for urine output, oral intake, intrave-
tion caused by operative trauma,2,12,13,16 or CSW, which is nous fluid intake, and other output such as emesis or bowel
a state characterized by hypovolemia and poor resorption movements.
of sodium in the kidney. A prior report regarding pitu- When patients are hypernatremic, in the setting of
itary surgery found that combined DI and hyponatremia large amounts of dilute urine (specific gravity < 1.005) and
proceeded with hypernatremia on the 1st–3rd postopera- sodium values are greater than 150 mEq/L, we place them
tive days followed by hyponatremia during postopera- on a regimen of DDAVP and encourage free water intake.
tive Days 7.5–10.5 Another study reviewed (in rats) the For patients who are hyponatremic, the management
so-called triphasic response that occurs after pituitary is more difficult and involves the assessment of volume
stalk sectioning: DI, followed by hyponatreria, and DI status. Cerebral salt wasting requires vigorous salt replace-
again.16 These authors suggested, as did others, that leak- ment, whereas SIADH requires fluid restriction.11 Fluid
age of vasopressin from damaged hypothalamiconeuro- restriction, however, as well as diuretic use in symptom-
hypophysial tracts and the posterior pituitary causes an atic patients, was suggested by 1 group as the sole treat-
“isolated second phase” (hyponatremia).5,16 In other stud- ment of hyponatremia occurring after pituitary surgery if
ies, hyponatremia has been previously shown not to be the sodium level is less than 130 mEq/L (given that their
associated with high levels of ADH but rather associated group believed it to be a form of SIADH).5 The same group
with elevated ANP, thought to have a role in CSW, after suggested that hyponatremia is the more troublesome of
pituitary region surgery.12,18 the 2 extremes and could lead to significant morbidity and
What likely happened in these 4 patients after resec- mortality if untreated but also warned that 1 patient devel-
tion of these very large tumors, most of which were re- oped acute renal failure during a period of fluid restriction
moved aggressively, is a similar occurrence to the tripha- during a hyponatremic state (sodium level of 127 mEq/L).5
sic response. However, it is unclear and not possible to say In addition to restricting fluid, we have used diuretics
with certainty that the hyponatremic states that occurred and saline replacement for treating hyponatremia. Saline
with the lowest values of sodium between Days 5 and 8 replacement, however, is not without risk and can have
was due to CSW or SIADH. Whether high circulating iatrogenic consequences just as DDAVP can. Overaggres-
ADH levels due to release from injured posterior pitu- sive treatment of hyponatremia can lead to central pon-
itary cells caused SIADH in these patients as shown in tine myelinosis with correction of sodium too quickly.
the triphasic response in rats or high circulating values of Correction of sodium at a rate of not more than 1 mEq/L
ANP caused CSW is unknown here. More likely is that every 2 hours has been our institutional goal. In those
SIADH occurred given the overall weights of the patients patients who can tolerate oral intake, salt tablets are also
when comparing discharge with baseline weight. Also, 2 prescribed. Central pontine myelinosis was not observed
additional periods (“double-dip”) of hyponatremia cannot in this cohort of patients.
be explained by iatrogenic causes (that is, administration
of DDAVP) in 2 of the 4 patients. One patient, in addi- Complications Associated With Hyponatremia
tion to experiencing a second decrease in sodium below Two areas of concern related to hyponatremia in-
the normal range, did have 2 hyponatremic episodes later clude the development of seizures and the development of
during hospitalization that were attributable to DDAVP, altered level of consciousness presumably due to cerebral
likely from routine twice daily administration. edema and increased intracranial pressure. For this rea-
son, if we have to err outside the normal range, we prefer
Management for patients to be slightly hypernatremic rather than hypo-
We have not routinely measured urine osmolalities, natremic. We treated 3 patients who experienced seizures
urine free sodium, or fractional sodium excretion to evalu- during their hyponatremic periods; 1 of these 3 patients
ate hyponatremia and hypernatremia. We also have not also experienced a decreased level of consciousness. The
measured central venous pressure in these young patients patient in Case 1 experienced seizures within 48 hours of
as a way of determining volume status, which may be more surgery of a tonic seizure semiology during a period of
invasive and add an unnecessary risk. In contrast to exten- hyponatremia. Tonic seizures and gelastic seizures were
sive testing used by other authors in a prospective investi- part of her preoperative seizure semiology. The patient
gation into the causes of aberrations in sodium physiology in Case 2 developed seizures during hyponatremia on
after pituitary region surgery,12 our management is de- postoperative Day 6 that were similar to previous gelas-
scribed below. Those authors measured ADH levels, ANP tic seizures but also had episodes of staring to the right,
levels, central venous pressure, creatinine clearance, free with right nystagmus and left arm rhythmic movement
water resorption, fractional sodium excretion, and daily followed by emesis. Seizures for the patient in Case 4 are
sodium output. They focused on central venous pressure described previously.
as an indicator of the fluid volume resuscitation needed. As stated, volume status in children (or adults) may

6 Neurosurg Focus / Volume 30 / February 2011

Unauthenticated | Downloaded 12/20/23 08:50 AM UTC
Syndrome of alternating hypernatremia and hyponatremia

not be easy to quantify;12 assessing the latency of periph- References

eral capillary refill can be useful in demonstrating dehy- 1. Abla AA, Rekate HL, Wilson DA, Wait SD, Uschold TD,
dration but remains a crude measurement. More invasive Prenger E, et al: Orbitozygomatic resection for hypothalam-
evaluations in children such as central venous pressure ic hamartoma and epilepsy: patient selection and outcome.
testing were forgone in our management. We recorded Childs Nerv Syst [epub ahead of print], 2010
daily weights and measurements of input and output of 2. Cusick JF, Hagen TC, Findling JW: Inappropriate secretion of
antidiuretic hormone after transsphenoidal surgery for pitu-
fluids in these patients. Although they are also crude itary tumors. N Engl J Med 311:36–38, 1984
measures of volume status, these values can potentially 3. Freeman JL, Zacharin M, Rosenfeld JV, Harvey AS: The en-
predict a deficit or surplus in a patient’s volume status, docrinology of hypothalamic hamartoma surgery for intra­
helpful in differentiating between salt wasting and SI- ctable epilepsy. Epileptic Disord 5:239–247, 2003
ADH. Daily weights, however, can be inconsistent or un- 4. Hans P, Stevenaert A, Albert A: Study of hypotonic polyuria af-
reliable as can a reliable quantification of exact fluid input ter trans-sphenoidal pituitary adenomectomy. Intensive Care
Med 12:95–99, 1986
and output measurements. 5. Kristof RA, Rother M, Neuloh G, Klingmüller D: Incidence,
The potential volatile nature of volume status in chil- clinical manifestations, and course of water and electrolyte
dren, given a smaller circulating fluid volume, lends a pos- metabolism disturbances following transsphenoidal pituitary
sible explanation as to why the alternating values of high adenoma surgery: a prospective observational study. Clinical
and low serum sodiums can occur so quickly in young pa- article. J Neurosurg 111:555–562, 2009
6. Lipsett MB, MacLean JP, West CD, Li MC, Pearson OH: An
tients; our impression is that small fluid shifts in children analysis of the polyuria induced by hypophysectomy in man.
could potentially result in large changes in serum sodium J Clin Endocrinol Metab 16:183–195, 1956
(or other osmolality) values. A prior study found a trend for 7. Lyen KR, Grant DB: Endocrine function, morbidity, and mor-
water electrolyte disturbances to tend to occur in younger tality after surgery for craniopharyngioma. Arch Dis Child
patients but did not address the mechanism and suggested 57:837–841, 1982
8. Newman CB, Levine LS, New MI: Endocrine function in chil-
rather that it may occur since patients with hormone-pro- dren with intrasellar and suprasellar neoplasms: before and
ducing pituitary tumors tend to be younger.5 after therapy. Am J Dis Child 135:259–266, 1981
9. Ng YT, Rekate HL, Prenger EC, Chung SS, Feiz-Erfan I, Wang
NC, et al: Transcallosal resection of hypothalamic hamartoma
Conclusions for intractable epilepsy. Epilepsia 47:1192–1202, 2006
10. Ng YT, Rekate HL, Prenger EC, Wang NC, Chung SS, Feiz-
We have shown that sequential alterations in hypo- Erfan I, et al: Endoscopic resection of hypothalamic hamar-
natremia and hypernatremia, previously demonstrated in tomas for refractory symptomatic epilepsy. Neurology 70:
pituitary region surgery, can occur in the largest of HHs. 1543–1548, 2008
Careful management of this entity requires frequent as- 11. Palmer BF: Hyponatremia in patients with central nervous sys-
sessment of sodium values, urine-specific gravity, and tem disease: SIADH versus CSW. Trends Endocrinol Metab
volume status. The determination of the cause of hypo- 14:182–187, 2003
12. Poon WS, Lolin YI, Yeung TF, Yip CP, Goh KY, Lam MK,
natremia can be further demonstrated with laboratory et al: Water and sodium disorders following surgical excision
testing of circulating ADH and ANP levels. The sodium of pituitary region tumours. Acta Neurochir (Wien) 138:921–
homeostasis derangements demonstrated here seem to 927, 1996
mimic the triphasic response; however, additional peri- 13. Sane T, Rantakari K, Poranen A, Tähtelä R, Välimäki M, Pel-
ods of hyponatremia independent of DDAVP administra- konen R: Hyponatremia after transsphenoidal surgery for pitu-
tion that occurred here suggest that there is more to the itary tumors. J Clin Endocrinol Metab 79:1395–1398, 1994
14. Sata A, Hizuka N, Kawamata T, Hori T, Takano K: Hypona-
phenomenon than is currently understood. A syndrome tremia after transsphenoidal surgery for hypothalamo-pitu-
of alternating hypernatremia and hyponatremia after HH itary tumors. Neuroendocrinology 83:117–122, 2006
surgery seen here requires further investigation into the 15. Thomsett MJ, Conte FA, Kaplan SL, Grumbach MM: Endo-
exact cause of hyponatremia occurring after HH surgery. crine and neurologic outcome in childhood craniopharyngi-
It appears to be independent of DDAVP administration, oma: Review of effect of treatment in 42 patients. J Pediatr
but whether it is CSW, SIADH, or another possible ab- 97:728–735, 1980
normality related to removal of HH lesional tissue is not 16. Ultmann MC, Hoffman GE, Nelson PB, Robinson AG: Tran-
sient hyponatremia after damage to the neurohypophyseal
known. Although an HH often causes a pathological state tracts. Neuroendocrinology 56:803–811, 1992
of seizures, it potentially could serve a secretory role in 17. Wait SD, Garrett MP, Little AS, Killory BD, White WL: En-
normal physiological sodium homeostasis. docrinopathy, vision, headache, and recurrence after transs-
phenoidal surgery for Rathke cleft cysts. Neurosurgery 67:
Disclosure 837–843, 2010
18. Yamamoto N, Miyamoto N, Seo H, Matsui N, Kuwayama A,
The authors report no conflict of interest concerning the mate- Terashima K: [Hyponatremia with high plasma ANP level—re-
rials or methods used in this study or the findings specified in this port of two cases with emphasis on the pathophysiology of cere-
paper. bral salt wasting.] No Shinkei Geka 15:1019–1023, 1987 (Jpn)
Author contributions to the study and manuscript preparation 19. Zada G, Liu CY, Fishback D, Singer PA, Weiss MH: Recogni-
include the following. Conception and design: Abla, Ng. Acquisition tion and management of delayed hyponatremia following trans-
of data: Abla, Kerrigan. Analysis and interpretation of data: Abla, sphenoidal pituitary surgery. J Neurosurg 106:66–71, 2007
Wait, Forbes, Ng. Drafting the article: Abla, Ng. Critically revising
the article: Abla, Forbes, Kerrigan, Ng. Reviewed final version of Manuscript submitted October 21, 2010.
the manuscript and approved it for submission: all authors. Statisti- Accepted December 6, 2010.
cal analysis: Abla, Kerrigan, Ng. Administrative/technical/material Address correspondence to: Adib A. Abla, M.D., Barrow Neuro-
support: Abla, Wait, Forbes, Pati, Kerrigan, Ng. Study supervision: logical Institute, 350 West Thomas Road, Phoenix, Arizona 85013.
Abla, Kerrigan, Ng. email: [email protected].

Neurosurg Focus / Volume 30 / February 2011 7

Unauthenticated | Downloaded 12/20/23 08:50 AM UTC