General Practice

High ambient temperature: a spurious cause of hypokalaemia

BMJ 1996; 312 doi: (Published 29 June 1996) Cite this as: BMJ 1996;312:1652
  1. P W Masters, senior registrara,
  2. N Lawson, consultant biochemista,
  3. C B Marenah, consultant chemical pathologista,
  4. L J Maile, general practitionerb
  1. a Department of Clinical Chemistry, Nottingham City Hospital, Nottingham NG5 1PB
  2. b The Surgery, St Wilfred's Square, Calverton NG14 4FP
  1. Correspondence to: Dr Masters.
  • Accepted 4 March 1996

During the exceptionally hot summer of 1995 we noticed an increase in the number of cases of hypokalaemia reported by this laboratory among patients seen in general practice, though quality control data showed that the laboratory's methods and performance had not changed. Samples are collected from surgeries once a day and delivered by van to the laboratory, often several hours after venepuncture. We postulated that the high ambient temperature in the interim was directly responsible for the increase in hypokalaemia by stimulating cellular uptake of potassium.

Subjects, methods, and results

Daily means for plasma potassium concentrations for samples from this hospital and general practices were obtained separately from the laboratory computer for the period 1 January to 10 August 1995. These means were compared with the maximum daily dry bulb temperatures recorded at the Nottingham and Warsop weather stations (data were supplied by the Meteorological Office at Bracknell).

Venous blood was collected from five healthy, non-fasting volunteers at 9 am and aliquoted into Vacutainers (Becton Dickinson, Oxford) containing lithium heparin. Samples were kept unseparated at 4°C, 37°C, and 23°C (the temperature of an airconditioned room). Plasma potassium was measured on an Olympus AU 800 analyser after 0, 4, and 24 hours. Changes at the higher temperatures were investigated further with another 10 volunteers, storing aliquots at 23°C and 37°C and measuring potassium after 0, 0.5, 1, 2, 3, 4, 5, and 8 hours. The effect of continuous sample mixing, as might occur during transit in a van, was investigated with samples from a further five volunteers, divided equally between two 37°C water baths, one with motorised sample agitation. Differences were compared to baseline by analysis of variance; results from different experiments were combined when measurements were made at the same temperature and time points.

The daily mean potassium concentration for hospital patients was relatively constant over the period 1 January to 30 June, with a slight fall in July and August (fig 1). The daily mean for general practice patients, however, was highly significantly correlated with maximum daily temperature (r = -0.91, P<0.0001). The widest divergence between the two populations coincided with the highest temperatures.

Fig 1
Fig 1

(Top) daily mean potassium concentrations from hospital patients (dotted line) and patients from general practice (solid line); (bottom) maximum daily temperature

In all the stability experiments, baseline potassium concentrations were in the range 3.83-4.50 mmol/l. Potassium rose rapidly when samples were kept at 4°C, as previously reported.1 2 After 4 hours the mean rise was 1.00 mmol/l (95% confidence interval 0.56 to 1.44; P<0.001). At 23°C potassium did not change significantly for up to 8 hours. After 24 hours, however, there was a mean increase of 1.12 mmol/l (0.87 to 1.37; P<0.001). At 37°C potassium showed a small initial fall of -0.22 mmol/l at 4 hours (-0.44 to -0.01; P<0.001) but a rise at 24 hours of 6.27 mmol/l (6.02 to 6.53; P<0.001). No differences were found between agitated and static samples.


Delayed sample separation is a well recognised cause of spurious hyperkalaemia. In hot weather it should be considered in the differential diagnosis of hypokalaemia. Laboratories and clinicians should be aware of this phenomenon so that results may be confirmed before starting unnecessary investigations and treatment. The corollary is that there may also be an increased prevalence of pseudonormokalaemia in patients who have true hyperkalaemia.

The apparent stability of potassium at 23°C is in agreement with a previous study which reported no change over 16 hours at 18°C.3 Another report indicates, however, that potassium concentration may fall by a mean of 0.22 mmol/l in 2 hours at 25°C.4 In vitro changes in potassium are more complex than is usually stated in textbooks.5 The exact timing and temperature are clearly critical in determining the result that is actually reported by the laboratory.

We thank the laboratory staff of the department of clinical chemistry for volunteering as subjects and for performing the potassium measurements.


  • Funding None.

  • Conflict of interest None.


  1. 1.
  2. 2.
  3. 3.
  4. 4.
  5. 5.
View Abstract