Hereditary spherocytosis

Abstract: Unusual dominantly inherited conditions of the red cell, collected under the generic title `hereditary stomatocytosis and allied disorders', exist, in which the red cell `leal-s' the univalent cations sodium (Na+) and potassium (K'). In some kindreds with these disorders, bizarre temperature effects can occur that have profound effects on the way in which the cells behave when removed from the body and cooled to either room or refrigerator temperatures. In some types, the cells lose K' at room temperature, giving rise to pseudohyperkalaemia; in others, this occurs in concert with swelling of the red cell and pseudomacrocytosis. In some of these conditions, a red-cell abnormality is clearly demonstrated by the presence of haemolytic anaemia; however, routine haematology can be virtually normal in the milder versions. All are inherited as dominants, although new mutations can be seen.

Key words: Erythrocytes. Pseudohyperkalaemia. Stomatocytosis. Temperature.

Red-cell-based pseudohyperkalaemia

Twenty years ago, this group identified a patient in Edinburgh who presented with recurrent but variable hyperkalaemia.1 Her general practitioner (GP) had started her on diuretics for ankle oedema and had later sent blood for estimation of urea and electrolytes. The doctor was surprised when he was told that the potassium (K+) concentration in the plasma (plasma [[K+]) was 8.2 mmol/L, without obvious lysis. Apart from the oedema, for which no cardiac, renal or hepatic cause was found, the patient was apparently healthy. Hospital investigations showed no evidence of any kind of serious metabolic disease, such as renal failure, mineralocorticoid deficiency or acidosis, which might result in such hyperkalaemia; and the electrocardiograph (ECG), a useful indicator of in vivo plasma [K'], was always entirely normal. Repeated tests by the GP (whose surgery was outside the city) were always high; however, hospital tests were always normal.

Pseudohyperkalaemia, due to an in vitro artefact, was suspected. There was no excess of leucocytes or platelets, sometimes associated with pseudohyperkalaemia;2-5 nor was there any lysis of the red cells.

A test was set up in which heparinised blood from the patient was stored at room temperature for a period of six hours, during which time samples were taken for separation of plasma from red cells, and plasma[K+] estimated. It became clear immediately that whilst the plasma [K+] in normal blood showed no significant change over this time period (as expected from common experience), there was a relentless rise in plasma [K+] in the patient. Family studies revealed a further 15 similarly affected individuals linked by an autosomal dominant pedigree. Studies showed that this net K+ loss from the red cell did not occur at 37degC, and a simple temperature effect was suggested. The presence of this effect in three other unrelated families studied recently by us is illustrated in Figure 1.

We called this condition `familial pseudohyperkalaemia,' and to distinguish it from other variants that will be discussed later, we now call it `familiar pseudohyperkalaemia Edinburgh'. In the patient described above, the routine haematology was almost completely normal. There was a very mild, fully compensated haemolytic state. The reticulocyte count hovered around the upper limit of normal at approximately 2-3%. There was no jaundice or anaemia. Further such pedigrees, in which pseudohyperkalaemia occurs with essentially normal haematology, have been described.6-11 A case recently reported in The Lancet is almost certainly of this type.12

Potassium and sodium homeostasis in the normal red cell

As in so many other cell types, the proper control of intracellular sodium (Na+) and K' is vital to the red cell. It is by means of this control that the red cell exerts `volume control;' the ability to stop itself swelling and bursting under the osmotic pressure exerted by its non-diffusible intracellular contents. As is well known, like virtually all other cells the red cell shows a high concentration of internal K' (around 100 mmol/L cells), whilst the internal concentration of Na+ is low (around 8 mmol/L cells).

The red cell is easily available and has been studied intensively by membrane-transport physiologists, and it was early experiments in the human erythrocyte that indicated the presence of the universally distributed Na+K+ pump - an ATP-consuming membrane protein that pushes Na+ out of the cell, and K+ into it, and is responsible for the maintenance of these gradients. This protein is inhibited specifically by the digitalis glycosides, and the most soluble of these, ouabain, is used commonly in the laboratory to identify ion transport that is due to the Na+K+ pump.

Isotopic tracer methods, which measure 'unidirectional' tracer fluxes (influx or efflux) are extremely useful in the red cell. By these means, workers have identified further systems,"notably an Na+KCl cotransporter, inhibited by the loop diuretics furosemide and bumetanide; a Na+Ne exchange system, which may be related to Nat-proton exchange systems elsewhere; and a K' channel that is activated by high intracellular calcium (Ca^sup 2^). These last three are not directly relevant to major movements of Na+ and K+ across the red-cell membrane, since they are either silent (e.g. Ca^sup 2^-activated K' channel) or only operate exchange fluxes (e.g. Na+K+Cl- co-transport, Na+Na+ exchange).

Lastly, there is a so-called `passive leak' process that seems to be a simple diffusional process, obeying Fick's law of passive diffusion. For K', this is easily measured as that flux which persists in the presence of the two inhibitors, ouabain and bumetanide. Since it is simply diffusional, its activity can be measured equally well either by influx or outflux methods. Usually, we measure the technically simpler influx rates, although, of course, it is the efflux that is physiologically important because it balances the inwardly directed pumping action of the Na+K' pump.

Leaky red cells: the hereditary stomatocytoses and allied disorders

These isotopic tracer methods for analysis of ionic movements across the red-cell membrane can be applied usefully to pathological red cells, and the results can be very striking. Early studies of the original familial pseudohyperkalaemia Edinburgh family indicated that the basic defect lay in the so-called passive-leak process, rather than the Na+K+ pump. It was the temperature dependence of this leak that was abnormally shallow, rather than that of the Na+K+ pump.

This abnormality in the passive leak suggested a link between this pseudohyperkalaemic condition and a group of previously known human haemolytic anaemias - a diverse group known as the hereditary stomatocytoses - in which it was known that abnormalities in this same passive-leak process at 37degC were associated with overt cellular pathology in the form of haemolytic anaemia. These conditions are usually recognised by virtue of the characteristic stomatocytic (mouth-shaped) erythrocyte morphology (Figure 2), a form which was first recognised in the original description of this condition. The first case to be described, now labelled overhydrated hereditary stomatocytosis (HSt), remains the most dramatic of the group.

The word 'stomatocyte' was first coined by Lock, Sephton-Smith and Hardisty14 at Great Ormond Street Hospital, London, to describe the erythrocyte morphology in a pale, jaundiced baby with a haemolytic anaemia. In all other respects, it was identical to hereditary spherocytosis, except in the morphology, which showed a boat- or mouth-shaped red cell with a central slit of pallor rather than a disc.

American workers, investigating a similar case, first identified the ion leak in these cells, which at about 40 times the normal rate is markedly deleterious. Such patients have a moderately severe haemolytic anaemia with overt jaundice, haemoglobin (Hb) levels of 9-11 g/dL, a reticulocytosis of 10-20% and a macrocytosis of 110 fL.

Although here we focus on K, there is an equivalent effect on leaks to Na+. We later confirmed the ion leak in this original British family." This condition is rare; we know of only three pedigrees in the UK, and there are about six worldwide. Of course, stomatocytes are now well recognised in many conditions, including liver disease, but, in our experience, dominantly inherited stomatocytosis with haemolytic anaemia is associated typically with a membrane leak.

Thus, the general group of dominantly inherited conditions with which we are concerned here fall under the general title `hereditary stomatocytoses and allied disorders.' This heading is now a little old-fashioned because not all of the cases that we deal with are frankly stomatocytic, and there are certainly cases of congenital stomatocytosis that do not show an ion leak. Nevertheless, there is no better term, and the name is enshrined in the literature.

Recently, we have reviewed these conditions in more depth,16 and their major features are summarised in Table 1. We have named the more recently described haemolytic variants after their town of origin (e.g. HSt Blackburn, HSt Woking, etc.).

The most common form of these conditions is now known as `dehydrated HSt' or `hereditary xerocytosis'. This is a less severe condition, with a much lower cation leak. The mean cell haemoglobin concentration (MCHC) is typically high, whereas in the original (overhydrated) condition it is low. We know of five pedigrees in the UK. Our colleagues in Europe have identified many further cases, some in association with a peculiar transient state of oedema in the newborn, manifesting mainly as ascites, which resolves spontaneously with time."-" This is the only pathology found in these cases that cannot be directly attributed to red-cell disease.

Temperature effects in leaky red cells

These two original forms of the condition are not commonly marked by major temperature effects. Historically, the first temperature effect described among the different variants was cryohydrocytosis, a form of stomatocytosis in which the cells showed very marked swelling and lysis when stored at refrigerator temperatures. Later, it was recognised that Na+ and K+ transport was enhanced in these cells at these temperatures.20 Pseudohyperkalaemia, similar to that observed in our original Edinburgh family, turns out to be common in other pedigrees, if it is sought."

To investigate the familial pseudohyperkalaemia Edinburgh kindred, we developed methods to look at the temperature dependence of the passive-leak fluxes of K+ and Na+, and it appears particularly easy and informative to measure the ouabain+ bumetanide-insensitive influx of K+. Results of such studies on seven different pedigrees are shown in Figure 3, representing a total of 15 families known to us in the UK.

These comparisons, which may appear bewildering at first sight, illustrate the genetic heterogeneity of these conditions. The flux at 37degC is proportional to the leak at body temperature and, roughly, to the degree of haemolysis, which is, in turn, proportional to the degree of stomatocytosis (only a proportion of the cells will be overtly stomatocytic in any one blood film). In our experience, the profile of the temperature curve not only distinguishes one family from another (a useful diagnostic tool), but predicts also how the cell behaves on cooling.

First, it is necessary to point out that the temperature dependence of this 'leaky' K+ flux in normal cells is not simple: a minimum is seen at about 8 degC.22 Those conditions in which the temperature curve is parallel to normal, or shows no minimum (dehydrated HSt, HSt Woking, overhydrated HSt), do not show unusual temperature effects such as pseudohyperkalaemia. In others, in which the temperature profile shows either a `shallow slope' variant (familial pseudohyperkalaemia Edinburgh, HSt Blackburn), or U-shaped (cryohydrocytosis) or 'shoulder' pattern (familial pseudohyperkalaemia Chiswick) show pseudohyperkalaemia, because, as the cell is cooled to room temperature, there is a disparity between the temperature dependence of the leak, reflected by these curves, and the opposing Na+K+ pump, which always shows a simple monotonic fall. The leak wins and there is a net loss of K' into the surrounding plasma. Those with very high fluxes at OdegC (cryohydrocytosis and, to a lesser extent, HSt Blackbum) show frank lysis on overnight storage at refrigerator temperatures.

As pointed out above, the leak applies to Na+ as well as to K+ and pseudohyponatraemia can occur in some of these conditions, as the Na+ leaks into the cells at low temperatures; however, this does not cause the same clinical concern.23

These curves are consistent with the studies of pseudohyperkalaeniia in Figure 1. Whole heparinised blood from representatives of three different kindreds was stored at 37degC, 20degC and OdegC, and samples were taken at intervals for separation of plasma and estimation of plasma [K+]. In all three of these pedigrees, the plasma [K+] at 20degC showed a marked rise of about 1-1.5 mmol/L per hour, whilst at 37degC all showed a slight fall (discussed below). At 0degC, there were marked differences in the rate of K+ accumulation.

Family A (the cryohydrocytosis variant2l) show a very rapid rise, whilst family B (HSt Blackburn24) show an intermediate rise at 20degC, and family C (familial pseudohyperkalaemia Chiswic25) show a rise that is virtually the same as that at 20degC. These differences can be related simply to differences in temperature dependence of the passive leak. The cryohydrocytosis cells, with the U-shaped profile in Figure 3, show the greatest isotopic flux at O'C and the greatest net loss of K' on storage of whole blood at this temperature. Both of these quantities are intermediate in value in the HSt Blackburn (shallow slope) case, and least in familial pseudohyperkalaemia Chiswick.

Consequent upon these major ionic shifts at low temperatures in cryohydrocytosis, the cell begins to swell and burst, and the effect on the mean cell volume (MCV), as determined on the Sysmex SE9500 haematology analyser, is illustrated in Figure 4. The MCV rises at O'C (panel A) and the red-cell distribution width also increases markedly (panel B). Very marked autohaemolysis occurs in these cells at low temperatures (50-70% after 48 hours). In fact, during family studies, we found that we were able to rule out this diagnosis absolutely if a heparinised blood sample failed to show lysis after overnight storage on ice.

Although the variants illustrated in Figure 3 may seem complex, almost certainly it is not exhaustive, and we are continually surprised by new variations. To get to the genetic basis of these conditions, we have performed genetic mapping studies on the large pedigrees available to us. Dehydrated hereditary stomatocytosis, the most common condition, maps to chromosome 16,26 as does the original familial pseudohyperkalaea Edinburgh family." Cryohydrocytosis and HSt Blackburn do not map to this locus (A Iolascon, unpublished data).

Apart from the research interest in a condition in which the vital process of control of Na+ and K' transport is abnormal, diagnosis of these conditions is not a triviality. Many of these conditions are mistaken for hereditary spherocytosis and treated surgically; splenectomy may be followed many years later by tragic thromboembolic complications, which can include pulmonary hypertension and death.28

There are more families out there

We encourage you to keep an eye out for these conditions. Mysterious and worrisome hyperkalaemia can be explained, and accurate haematological diagnosis can be made with avoidance of unnecessary and dangerous splenectomy. The pointers to look out for are as follows:

* Variable hyperkalaemia, sometimes with macrocytosis, the hyperkalaemia (and, if present, macrocytosis) typically being more marked on a GP specimen than on a fresh hospital specimen;

* Stomatocytes on the blood film in frankly anaemic cases; and,

* Splenectomy failure in a family member, a lack of haematological response with persistent thrombocythaemia and possibly clotting events, such as deep venous thrombosis and pulmonary emboli.

Simple diagnostic tests

The most informative test on these conditions is to measure isotopic K' or Na+ fluxes, but these tests, although simple, are not widely available. Measurements of intracellular Na+ and K' on cells washed free of extracellular Na+ and K' in magnesium chloride media are even simpler. The blood film remains the mainstay of haematological diagnosis, but suspicions of this diagnosis are strengthened by a history of pseudohyperkalaemia, or by a family history of failed splenectomy (i.e. the knowledge that in an affected family member, splenectomy was not followed by a decrease in reticulocyte count and improvement in haemoglobin, and possibly was accompanied by chronic thrombocythaemia).

The frequent misdiagnosis as spherocytosis may not be entirely accidental. Quite apart from the rarity of these conditions, it is possible that this apparent spherocytosis represents yet another temperature effect in some families, for it can be imagined that as cells swell under the load of the inward Na+ leak, then the cell will move from a stomatocytic shape to a spherocytic shape as it 'inflates' before finally bursting. But these are really 'macrospherocytes,' and the high MCV can help to distinguish these conditions from 'common' hereditary spherocytosis.

The underlying pathophysiology

The underlying molecular basis of these diseases is enigmatic. It cannot be a coincidence that stomatocytosis is associated so frequently with an ion leak. Stomatocytosis is generally accepted to be the result of expansion of the inner leaflet of the cell membrane's lipid bilayer, at the expense of the outer leaflet - analogous to the distortion of the skin of one's fingertips after too long in a hot bath, due to expansion of the skin's superficial layers.

The presence of stomatocytosis in conditions that quite clearly are not leaky shows that, in itself, it is not a cause of an ion leak. It seems unlikely to be a result of the ion leak (a swollen spherocyte is more likely). It is possible that an underlying defect of membrane management is responsible for this expansion and for the ion leak. The best clue to this condition comes from the study of the membrane protein stomatin, which is missing from the membrane in the original and most severe form, overhydrated HSt.

We and others have purified stomatin and cloned the gene encoding it,"' and, with expectancy, amplified the gene from the patients and sequenced it. However, it proved to be normal in both our patients and those of others; therefore, some other gene is the site of the mutation.

Hot weather and hypokalaemia

There is one further point that pertains to normal cells. Most marked in the patients but also evident in normal cases, both normal and abnormal cells typically show a fall in plasma [K] when stored at 37'C (Figure 1, diamonds). This is a consistent observation in our laboratory. Chemical pathologists have noted that plasma-W] levels among a standard population are generally lower in hot weather (the normal range in Barbados is lower than that in the UK),30-32 and this net movement of K+ into cells at 37 degC can explain it. The cause of this effect is unknown, but it reflects the point that the red cell contains a lot of K+ and that, even in the test tube, there is a dynamic interchange between cells and plasma, across the red-cell membrane

Other forms of stomatocytosis Stomatocytes are seen frequently in the peripheral blood in liver disease.33 In Australian, Ducrou and Kimber34 noted an association between stomatocytes and large platelets in patients of Mediterranean origin. In England, we found a girl with a mild, non-leaky stomatocytic anaemia, associated with large platelets and hyperlipidaemia.35 Recently, we have seen two families (unpublished data) with recessively inherited non-leaky stomatocytosis and large platelets. The explanation for this association remains obscure.

Conclusions

Leaky-membrane disorders of the red cell are rare but they do exist and are underdiagnosed. Accurate diagnosis can explain worrisome hyperkalaemia and avert needless and possibly harmful (even lethal) splenectomy. These conditions are of major physiological interest, and represent natural mutations of the control over some kind of Na+ and K' transporting system. Genetic mapping will lead to new understanding of these conditions, and of the genes and proteins goveming Na+ and K' handling by cell membranes. We are always keen to investigate new families.

We thank University College Special Trustees for funding, our patients for their cooperation, and Ms Sue Crouch and colleagues in the Department of Haematology, University College Hospital, for access to the Sysmex machine.

References

I Stewart GW, Corrall RJM, Fyffe JA et al. Familial pseudohyperkalaemia. A new syndrome. Lancet 1979; ii: 175-7.

2 Mazarelo F, Donaldson D. Lessons to be learned: a case study approach: pseudohyperkalaemia due to thrombocytosis in a case of tubo-ovarian abscess. JR Soc Health 1999; 119: 196-9.

3 Michiels JJ. The myeloproliferative disorders. An historical appraisal and personal experiences. Leak Lymph 1996; 22: 1-14.

4 Clark BA, Brown RS. Potassium homeostasis and hyperkalemic syndromes. Endocrinol Metab Clin North Am 1995; 24: 573-91.

5 Colussi G, Cipriani D. Pseudohyperkalemia in extreme leukocytosis. Am J Nephrol 1995; 15: 450-2.

6 Axford JS, Rampton DS, Mayne PD. Pseudohyperkalaemia. BMJ 1985; 291:1202.

7 James DR, Stansbie D. Familial pseudohyperkalaemia: inhibition of erythrocyte K efflux at 4'C by quinine. Clin Sci 1987; 73: 557-60.

8 Leadbetter S, O'Dowd TC. Possible screening test for familial pseudohyperkalaemia. Lancet 1982; ii: 103-4.

9 Luciani J-C, Lavabre-Bertand T, Fourcade J et al. Familial pseudohyperkalaemia. Lancet 1980; i: 491.

10 Meenaghan M, Follett GF, Brophy PJ. Temperature sensitivity of potassium flux into red blood cells in the familial pseudohyperkalaemia syndrome. Biochim Biophys Acta 1985;821:72-8.

11 Vantyghen MC, Dagher G, Doise B et al. Pseudo-hyperkaliemie. A propos d'une observation familiale. Ann Endocrinol 1991; 52: 104-8.

12 Smit WJ. Hyperkalaemia in a Thai man. Lancet 1999; 354:1182.

13 Ellory JC, Gibson JC, Stewart GW. Pathophysiology of abnormal cell volume in human red cells. In: F Lang, ed. Cell volume regulation. Basel: Karger, 1998: 120-239.

14 Lock SP, Sephton Smith R, Hardisty RM. Stomatocytosis: a hereditary haemolytic anomaly associated with haemolytic anaemia. BrJ Haematol 1961; 7: 303-14.

15 Stewart GW, Hepworth-Jones BE, Keen JN et at. Isolation of cDNA coding for a ubiquitous membrane protein deficient in high Na, low K stomatocytic erythrocytes. Blood 1992; 79: 1593-601.

16 Stewart GW, Turner EJ. The hereditary stomatocytoses and allied disorders: congenital disorders of erythrocyte membrane permeability to Na and K. Baillieres Best Pract Res Clin Haematol 1999; 12: 707-28.

17 Entazami M, Becker R, Mensen H, Marcinkowki M et al. Xerocytosis with concomitant intrauterine ascites: first description and therapeutic approach. Blood 1996; 90: 5392-3.

18 Grootenboer S, Schischmanoff PO, Cynober T et al. A genetic syndrome associating dehydrated hereditary stomatocytosis, pseudohyperkalaemia and perinatal oedema. Br J Haematol 1998;103:383-6.

19 Grootenboer S, Schiscmanoff P, Cynober T et al. Chronic hereditary xerocytosis and pseudohyperkalemia and severe perinatal ascites: the extension of a novel genetic syndrome. Blood 1997; 90(Supp 1): 14-56.

20 Miller G, Townes PL, Macwhinney JB. A new congenital hemolytic anaemia with deformed erythrocytes ('stomatocytes') and remarkable susceptibility of erythrocytes to cold hemolysis in vitro. Pediatrics 1965; 35: 906-15.

21 Coles SE, Chetty MC, Ho MM et al. Two British families with variants on the 'cryohydrocytosis' form of hereditary stomatocytosis. Br J Haematol 1999; 105: 1055-65.

22 Stewart GW, Ellory JC, Klein RA. Increased human red cell cation passive permeability below 12'C. Nature 1980; 286: 403-4.

23 Kilpatrick E, Burton I. Pseudohyperkalaemia, pseudohyponatraemia and pseudohypoglycaenia in a patient with hereditary stomatocytosis. Ann Clin Biochem 1997; 34: 561-3.

24 Coles SE, Ho MM, Chetty, MC et al. Hereditary stomatocytosis with marked pseudohyperkalaemia. Br J Haematol 1999; 104:275-83.

25 Haines P, Crawley C, Chetty M, Jarvis H, Coles SE, Fisher J et al. Familial pseudohyperkalaemia Chiswick: a novel congenital thermotropic variant of K and Na transport across the human red cell membrane. J Physiol 2000.

26 Carella M, Stewart GW, Ajetunmobi JF et al. Genome-wide search for dehydrated hereditary stomatocytosis (hereditary xerocytosis): mapping of locus to chromosome 16 (q23-ter). Am J Hum Genet 1998; 63: 810-6.

27 Iolascon A, Stewart GW, Ajetunmobi JF et al. Familial pseudohyperkalemia maps to the same locus as dehydrated hereditary stomatocytosis (hereditary xerocytosis). Blood 1999; 93: 3120-3.

28 Stewart GW, Amess JAL, Eber S et al. Thrombo-embolic disease after splenectomy for hereditary stomatocytosis. Br J Haematol 1996; 93: 303- 10.

29 Hiebl-Dirschmied CM, Entler B, Glotzmann C et al. Cloning and nucleotide sequence of cDNA encoding human erythrocyte band 7 integral membrane protein. Biochim Biophys Acta 1991;1090:123-4.

30 Masters PW, Lawson N, Marenah CB, Maile LJ. High ambient temperature: a spurious cause of hypokalaemia. BMJ 1996; 312:1652-3.

31 Buckley-Sharp MD, Gardner DA. Commentary: replication of results. BMJ 1996; 312: 1653.

32 Masters PW, Ismail AA. Minimising fictitious hyperkalaemia. Samples should be centrifuged after collection in general practices. BMJ 1997; 315: 190.

33 Lux S, Palek J. Disorders of the red cell membrane. In: R Handin, S Lux, T Stossel, eds. Blood: principles and practice of hematology. Philadelphia: JB Lippincott, 1995: 1701-818.

34 Ducrou W, Kimber RJ. Stomatocytes, haemolytic anaemia and abdominal pain in Mediterranean migrants. Some examples of

a new syndrome. Med JAustral 1969; ii: 1087-91.

35 Stewart GW, O'Brien H, Morris SA et al. Stomatocytosis, abnormal platelets and pseudo-homozygous hypercholesterolaemia. Eur J Haematol 1987; 38: 376-80.

36 Meadow SR. Stomatocytosis. Proc Roy Soc Med 1967; 60: 13-5.

37 Stewart GW, Ellory JC. A family with mild xerocytosis showing increased cation permeability at low temperatures. Clin Sci 1985;69:309-19.

M.C. Chetty and G.W. Stewart Department of Medicine, Universtiy college London, Rayne Institute, University Street, WCIE 6JJ, UK (Accepted 28 September 2000)

Correspondence to: Dr G. W. Stewart. E-mail: g.stewart@ucl.ac.uk

Copyright Royal Society of Medicine Press Ltd. 2001
Provided by ProQuest Information and Learning Company. All rights Reserved