Emma
McPherson* MB,ChB
Department
of Orthopaedics
#Consultant Orthopaedic Surgeon
°corresponding author
We
performed a retrospective review of hyponatraemia in patients with hip
fractures, before and after surgery. All
patients admitted with fractures of the neck of femur who had a surgical
intervention to deal with the fracture were included.
Results were determined using two definitions for hyponatraemia.
The incidence of pre-operative and post-operative hyponatraemia were both
2.8% if hyponatraemia was defined as [Na]<130mmol/l.
No cases of hyponatraemia were found pre-operatively when hyponatraemia
was defined as [Na]<125mmol/l. Using
this definition the post operative incidence of hyponatraemia was 0.93%.
The incidence of hyponatraemia in this group of patients is small.
However the potentially severe affects of hyponatraemia warrant close
monitoring of these patients and the establishment of methods to prevent this
problem from occurring.
A recent editorial in the
British Medical Journal has highlighted the dangers of hyponatraemia in
post-operative patients, particularly the elderly (1).
It emphasised the problem in orthopaedic surgery and the “apparent
ignorance regarding the risk of hyponatraemia after joint replacement”.
This article resulted in much correspondence, in particular from the
orthopaedic community (2-7).
The problem of hyponatraemia
appears to be often related to the normal responses to trauma in the elderly
patients and the use of thiazide diuretics in this group (8).
Thiazide diuretics are used frequently for concomitant medical conditions
in the elderly population requiring joint replacement surgery.
The association between thiazide diuretics and hyponatraemia has been
reported in orthopaedic practise (9,10). We
wanted to determine the scope of this problem by determining the incidence of
hyponatraemia before and after surgery. Patients
admitted for elective total hip and knee replacement tend to have electrolyte
problems resolved before surgery via the pre-assessment clinic.
Those elderly patients admitted as emergencies will 0have more associated
co-morbidity and drug related problems. We
therefore chose to investigate the incidence of pre- and post-operative
hyponatraemia in-patients admitted with fractures of the femoral neck.
This retrospective analysis
involved all patients admitted to our unit with a fractured neck of femur from
December 2000 – June 2001. Patient
details were recorded on our STAG (Scottish Trauma Audit Group) database.
Any patient not operated on was excluded from the study.
Results of pre-operative and immediate post-operative sodium
concentrations were recorded from the hospital Biochemistry computer database. If data was not found here a search of the casenotes was
performed.
Several local hospitals were
phoned to determine their normal levels for sodium. The accepted normal range for sodium in the west of Scotland
is 135-145 mmol/l. The Biochemists
at these hospitals agreed that hyponatraemia was not significant until the
sodium level fell below 130 mmol/l. We
accepted a value of [Na]<130 mmol/l as significant as have previous authors
(8,11). Previous reports have also
taken a sodium concentration of <125 mmol/l as severe (4,5,10,12,13).
We calculated our incidence of hyponatraemia using both these definitions
of the condition.
The database showed that 112
patients had received operations for fractured neck of femur in the study
period. The mean age of these
patients was 79.1 years (range 37 – 101) and they were predominantly female
(88 female, 24 male). Five patients
were excluded from the analysis as details could not be found on the computer
nor could their casenotes be traced. This
left 107 patients in the study.
Of the 107 patients three
patients (2.8%) were noted to be hyponatraemic pre-operatively using the
definition of hyponatraemia of <130mmol/l.
This discrepancy was corrected before surgery in each case.
Three patients (2.8%) were hyponatraemic immediately post-operatively.
None of the patients noted to be hyponatraemic pre-operatively developed
the problem after surgery.
When our data was
re-examined, using a level of <125mmol/l to denote hyponatraemia, only one
post-operative patient (0.93%) developed this problem.
No pre-operative patients were hyponatraemic using this definition.
The incidence of
hyponatraemia varies in the literature from 3% to 30% (8,9,11).
The findings of this study suggest the incidence of hyponatraemia in our
group is consistent with those series previously published.
We did not detect a difference in the incidence of hyponatraemia in
patients before or after surgery. However,
those patients noted to be hyponatraemic pre-operatively did not develop the
problem in the post-operative period. This
suggests that, once alerted to the potential problem, close post-operative
monitoring helps to prevent recurrence of the problem.
The incidence of hyponatraemia after surgery is small: approximately 3%
using a level of <130mmol/l
to define the problem. Using a
sodium level of <125mmol/l to define hyponatraemia our results show an
incidence of <1%. This is again
similar to previous reports (11,13,14,15).
In each of these cases the sodium concentrations returned to normal by a
combination of fluid restriction and altering or stopping the diuretic used.
No adverse affects of hyponatraemia were noted in any of our affected
patients. Our results suggest that
hyponatraemia is a relatively uncommon sequela in our patient population and is
not as widespread as suggested by Lane and Allen (1).
The development of
hyponatraemia in elderly patients reflects their response to the effects of
trauma and surgery. In
post-operative patients the levels of antidiuretic hormone (ADH) are increased
(11,16,17). Therefore water
excretion by the kidneys is reduced. Post-operative
fluid replacement often uses 5% Dextrose solution.
Although this solution is isotonic, upon infusion into the post-operative
patient the glucose is rapidly metabolised thus effectively producing a
hypotonic solution (8). This in combination with the elevated ADH produces
hyponatraemia by a dilutional effect.
The problem of hyponatraemia
is further aggravated by the use of thiazide diuretics.
The association of hyponatraemia with thiazide diuretics has been known
for some time (18). Up to twenty percent of the population >65 years are
estimated to take a diuretic (19). Diuretics
commonly produce adverse drug reactions (20), with up to 20% of elderly patients
developing some degree of hyponatraemia (9,21).
These drugs act by increasing sodium excretion in the proximal convoluted
tubule of the nephron (8). Potassium
sparing diuretics compound the problem further by increasing sodium excretion at
the distal convoluted tubule (8,21). There
is also some evidence that thiazide diuretics might directly induce the
secretion of ADH (22,23). The
effects of electrolyte dilution by water retention and increased sodium
excretion together can produce marked levels of hyponatraemia.
Hyponatraemia can lead to
problems ranging from mild cognitive dysfunction to death (1,8-13,17-20).
The homeostatic changes associated with hyponatraemia cause cellular
hypo-osmolality. This is
responsible for the common clinical manifestations of muscle cramps, weakness,
fatigue, mental confusion and disorientation.
Severe hyponatraemia can result in brain oedema, brain herniation,
seizures and respiratory arrest (16,17). If
hyponatraemia is corrected too rapidly a demyelinating disorder, central pontine
myelinolysis, can occur. This
presents with quadriplegia with dysarthria and dysphagia (24).
Thus early recognition of the problem and appropriate corrective measures
can minimise or prevent problems associated with hyponatraemia.
We agree with Lane and Allen
that the effects of hyponatraemia are potentially devastating and all possible
measures should be taken to avoid the problem.
For this reason it is important to try to anticipate the development of
this problem, be aware of measures to try to prevent it and be aware of how to
treat it when it occurs. The
problem often results from the management of postoperative fluid regimes being
left to relatively inexperienced doctors out of hours.
Isotonic glucose solutions are used routinely in post-operative patients,
but has been shown to predispose patients to the development of hyponatraemia.
Regular teaching programmes for medical and nursing staff are required to
ensure the correct resuscitation and monitoring techniques are employed. Systems are required to allow Biochemists to be involved in
this monitoring (4,5). Regular
audit of the hospital systems for monitoring has been shown to effectively
reduce this problem and should be adopted (4).
It has been suggested that hyponatraemia levels <130mmol/L have been
associated with a sixty fold increase in mortality, particularly in women (17).
By taking a multi-disciplinary approach to post-operative hyponatraemia
the adverse effects of this metabolic disturbance could be avoided.
REFERENCES
1)
Lane N, Allen K. Hyponatraemia
after orthopaedic surgery. Br Med J 1999:318:1363-1364
2)
Severin AM, Dodds C.
Hyponatraemia after orthopaedic surgery: Failsafe system is needed.
Br Med J 1999:319:514 (letter).
3)
Harrington P. Hyponatraemia
after orthopaedic surgery: General journals must not alienate particular
specialities. Br Med J 1999:319:
514-515 (letter).
4)
Marino A, Krikler S, Blackmore M.
Hyponatraemia after orthopaedic surgery: Rigorous audit and introduction
of guidelines decreased hospital’s figures.
Br Med J 1999:319: 515 (letter).
5)
Mojiminiyi OA. Hyponatraemia
after orthopaedic surgery: Laboratory must play a part in patient management.
Br Med J 1999:319:515 (letter).
6)
Hoffbrand B. Hyponatraemia
after orthopaedic surgery: Hypotonic solutions should be used infrequently.
Br Med J 1999:319:515 (letter).
7)
Electronic responses.
Hyponatraemia after orthopaedic surgery. eBMJ 1999:318 (www.bmj.com/cgi/content/full/318/7195/1363
responses)
8)
Malin JW, Kolstad K, Hozak WJ, Rothman RH.
Thiazide induced hyponatraemia in the postoperative total joint
replacement patient. Orthopedics
1997:20:681-683.
9)
Hornick P, Allen P.
Acute hyponatraemia following total hip replacement.
Br J Clin Prac 1990:44:776-777.
10)
Tolias CM. Severe
hyponatraemia in elderly patients: cause for concern.
Ann R Coll Surg Engl 1995:77:346-348.
11)
Chung HM, Kluge R, Schrier RW, Anderson RJ.
Postoperative hyponatraemia. Arch
Int Med 1986:146:333-336.
12)
Arieff AI, Llach F, Massry SG.
Neurological manifestations and morbidity of hyponatraemia: correlation
with brain water and electrolytes. Medicine
1976:55:121-129.
13)
Natkunam A, Shek CC, Swaminathan R.
Hyponatraemia in a hospital population.
J Med 1991:22:83-96.
14)
Flear CTG, Gill GV, Burns J. Hyponatraemia: mechanism and management.
Lancet 1981:2:26-31.
15)
Kennedy PGE, Mitchell DM, Hoffbrand BI.
Severe hyponatraemia in hospital inpatients.
Br Med J 1978:2:1251-1253.
16)
Oh MS, Carroll HJ.
Disorders of sodium metabolism: hypernatraemia and hyponatraemia.
Crit Care Med 1992:20:94-103.
17)
Arieff AI. Hyponatreamia,
convulsions, respiratory arrest and permanent brain damage after elective
surgery in healthy women. N Eng J Med 1986:314: 1529-1535.
18)
Fichman MP, Vorherr H, Kleeman CR, Telfer N.
Diuretic induced hyponatraemia. Ann
Intern Med 1971:75:853-863.
19)
Castleden CM, George CF. In: O’Malley K (Ed). Clinical
Pharmacology and Drug Treatment in the Elderly.
Edinburgh: Churchill Livingston, 1984:71-98.
20)
Williamson J, Chopin JM. Adverse reactions to prescribed drugs in the elderly: a
multicentre investigation. Age
& Ageing 1980:9:73-80.
21)
Sunderam SG, Mankikar GD. Hyponatraemia in the elderly.
Age and Ageing 1983:12:77-80.
22)
Luboshitzky R, Tal-Or Z, Barzilai D.
Chlorthalidone induced syndrome of inappropriate secretion of
antidiuretic hormone. J Clin Pharmacol 1976:18:336-339.
23)
Horowitz J, Keynan A, Ben-Ishay D.
A syndrome of inappropriate ADH secretion induced by cyclothiazide.
J Clin Pharmacol 1972: 12: 337-341.
24)
Sterns RH, Riggs JE, Schochet SS Jr.
Osmotic demyelination syndrome following correction of hyponatraemia.
N Eng J Med 1986:314:1535-1542.
We would like to thank Sister Kathleen Duncan who allowed us to retrieve our data from the Scottish Trauma Audit Group (STAG) data on hip fracture patients collected in our hospital. We would also like to thank the Biochemists in the west of Scotland who gave us advice.